The Information: All

Microsoft’s Executive Exodus Foreshadows Business Transition [The Briefing]

2026-10-02T00:07:46+00:00

Is there a fire alarm going off at Microsoft’s headquarters? A slew of Microsoft executives have bolted for the exits in recent weeks, most recently Ryan Roslansky, executive vice president of Office apps, who said Thursday he would leave at the end of this year. A day earlier it was Microsoft Science President Peter Lee’s turn. Both follow a bunch of other senior exits over the past few months, most prominently Office and Windows chief Rajesh Jha.

Those exiting are mostly veterans who had been at the company for 15 years or more. Each had their own reasons for leaving, we’d guess—for example, new jobs elsewhere. But it’s likely not a coincidence they’re choosing to leave as Microsoft, like other companies, is having to rethink how it operates to deal with both the opportunities and the threats posed by AI. In this environment, it makes sense that people who have been at Microsoft for years and probably have made a lot of money on the stock (it’s up 1,289% since 2014, according to Koyfin) would decide to get out now while the going is good.

In Robotics, Google Goes Android to Tesla’s Apple [Applied AI]

2026-10-01T21:53:06+00:00

While interviewing Google’s Koray Kavukcuoglu last week, I had a flash of deja vu.

In his first sit-down since taking over as CEO of Google DeepMind, Kavukcuoglu offered some of his most detailed comments yet on where the company's robotics ambitions are headed and how they differ from the rest of the industry.

OpenAI Fires Three Safety Researchers [Briefing]

2026-10-01T19:31:26+00:00

OpenAI said it fired three safety employees for mishandling corporate information, which a person familiar with the matter said included sensitive information with an outside organization that does AI evaluations.

“We have parted ways with three individuals for violating our policies on ...

Federal Court Dismisses Publishers’ Suits Over Google’s AI Overviews [Briefing]

2026-10-01T18:24:15+00:00

A federal judge on Wednesday dismissed lawsuits from education technology company Chegg and Penske Media, the owner of Rolling Stone, Variety and other publications, which had alleged Google had violated antitrust law when its AI Overviews summaries used their content. The court rejected the ...

Exclusive: Nvidia, SoftBank Make Final $20 Billion Investment in OpenAI’s Last Round [Briefing]

2026-10-01T18:15:57+00:00

Nvidia and SoftBank have each made the final $10 billion investment in each of their $30 billion pledges to OpenAI’s last funding round, according to a person with knowledge of the investments and a statement from SoftBank.

OpenAI in March said it had $122 billion in commitments in a round that ...

Sharon AI Announces $356 Million GPU-Backed Loan [Briefing]

2026-10-01T16:32:01+00:00

Australian neocloud Sharon AI has secured $356 million through its first loan backed by its graphics processing units, making it the latest cloud provider to finance their expansion by borrowing against chips and cash flows from customer contracts. The senior secured facility carries an interest ...

Data Center Infra Provider Accelevation Sinks In Post-IPO Trading [Briefing]

2026-10-01T16:28:01+00:00

Data center infrastructure provider Accelevation’s shares were trading down roughly 5% on Thursday morning, as the stock continued to sink following its initial public offering earlier this week.

The company on Sept. 29 raised $540 million, moving ahead with the offering even as some ...

Why Coding Agents Are Choosing Vercel as Often as Humans Do [AI Agenda]

2026-10-01T14:01:40+00:00

The rise of AI agents that write software continues to pay off for the companies that host such software. Among them is Vercel, whose software helps developers make and host websites and AI apps, including so-called agents that automate white-collar tasks.

The 11-year-old company is generating $600 million in annualized revenue, or $50 million a month, up 148% from a year earlier and up from $500 million in July. The number of paying customers also rose by a similar amount, around 20%, according to a person close to the company.

Valued at $9.3 billion in a financing a year ago, Vercel has since changed its business model and mostly charges customers based on their usage rather than through subscriptions. The company competes with an array of software firms, including Netlify and Cloudflare, as well as cloud providers like Amazon. And it’s also competing with OpenRouter, soon to be owned by Stripe, in helping app developers pick and choose the right AI models from one application programming interface.

Coding AI agents that produce new code now account for about half of Vercel’s new business, up from less than 3% at the start of the year, CEO Guillermo Rauch said in June at Vercel’s Ship conference in London.

How SpaceX’s AI Unit Turned Itself Into an AI Cloud Firm

2026-10-01T13:01:44+00:00

SpaceX’s AI unit held talks over the summer about leasing computing capacity to Microsoft, say people familiar with the matter, a sign of how fully Elon Musk has embraced the neocloud business after his initial reluctance.

Selling computing power has quickly become one of SpaceX’s most lucrative businesses, generating billions of dollars a month in commitments from compute-constrained AI labs including Anthropic and Google, with more deals still to kick in. The status of the Microsoft talks isn’t clear, the people said. A spokesperson for Microsoft declined to comment.

OpenAI Accuses Moonshot of Distillation Campaign [Briefing]

2026-10-01T08:51:46+00:00

OpenAI said it recently identified and mitigated a “coordinated model-distillation campaign” involving individuals associated with Moonshot AI, the Chinese developer of Kimi models.

OpenAI said the operators of the campaign tried to extract protected reasoning—the model’s hidden thinking process ...

Behind Factory Founder’s Spat With Cognition Over Key Hire [Dealmaker]

2026-10-01T00:49:09+00:00

A dispute over high-flying AI coding firm Cognition’s hiring of a prominent sales leader who had advised rival Factory erupted into a public feud, dividing startups as well as investors within the same venture firm.

Spats between competing startups and their investors flare up every so often in Silicon Valley. Even by those standards, Wednesday’s brawl stood out—exposing how the fight over AI talent has spilled beyond top researchers into other key jobs, and underscoring that a long-standing taboo against VC firms backing rivals still has its adherents.

Why DoorDash Needs Its Own AI Assistant [The Briefing]

2026-10-01T00:31:53+00:00

Tech companies may have already perfected the art of separating consumers from their money—but AI promises to take that capability to a whole new level. The latest demonstration of that is DoorDash’s Wednesday announcement that customers will soon be able to order restaurant meals and groceries by text. The new service is a result of DoorDash integrating the capabilities of its AI assistant, which is already in its app, into iMessage.

So if you’re like me and you like to reorder the same restaurant meal every Friday night, now you can do so with a quick text. To be honest, the convenience of this seems illusory, as DoorDash’s mobile app already makes reordering a meal as easy as pie. But the AI assistant, whether it’s in the app or outside it, is an advance. If its a chore to figure out where to order food for a special dinner from, it can suggest places. The assistant is even more handy for groceries: It can list the ingredients you need for a meal and order them.

Revenue Quintupled at AI Memory Maker Micron [Briefing]

2026-09-30T23:46:12+00:00

Memory maker Micron said revenue nearly quintupled to $54.2 billion in the quarter ending Sept.3, as demand for high-bandwidth memory hardware for AI chips showed no sign of slowing down.

Micron’s gross margin nearly doubled to $86.8%, thanks to its price hikes, allowing it to generate an ...

Microsoft Science President Peter Lee to Step Down [Briefing]

2026-09-30T23:05:42+00:00

Microsoft veteran Peter Lee, who for years led the company’s internal research unit, will leave the company, he announced Wednesday.

Lee is the latest in a string of senior executive departures from Microsoft announced this year. So far this year, Office and Windows boss Rajesh Jha, Xbox CEO ...

Google Unveils Gemini 4 Argon, Pricing It Well Below Rivals [Briefing]

2026-09-30T23:05:17+00:00

Google launched Gemini 4 Argon, its latest flagship AI model, touting its abilities in cybersecurity defense and coding, legal and finance work. The company hasn’t launched such a frontier model in nearly a year, prompting questions about whether it could stay in the race with Anthropic and ...

Exclusive: Meta’s Muse Tops 3 Million Weekly Users [Briefing]

2026-09-30T23:05:17+00:00

Meta Platforms’ Muse now has more than 3 million users who give the AI agent at least one prompt a week, according to internal data reviewed by The Information, giving the product a sizable active audience just weeks after its launch.

That figure is a subset of a broader group totaling more than ...

DeepSeek Open-Sources Software Tools for Huawei AI Chips [Briefing]

2026-09-30T23:05:16+00:00

DeepSeek said Wednesday that it would open-source software tools that are customized for Huawei’s AI chips, as the Chinese AI firm steps up its efforts to use domestic semiconductors for training and running its models.

The software tools include a Huawei-compatible version of TileLang—a ...

Banks Including SocGen, MUFG Pull Back From Data Center Loan Deals [Briefing]

2026-09-30T21:04:00+00:00

Some of the largest lenders to the artificial intelligence build out are becoming more choosy about the deals they take on as market volatility and rising yields give lenders reasons to choose safer deals over more speculative ones.

Lenders including Société Générale, Sumitomo Mitsui Banking ...

Why Shopping Agents Are a Security and Payments Headache [Special Report]

2026-10-01T02:03:05+00:00

When venture capitalist Yuechen Zhao asked Instinct, the popular consumer AI agent, to look up hotels for him on Booking.com this month, he discovered something unsettling.

Zhao, a partner at venture firm Informed Ventures, said the agent promised it could fill out the booking form for him if he entered his credit card information into the Vault, Instinct’s credential manager, which allows users to save and autofill their payment and other credentials such as login passwords. The agent, when prompting users to share the information, says it won’t actually see sensitive details like card numbers directly.

Zhao tested the process with a fake credit card number, and challenged the agent to see if it was technically feasible to read the numbers back. Turns out the agent acknowledged it could indeed see his credit card and passwords and read them back to him in plain text by running a Javascript on the webpage.

“These companies have been doing a lot to try to protect user information and try to create a magical functionalist experience for users,” Zhao said. “But it does raise the question, just in general, for these personal agents—how should people think about security?”

Asked about the episode, a spokesperson for Instinct, a Sequoia Capital–backed startup that raised at a $10 billion valuation this week, said, “Securing user data is a top priority for us, and users are always in control of their data and credentials.”

AI companies have been promising seamless, AI-powered shopping for several years. But that’s mostly been hard to pull off at a large scale, with OpenAI’s high-profile shopping pullback earlier this year showing how hard it is for AI companies to build a shopping business.

Now, there’s been a sudden spate of new agents gaining traction—most recently Meta Platforms’ Muse and Instinct—that can go shopping and handle transactions for users. That has forced merchants, payment providers and AI companies to confront the potential security and financial risks of having an agent make a purchase instead of a human.

Payment firms have been jumping in to support various levels of agentic transactions, from fully autonomous shopping to transactions where a human is still pushing the final “buy” button. Stripe has partnered with both Meta and Instinct to enable agents to make payments. PayPal announced last week it’s also teaming up with Muse to enable checkout with PayPal’s merchants. And Shopify is rolling out a one-click checkout button in Muse to allow users to pay with the cards and shipping address already in their Shop Pay account.

Still, the path to having agents seamlessly pay for purchases is not easy. By default, online shopping websites have been designed to block bots to reduce fraud ever since the emergence of eBay and Amazon in the ’90s. Those same protections can make it hard for shopping agents to navigate sites and make payments.

“For 30 years, we’ve screamed, ‘Bots are bad, bots are bad, bots are bad,’ and all of a sudden, in 2025, we started changing it around, saying, ‘Some bots are good,’” said Dan Coates, product management director at ACI Worldwide, who leads strategy for providing payment platforms for merchants such as Wendy’s. “We spent 30 years building up this infrastructure that blocked all the bots.

“That’s a very significant change,” he said. “We are having to figure out, how do we poke holes in the supposed firewall? How do we let the good ones in and block the other ones?”

Visa and Mastercard, for example, have introduced an agent registry that can verify agents and distinguish them from bots. Such a feature “enables merchants to verify that they are interacting with a legitimate AI agent and [shows] who is behind it,” said Rubail Birwadker, global head of growth at Visa.

Among the risks, preventing an agent from misspending money is the biggest concern. Payment firms have introduced ways to set caps on spending to prevent agents from making an accidental splurge without users signing off first.

Stripe, for example, can issue single-use virtual cards for Muse agents with a spending limit based on the specific approved purchase. The virtual card is linked to the cards and bank accounts already stored in a user’s Stripe Link wallet, but the agents don’t see the underlying payment details. In a blog, Meta said Muse has no visibility into people’s passwords or payment methods, and any credentials a person shares, including passwords a person types into the browser themselves, go into secure storage, so Muse can use them without seeing them.

Then there’s the question of who bears the cost if a shopper disputes a charge. A consumer using an agent to buy something may say their agent made the wrong purchase or didn’t follow their instructions, when initiating a charge reversal request with their card provider. Typically, if a cardholder disputes a charge, issuing banks will make a call on whether to ask the merchant to return the money. The process often involves a lot of back-and-forth over disputes in which merchants may fight over who has to eat the loss.

How disputed charges play out can depend on what kind of evidence merchants have that a shopper really meant to buy something and the goods came as promised. For instance, for in-person shopping, a merchant might show receipts, shipping confirmations or customer communications.

Card networks have started introducing new measures to check what users authorize their agents to do as well. Mastercard, for example, recently introduced a process called “verifiable intent” to record what authority the user had delegated to their agent, and whether the agent followed those instructions when making purchases.

“If something were to go wrong and someone did want to charge back a purchase, the issuer would be able to have access to that intent data in our regular charge-back system,” said Sherri Haymond, Mastercard’s global head of digital commercialization. That paper trail would help determine which party, such as the bank, the merchant or the shopper, would bear the cost of a disputed agent payment.

Even with new innovations in managing payment risks, how much merchants and shoppers will embrace using agents for shopping remains an open question. Merchants are worried AI agents could reduce the role of their own sites in the shopping process and eliminate the human browsing that provides a lot of valuable data on consumer behavior. And some online sellers are wary of introducing another company’s technology into the mix.

When Amazon banned Muse from accessing its site, one of its concerns was about undisclosed third parties accessing customer accounts, processing transactions and handling sensitive data without Amazon’s knowledge or consent, the company said.

And consumers too are worried about agents mispending money, which could be a hard perception to change. In a survey by ACI Worldwide and YouGov, only 7% of fashion shoppers said they trust AI to make purchases for them without approval.

Sydney Sweeney-Backed Novig Raised at $2 Billion Valuation [Briefing]

2026-09-30T21:25:49+00:00

Novig, a sports-only prediction market backed by Sydney Sweeney, has raised new funding at a $2 billion valuation, the Wall Street Journal reported, citing people familiar with the matter.

Founded in 2021 by Harvard graduates Jacob Fortinsky and Kelechi Ukah, Novig only offers trading on sports ...

Cracks Emerge in AI’s Debt-Fueled Data Center Boom

2026-10-01T01:57:17+00:00

Bond markets are getting tough for lower-rated borrowers. One example: CleanSpark, which is developing a data center for Meta Platforms, had to offer investors big concessions to land financing earlier this month.

That deal is among the clearest signs yet that financing the data center boom is getting expensive across the board, and investors are getting picky about which new projects they’ll back. The bank loan market too is showing signs of strain, with some lenders like Société Générale, Sumitomo Mitsui Banking Corp. and Mitsubishi UFJ Financial Group becoming more selective in lending to data center projects, people arranging the deals say. Those dynamics mean companies could struggle to borrow for planned data centers, jeopardizing growth plans for the broader AI industry.

Apple Plans Aggressive New Push Into Smart Home [Briefing]

2026-09-30T18:49:13+00:00

Apple plans to launch a smart home hub device on October 13th, Bloomberg reported.

The hub has a roughly 6-inch square display that can sit on countertops or be mounted on walls, the news outlet reported. The device will showcase the capabilities of Apple’s overhauled Siri, which was recently ...

Why OpenAI’s DevDay Looked Like Catch-Up [AI Agenda]

2026-09-30T14:01:35+00:00

While the rest of us were distracted by details within Anthropic’s IPO filing and OpenAI’s plans to raise $30 billion in a new round that could value it around $1.4 trillion, app developers on Tuesday flooded to San Francisco’s Fort Mason for OpenAI’s annual developer conference, where it made a flurry of new product and feature announcements.

All in all, OpenAI’s DevDay felt a bit reactionary. Take “Dots,” an always-on or persistent agent product—which we broke news about earlier this month—that promises to automate complex processes like turning customer feedback into improvements for an app or filling out invoices for employees.

In case you’re counting, this is OpenAI’s fifth attempt to launch an agent product, counting past tries like Operator, Deep Research, ChatGPT Agent and ChatGPT Work. The company said that this time was different, though, because of how good frontier AI like Astra has become (and they might be right!).

Dots looks similar to recent breakout agent products like Instinct, Meta Platforms’ Muse and SpaceX’s Grok Bot, which allow people to talk to agents through text and calls. To be fair, though, OpenAI’s version felt distinctly productivity- and work-focused compared to some of these other examples, especially with features like having Dots work together to create documents and presentations. (Separately, can anyone tell me how to test out all these different agent products? I’m personally overwhelmed.)

Wall Street and Silicon Valley Split Over AI’s Price Tag

2026-09-30T13:01:17+00:00

Shifting economic forces are driving a widening divide between the financial views of Silicon Valley and Wall Street, clouding the outlook for trillions of dollars invested in AI.

Stock market angst has already caused delays for several medium-sized initial public offerings, and it is starting to cool once-red hot sentiment about Anthropic’s giant IPO, which is coming later in the year than investors expected. Two large public investors said Tuesday that they thought bankers should value Anthropic closer to $1.5 trillion, rather than the $2 trillion that some bankers have floated in early conversations. The AI firm is still likely weeks away from an official IPO roadshow.

Exclusive: GPU Cloud Provider GMI Raises $668 Million From Nvidia and Others [Briefing]

2026-09-30T10:55:47+00:00

GMI Cloud, a five-year-old provider of Nvidia chip servers to enterprises, said Wednesday it raised $668 million in equity and debt financing. The new fundraise indicates Nvidia is continuing to help smaller, upstart cloud providers, especially those that plan to boost Nvidia in more ways than ...

Alibaba’s AliExpress Tests Membership Program in Challenge to Amazon

2026-09-30T07:01:09+00:00

AliExpress, Alibaba Group’s international e-commerce site, is testing a new VIP membership program in some markets, as it tries to attract and retain high-spending customers in a challenge to Amazon, according to a person with direct knowledge of the efforts.

The VIP program, called Elite, is available on an invitation-only basis to customers who have spent more than $800 a year on AliExpress purchases, the person said. The company is piloting the program in Australia, the U.K., Germany and Spain, and plans to bring it to more European markets as well as South Korea next month, they added.

Oura, World Labs Show Divergence in VC Exits [Dealmaker]

2026-09-30T01:11:00+00:00

It’s turning out to be a great year for venture exits—as long as you’re talking about M&A.

Take Monday’s news that Advanced Micro Devices would purchase Fei-Fei Li’s World Labs. The $8.2 billion all-stock purchase of the two-year-old AI startup adds to the growing list of such deals this year, which include SpaceX’s purchase of Cursor, Nvidia’s purchase of Hugging Face and Stripe’s acquisition of OpenRouter. The sale is also another win for Martin Casado’s infrastructure team at Andreessen Horowitz, which was World Labs’ largest shareholder and owns nearly 14% of the company, I reported.

An expected IPO gusher, on the other hand, is at serious risk of turning into a drought. On Tuesday, health-tracking wearable company Oura said it was delaying its IPO, previously expected to take place later this week, blaming “market uncertainty.”

Oura’s IPO Delay Signals Broader Market Stall [The Briefing]

2026-09-30T00:07:14+00:00

The IPO market appears to have stalled, which could be a problem for Anthropic. Fitness ring maker Oura’s postponement of its IPO on Tuesday, citing “uncertainty in the IPO market,” followed similar moves by metals producer Amaero last week and Holtec Nuclear Corp. the week before that. Then there’s SB Energy, the SoftBank-controlled power developer for data centers, which made public its IPO paperwork on Sept. 1 (two days before Oura did so) but hasn’t yet begun to market the offering.

You don’t have to be a rocket scientist to figure out what’s going on. In its statement explaining its IPO postponement two weeks ago, Holtec mentioned “headwinds” including rising energy costs, trade tensions, military conflicts and rising interest rates. In short, it’s the world we live in. On top of that, Holtec—which sells to data centers—cited “uncertainty over data center development.”

Google Is Paying About 100 Digital Publishers for AI Overviews

2026-09-29T23:09:42+00:00

Google has been paying roughly 100 digital publishers for how much their content contributes to AI-powered answers to search and other products, according to people familiar with tests. The pilot program signals a further shift by the company toward paying publishers for content that shows up in its powerful search engine.

Google has historically taken the attitude that it shouldn’t pay publishers whose content answers users’ queries, on the grounds that the questions direct traffic to their websites. But the advent of AI-powered summaries, called AI Overviews, has drastically reduced the amount of traffic Google’s search refers to websites, prompting widespread complaints about Google, including lawsuits from some publishers.

President Trump Announces ‘Super Intelligence’ Accord, Renames AI [Briefing]

2026-09-29T22:49:15+00:00

President Donald Trump announced on Tuesday that leading AI companies had agreed to non-binding commitments on AI safety provisions, including internal controls on model alignment and external auditing. The “White House Accord on Super Intelligence,” which Trump posted on Truth Social, was ...

Google, Microsoft Join Industry Group to Standardize Enterprise Data For AI [Briefing]

2026-09-29T23:06:51+00:00

Google is in the process of joining Apache Ossie, a software industry group with members such as Snowflake and Nvidia that is trying to make it easier for AI tools to access data from applications and databases, according to a company spokesperson.

The search and cloud provider will ...

How Investors Are Financing On-Site Power to Break AI Bottlenecks [Special Report]

2026-09-29T18:18:56+00:00

Companies building and operating data centers are hungry for ways to secure the power needed to move ahead with massive AI build-outs. That’s prompted big financing pledges and creative dealmaking for on-site power that can cut data centers’ reliance on electric grids.

Investors are getting in on the crunch through several angles, including large investments in specific projects, such as new on-site power plants. Some investors are also taking equity stakes in companies that provide on-site power, or providing outside capital to finance equipment like fuel cells for end customers such as AI cloud providers.

A Blackstone-led group, including KKR and Apollo, this summer agreed to invest $5.3 billion for a 49% stake in five gas-fired power projects pipeline giant Williams is developing to supply on-site power for data centers. Of the $5.3 billion, $4.4 billion will cover the investors’ share of expected project costs, Williams said.

Though on-site power can be pricier than connecting to the grid, availability and speed is an obvious advantage. Williams, which recently expanded into developing and running on-site generation for data centers, said last month it brought its first 200 megawatts online less than 18 months after commercialization. By comparison, Fitch Ratings estimates that building new infrastructure such as grid-connected power plants and high-voltage transmission lines can take five to 10 years.

Many power-generation deals have been financed primarily on a project-by-project basis. But some investors have also started buying stakes in companies that provide on-site power generation, whether that be gas plants, fuel cells or solar projects, giving investors exposure to broader growth rather than a specific project. For instance, Blackstone and Halliburton in May announced a $1 billion equity investment in VoltaGrid, which builds and operates on-site gas-powered systems for data centers.

“With all the private equity and other money that has come into the market, there’s different structures. And when there’s private equity, they want to have an ownership interest in the developer entity,” said Carl Bivens, a real estate partner at Troutman Pepper Locke.

Meanwhile, the up-front costs of bringing power on-site to data centers quickly has created a need for outside capital to provide financing for customers such as smaller AI cloud providers. That’s similar to other areas in AI infrastructure, where some cloud providers have been relying on special purpose vehicles that use outside capital to finance expensive assets like graphics processing units.

On the power side, Bloom Energy announced last fall that Brookfield Asset Management had agreed to provide up to $5 billion to finance future power projects for AI data centers, with Bloom providing the fuel cells. In June, Bloom announced that Brookfield had expanded the amount of potential financing to $25 billion.

Bloom has relied for more than a decade on infrastructure funds and investment firms to form SPVs that finance deployments of its products, but investor interest accelerated more recently, a person familiar with the financings said.

Investors are also financing on-site power projects for Bloom customers, separately from the broader Brookfield framework. In July, project developer Industrial Development Funding and investment firm Oaktree Capital Management announced a $1.7 billion project investment to install Bloom fuel cells at Nebius data centers. Oaktree is contributing minority equity financing, with MUFG leading senior debt financing, while Nebius will pay Bloom for power.

But the rush of financing for on-site power creates a potential timing mismatch across different parts of what’s needed to get an AI data center up and running. Some power assets can operate for decades, while big AI compute contracts are often much shorter and the GPUs inside data centers can become outdated in just a few years.

For developers and the investors financing power infrastructure, longer contracts that lock in predictable cash flow are preferable. Investors are pushing for longer customer commitments to underpin these deals, increasingly favoring 10- to 15-year commitments from large cloud providers with five-year renewal options, Bivens said.

“It provides a longer runway of guaranteed rent for the investors. It allows [institutional investors] to amortize their costs of the behind-the-meter solution or the data center build,” he said.

Customers of the on-site power may want more flexibility, particularly if they’re awaiting grid power. That points to the bigger question of how much data centers will need to rely on on-site power in the long term. For example, more grid power could become available or GPUs could become more energy efficient, and data centers’ on-site power needs might evolve by the time big contracts expire.

Industrial Development Funding, an infrastructure investment firm and the developer behind the Bloom-Nebius project, says on-site power projects can still be valuable even if the data centers’ power needs themselves change. The owners of the project could sell power into broader markets down the line, said IDF CEO Nik Nunes. For instance, if a data center tenant moves on when its initial contract term ends, the power asset could be mostly paid off but could still generate revenue.

"If I have a power block that is significantly amortized in six to eight years, I have a lot of optionality to sell that power into the grid or into other markets,” Nunes said.

Converting to Convertibles

Convertible bonds have been one particularly popular way for tech companies to raise cash in a banner year for financing across the sector, fueled by the AI build-out.

Technology companies have accounted for about 60% of overall U.S. convertible bond issuance so far in 2026, raising roughly $78 billion through Sept. 11, according to S&P Global Market Intelligence. That’s up from around 44% of that market last year. By comparison, tech companies accounted for 50% of equity follow-on offerings, 27% of investment-grade debt and 20% of high-yield issuance so far this year.

The convertible surge comes as financing costs for AI-specific debt have been rising. Investors are swamped with new supply and grappling with uncertainties about how AI investments will pay off. Convertible debt often carries lower coupons than straight corporate bonds since investors get an upside if the stock rises. That can eventually dilute the holdings of existing shareholders, but potentially less so than equity offerings.

“Obviously, there’s the benefits of being not as dilutive as a follow-on. So it has been one of the places of priority,” Chris Fenske, head of capital markets research at S&P Global Market Intelligence, said on a webinar last week. And S&P Global’s total doesn’t count recent convertible issuance like CoreWeave’s $4.2 billion sale last week. But as more companies tap convertibles, that form of financing too could get more expensive—we’ll see whether those firms shift to equity raises next.

New From Our Reporters

Exclusive

Wall Street’s GPU Futures Push Stalls at the CFTC

By Dakin Campbell

Jane Street-Linked Data Center Debt Sours

By Dakin Campbell

OpenAI Announces ‘Dots’ Agents at DevDay [Briefing]

2026-09-29T18:07:38+00:00

OpenAI announced its new agent product called “Dots” at its annual Developer Day on Tuesday, jumping into the increasingly crowded market for AI assistants

The announcement, which confirmed The Information’s reporting this month that OpenAI was working on a productivity-focused agent, follows ...

Use Google Gemini to Research Vendor Products, Services, and Features [Partner Content]

2026-09-29T18:00:15+00:00

Making a vendor recommendation for a major IT initiative can be paralyzing. Every company promises a “seamless” user experience and “robust” APIs, but buyers typically lack visibility into real-world implementation and operations until after they’ve already made a decision.

As a result, many tech leaders default to recommendations from people in their networks, or even avoid making a change at all, preferring to stick with the “devil they know” over risking the unknown.

Google Gemini can help you quickly research vendor products, services, and features, going beyond the marketing materials to surface real-world insights about implementation, integration, and total cost of ownership.

Step 1: Vent Your Frustrations

Open up a new chat in Gemini and list off everything that is driving you mad about your current product or platform so Gemini has comprehensive context about the problem you are looking to solve. If you’re looking for a new cloud vendor, the problem might be unpredictable costs, or it might be unacceptable levels of latency. Those are two very different problems, and each might be best solved by two different vendors.

Don’t sugar-coat here. Mention user adoption issues, unhelpful account managers, broken integrations, and anything else that has you considering a change. If Gemini knows what’s currently going wrong, it will be better equipped to make best-fit recommendations.

Step 2: Generate Recommendations

Give Gemini some basic background information about your company and what technology you’re looking to replace. Include your frustrations from Step 1.

Here’s a sample prompt:

I am the CIO at [company]. We are currently evaluating vendor options to [replatform our intranet]. Right now, we use [vendor], but we are looking to switch because [we are unhappy with limited functionality and account support]. Suggest additional vendor options to consider and include descriptions of their product and services and key features.

Gemini will evaluate the scope of the change you are looking to make, consider your specific context, and offer several options. If you explain that you lead the IT department at a global supply chain company with 14,000 employees, for example, Gemini will account for organization-specific considerations like mobile access and integration with shift-scheduling systems.

Review Gemini’s initial shortlist to identify top contenders, eliminate any obvious mismatches, and move forward with your finalists.

Step 3: Dig Deeper

Gemini’s initial responses will focus on the hero features and selling points of several options, drawing from various sources, including vendors’ own marketing materials. This means you may need to continue the conversation to fully address potential friction points like integration, pricing issues, and scalability challenges.

Follow up with Gemini, this time asking it to take a more skeptical look at your top choices. Here, you can ask pointed questions about specific details and more fully explore likely total costs. You can even prompt Gemini to scour the web for user reviews, helping you to discover stories from previous customers who had subpar experiences.

Try prompts like these:

What are some of the known integration challenges when connecting [Vendor] to [System A] and [System B]?

Summarize the three most common complaints that enterprise IT teams express about [Vendor’s] technical performance or customer service.

What is the realistic enterprise deployment timeline for [Vendor], as opposed to their estimated sales timeline? Are there any configuration issues I should watch out for?

By asking these types of questions, you can push Gemini to go beyond generic feature lists and give you more granular information that gets you closer to a decision.

Step 4: Compare and Contrast

Chat with Gemini for a bit. Go back-and-forth until you feel like you have a comprehensive understanding of the benefits and potential tradeoffs for each vendor, including how each will fit with your existing environment.

Then, ask Gemini to summarize its findings in a structured table. Try this prompt:

Create a matrix comparing [Vendor A] and [Vendor B] across deployment timeline, integration complexity, and potential friction points.

Gemini will generate a clean table that helps you cut through the weeds of your sprawling research session, take a big-picture view of your options, and more clearly how vendors compare with one another on your most important criteria.

You and your team will still need to make the final recommendation to your leadership team. But now, you will have the real-world information you need to help you make (and defend) your choice.

OpenAI in Early Talks to Raise $30 Billion Before an IPO [Briefing]

2026-09-29T17:57:54+00:00

OpenAI is in early talks with investors to raise a pre-IPO round and is targeting raising $30 billion, according to a person familiar with the matter.

The ChatGPT maker could seek a valuation around $1.4 trillion, though it hasn’t signed a term sheet yet with any investor, the person said.

The ...

Microsoft Tears Down Its AI Data Wall [Applied AI]

2026-09-29T16:41:02+00:00

Microsoft’s biggest products have generally thrived by letting other companies build on top of them. So Microsoft’s decision to join Apache Ossie, a 1-year-old software industry group aiming to make it easier for AI tools to access data from applications and databases, might not seem like a big deal. But actually it is.

Just four months ago, Microsoft moved to block partners from connecting their data management tools to Microsoft’s Power BI, which companies use to analyze data in charts and graphs. In doing so, Microsoft appeared to be protecting its own data management product, Fabric, at the expense of rival firms such as Databricks and Snowflake.

In other words, Microsoft was opening a new front in the AI data wars, part of traditional enterprise app firms’ efforts to stop customers from using AI from other companies to access data they store in traditional apps.

After our report in May, Microsoft took heat for its decision. We don’t know what’s happened behind the scenes since then, but the company must have had a change of heart. And its decision to join Ossie is in line with CEO Satya Nadella’s strategy of building products that worked those of rivals, such as developing a mobile version of Office for Apple iOS and Google’s Android operating systems over a decade ago.

Salesforce made a similar transition in the past couple of years and now is making it easy for customers to use AI tools such as Claude to access their Salesforce data—for a price.

Meta Adapts Muse Agent For Small Businesses [Briefing]

2026-09-29T16:09:50+00:00

Meta Platforms is adapting its AI agent Muse for small businesses, adding integrations with workplace software including Salesforce’s Slack, Intuit QuickBooks and Zoom.

In a blog post Tuesday, the company said Muse can connect to small-business owners’ professional Instagram accounts, Facebook ...

OpenAI Nears $70 Billion in Annualized Revenue [Briefing]

2026-09-29T15:44:18+00:00

OpenAI is nearing $70 billion in annualized revenue, up around 70% from its revenue pace at the beginning of the third quarter, according to a person briefed on the numbers.

That growth comes as the ChatGPT-maker has doubled down on selling to enterprises. OpenAI has rapidly cut the prices of ...

Anthropic Discloses up to $84.5 Billion in SpaceX Compute Agreements [Briefing]

2026-09-29T15:24:49+00:00

Anthropic has agreements to pay SpaceX up to $84.5 billion to use the rocket and AI company’s Nvidia-based computing capacity through 2029, according to a confidential initial public offering prospectus reported by Reuters on Tuesday. Those agreements can mostly be cancelled with a 90-day notice ...

Amazon Looks to Further Automate Media Buying With New Agentic Tools [Briefing]

2026-09-29T14:23:50+00:00

Amazon introduced new AI-powered tools to help marketers buy ad inventory across the vast variety of media it sells, as it steps up its effort to strengthen its position in the ad tech world.

One of these tools, full-funnel campaigns, is designed for advertisers who already sell on Amazon’s ...

General Catalyst Backs Coding Data Startup Proximal As It Reaches Revenue Milestone [AI Agenda]

2026-09-29T14:01:53+00:00

It’s no secret that leading AI developers are on the hunt for data to improve their models, especially in lucrative areas like software engineering. The insatiable appetite for data has lifted a whole host of so-called data labeling or data curation startups, including older ones like Scale and Surge and newer ones such as Fleet and Micro1.

The latest beneficiary is San Francisco-based Proximal, which is coming out of stealth with $15 million in seed funding led by General Catalyst at a $300 million valuation, the company told The Information exclusively.

Proximal raised its seed round at the end of last year, when it was generating a negligible amount of revenue, co-founder and CEO Calvin Chen said. Chen had cofounded the 40-person startup last fall with his roommate Justus Mattern, who previously worked at compute startup Prime Intellect, which itself is only two years old.

Now, just 10 months later, the company has surpassed $200 million in annualized revenue, he said. (Proximal calculates annualized revenue by taking the revenue generated in the quarter so far and multiplying it by four, Chen said.)

How Nscale’s Unbuilt Data Centers Undercut Its $35 Billion IPO Pitch [True Value]

2026-09-29T21:01:04+00:00

AI chip rental business Nscale, which has filed paperwork indicating its plan to go public, is a puny, relatively new entrant in the neocloud race. At first blush, it looks to hold a lot of promise, posting rapid growth in the past year and boasting a backlog of contracted revenue on par with that of older rival CoreWeave.

But nearly all of Nscale’s $103 billion in contract commitments rely on data centers it hasn’t yet built—or even raised all the money needed to build. That makes Nscale a symbol of the risks inherent in the AI boom. It also suggests the $35 billion valuation for which it is reportedly aiming in its initial public offering looks far too rich. The right valuation is probably less than half that.

Oura Postpones IPO, Citing “Uncertainty” in IPO Market [Briefing]

2026-09-29T11:58:07+00:00

Oura, the maker of a popular ring that tracks people’s health, said Tuesday it was postponing its initial public offering, “despite strong demand, due to uncertainty in the IPO market.”

Oura was scheduled to go public later this week. The company had made public its IPO filing at the beginning ...

Samsung Invests $1 Billion in KKR-Backed AI Infrastructure Firm Helix [Briefing]

2026-09-29T07:25:35+00:00

Samsung Electronics and five of its affiliates are investing $1 billion in Helix Digital Infrastructure, an AI infrastructure venture established in June by private equity firm KKR, the companies announced on Tuesday.

Soaring AI demand has prompted tech companies, energy providers and investment ...

Anthropic’s Confidential Filing Reveals Customer Concentration [Briefing]

2026-09-29T03:21:19+00:00

Anthropic said nearly a quarter of its $4.6 billion in revenue last year came from just two customers, according to a confidentially filed IPO prospectus reported by Reuters on Monday. That’s one metric that could give investors pause ahead of its IPO, expected later this year. The report didn’t ...

Meta’s MongoDB CEO Poach [The Briefing]

2026-09-29T00:07:59+00:00

Mark Zuckerberg is on a roll. Fresh off his triumphant rollout of the Muse personal agent, he signaled his commitment to selling AI to companies by announcing the poaching of CJ Desai, MongoDB’s CEO, as head of a newly created Meta Enterprise Platform business. Desai has had a long career in enterprise software, at companies including ServiceNow and Cloudflare before MongoDB, making him a significant hire.

We still don’t know exactly how expansive Meta Platforms’ ambitions are for enterprise. Will Zuckerberg use some of its data center capacity to go into the neocloud business? Zuckerberg’s announcement—and Desai’s own background—suggest Meta is more interested in selling AI agents and software to businesses. Whatever the case, Zuckerberg must have made a good case to Desai about the scope of his ambitions. Desai only joined MongoDB last November, and presumably he wouldn’t have bolted without a very good reason (and/or a big paycheck). It’s a pity he couldn’t have negotiated a better title than chief enterprise platform officer, however.

OpenAI Will Not Ship GPT-6.1 Astra Due to Safety Concerns [Briefing]

2026-09-28T23:53:11+00:00

OpenAI has decided not to release the AI model that it had intended to release as GPT-6.1 Astra due to its results on safety tests, a spokesperson said Monday.

The model performed worse than GPT-6 Astra—OpenAI’s newest model, released earlier this month—on tests measuring whether it pursues ...

Startup NinjaTech AI Takes Aim at Agent Sticker Shock [Briefing]

2026-09-28T22:44:36+00:00

NinjaTech AI, an early-stage startup backed by Amazon and led by a former Google product management executive, unveiled a new product that promises to help large companies take the guesswork out of forecasting the costs of running agents around the clock to complete long-running tasks.

The ...

Recursive Self-Improvement Won’t Be Winner-Takes-All, Say Founders of Ricursive [TITV]

2026-09-28T23:01:07+00:00

Recursive self-improvement, the idea that AI models could get so good at research and development that they train themselves, without a need for human researchers, has captivated Silicon Valley. Some in the industry have suggested that the first company to achieve RSI will gain all the benefits: after all, once one company manages to cross that threshold, how could competitors still at least partly reliant on slow human labor ever hope to catch up?

But just because the pace of RSI will be exponentially fast doesn’t mean it’s winner-takes-all, argued Anna Goldie and Azalia Mirhoseini on the latest episode of The Information’s AI Deep Dive. They are the co-founders of Ricursive Intelligence, a year-old startup developing AI systems for designing AI chips. They previously worked at Anthropic and Google, where they led work on AlphaChip, an AI system for chip design.

Manus Unveils New Personal Agent App in Challenge to Meta’s Muse [Briefing]

2026-09-28T22:23:06+00:00

Manus, the AI agent startup that recently separated from Meta Platforms, launched on Monday a new standalone app for personal agents that could compete with Meta’s Muse.

The new app, Cue, provides each user with multiple personal AI agents. Each agent has access to its own email, phone number, ...

AMD to Buy Fei-Fei Li’s World Labs for $8.2 Billion [Briefing]

2026-09-28T20:34:34+00:00

Advanced Micro Devices said Monday it has agreed to buy World Labs, the two-year-old maker of models that can simulate three-dimensional spaces, for about $8.2 billion in an all-stock transaction.

The deal is expected to close at the end of this year, AMD said. World Labs CEO and AI researcher ...

Florida AG Seeks to Bar OpenAI from Developing New Models Without Guardrails [Briefing]

2026-09-28T21:14:57+00:00

Florida Attorney General James Uthmeier on Monday asked a Florida state court for an injunction that would bar OpenAI from developing new AI models “without third-party approved safety guardrails” and from allowing minors to use ChatGPT. The injunction would be effective throughout Florida.

...

Exclusive: Instinct Hires Coatue’s Schwerin as Chief Business Officer [Briefing]

2026-09-28T18:39:31+00:00

Instinct, Silicon Valley’s hot AI assistant of the moment, is hiring Ben Schwerin, an investor at Coatue Management, as its first chief business officer.

In Schwerin, the year-old company—which Monday said it raised $1 billion at a $10-billion valuation—is getting a hot consumer tech connection ...

Nvidia Authorizes $150 Billion Stock Buyback Increase, Largest In History [Briefing]

2026-09-28T15:48:47+00:00

Nvidia’s board approved adding $150 billion to its stock repurchase program, bringing the total amount currently authorized to $235 billion.

Nvidia said it is the largest stock buyback authorization increase in history, and said it expects to buy back all of that $235 billion in stock by the end ...

Nvidia Releases Products To Stop AI From Going Rogue [Briefing]

2026-09-28T15:47:05+00:00

Nvidia on Monday released new software and hardware products to prevent hacking incidents such as the breach in July of Hugging Face and other companies by OpenAI’s AI. Other AI developers such as Anthropic and Google have disclosed similar hacking incidents involving their AI agents, which took ...

Exclusive: LiveRamp Expands OpenAI Partnership to Let Brands Target ChatGPT Ads [Briefing]

2026-09-28T15:43:37+00:00

LiveRamp, which helps advertisers use customer data to target and measure ads, said Monday it is expanding its partnership with OpenAI so advertisers can target their ads in ChatGPT based on customer data they collect, a feature already available with Google and Meta.

ChatGPT currently matches ...

Meta Taps MongoDB CEO to Lead New Enterprise AI Division [Briefing]

2026-09-28T15:42:24+00:00

Meta Platforms is launching a new division to sell its AI tools to businesses, tapping MongoDB Chief Executive Chirantan Desai to lead the unit as chief enterprise platform officer.

In a post on X Monday, Meta CEO Mark Zuckerberg said the new division, called Meta Enterprise Platform, will focus ...

SpaceX Launches Starship Into Orbit For The First Time [Briefing]

2026-09-28T15:22:32+00:00

SpaceX launched its Starship rocket into orbit on Monday morning in a historic first for the company.

It was the 14th test flight of the massive rocket ship but the first that has reached orbit. The rocket released 26 of SpaceX’s’s new Starlink V3 satellites about 10 minutes after reaching ...

Why AI's $15 Trillion Buildout Is Rewriting the Rules of Data Center Financing [AI Infrastructure]

2026-09-28T15:01:51+00:00

The unprecedented scale of capital and resources dedicated to the AI buildout is breaking the mold of how data centers get financed and constructed. It’s also raising new questions about how investors assess their risk, as new data and my interviews over the last week showed.

At The Information’s AI Agenda Live last week, I asked two experienced infrastructure developers—Ricard Boada, co-founder and CEO of new computing and funding platform Volta, and Chris Dolan, chief data center officer of developer Crusoe—how approaches are changing to keep up with this velocity of spending and building.

Valuations for Most Robotics Startups Will Fall By 2030, Says Vinod Khosla [AI Agenda]

2026-09-28T14:02:02+00:00

Vinod Khosla has been one of the most optimistic investors in robotics, predicting the field will have a “ChatGPT moment” in the next two years. But he is also warning that too much investment and lofty valuations in the hottest robotics startups mean a big shake-out is coming.

“I would say more than half will be below their current valuations by 2030,” he said in an interview, while adding that the ones that are not “will have huge multiples.”

Of course, it’s generally true that most venture capital bets don’t pan out, but Khosla’s comments are notable in light of similar warnings in China about its world-leading robotics industry. Last year, China’s economic planning agency warned that over 150 companies are working on creating humanoid robots, a number so great that it risks creating a bubble in the sector. Humanoid giant Unitree had a rocky debut on the Shanghai Stock Exchange last month, prompting Chinese regulators to tighten approvals for humanoid startups seeking to go public, The Information reported.

Anthropic Moves to Cut Off Discounts When Customers Hit Their Cap

2026-09-28T19:16:50+00:00

Anthropic is flexing its muscle with enterprise customers by taking a hard line on discounts once those customers use up all the tokens they purchased. That’s created an opening for OpenAI to take a more flexible approach.

While both Anthropic and archrival OpenAI provide discounts to customers that commit to spending millions of dollars a year on their products, Anthropic has taken the unusual step of ending customers’ discounts once they hit the usage limits in their contracts, according to managers from three software firms that buy AI from both companies. Customers that reach the cap must negotiate a new agreement or face higher prices, these people said.

Why Oura May Not Be the Healthiest IPO [The Briefing]

2026-09-27T22:08:05+00:00

So much is going on in the broadly defined AI world that it’s easy to overlook other stuff that’s happening. This week, for instance, fitness ring maker Oura is expected to go public, which is an event worth watching. It won’t be as earth shattering as Anthropic’s anticipated IPO, to be sure. But it will be a treat for anyone curious to see how easily overhyped these flash-in-the-pan businesses can be.

That might sound a bit harsh. Oura, one of the leaders in health-related wearables, has made quite the name for itself in the past few years. Revenue more than doubled in its fiscal 2025, ending last September, to $907.9 million and jumped another 74% in the nine months ending June of this year. Oura is generating real cash—$100 million in fiscal 2025. But we’ve been here before. As Arete Research analyst Richard Kramer notes, Oura has parallels to other “single product–plus-subscription models that didn’t end well,” such as Peloton, GoPro and Fitbit.

Anthropic’s Amodei to Dine With Trump at White House [Briefing]

2026-09-27T18:47:41+00:00

Anthropic CEO Dario Amodei is scheduled to have dinner with President Donald Trump at the White House on Sunday, Axios reported.

The private dinner will be the first meeting between Trump and Amodei, Axios reported, and could be a sign of warming relations between Anthropic and the White House. ...

China Weighs Allowing Purchases of New Nvidia Chips by ByteDance, Alibaba

2026-09-27T15:01:26+00:00

China’s government has signaled it could allow some domestic companies to buy a new Nvidia chip built for high-end professional computers, according to two people familiar with the matter, as the country’s AI firms strain for enough computing power to run their chatbots and agents.

The Ministry of Industry and Information Technology, which oversees China’s tech sector, recently asked companies including Alibaba Group and ByteDance to report on their plans to buy Nvidia’s RTX Pro 5500 chips, released this month—specifically how many they want to buy and what they’ll use them for, the people said. The ministry has told some Chinese companies the government intends to approve the purchases, they said. Even if that happens, though, it’s unclear when the approvals would come, how many chips officials would allow and what standards they would use to decide.

Why I’m Keeping Meta’s Muse on a Short Leash [The Weekend]

2026-09-26T15:04:12+00:00

• The Big Read: A biology contest pitting OpenAI against human scientists goes awry

• Why the big, bad bears aren’t ready to bet big on AI yet

• A gusher of oil money is washing up in Silicon Valley

• Plus, Recommendations—our weekly pop culture picks: “The Journey,” “Roll the Calls” and “Backrooms”

After returning from a mostly internet-free vacation a little over a week ago, I couldn’t believe the online buzz about Meta Platform’s new agent app, Muse. I immediately downloaded it and began assigning tasks to it like it was an intern in an era without labor laws.

I created an agent to monitor the internet for reviews of a new album by the artist Beck. I ordered Muse to keep an eye on Facebook Marketplace for obscure stereo equipment. And I used it to create alerts that flagged new SEC filings by companies I want to follow closely. Muse accomplished all of those jobs effortlessly, and using the app was even—dare I say it?—delightful.

These were all fairly banal assignments that I could have accomplished pretty easily without Muse. If I wanted more out of the app, I would need to give Muse access to things like my email inbox—which it promised to scour to build agendas for the day—and my credit card account, which it could use to flag recurring charges I might have forgotten about.

That’s when I began to wobble. Giving Muse access to some of my most sensitive personal accounts is the kind of thing that would make people like my wife—who is whatever the opposite of “AI-pilled” is—question my sanity. I had already read my colleague Abram Brown’s account of how Muse had double-booked hotel rooms for him. And yesterday another colleague, Jyoti Mann, reported on the discovery of a security vulnerability in Muse that could let a hacker snag personal information.

This is the trust purgatory I predict a lot of people are going to get stuck in when it comes to giving agents the digital keys to run their lives. Overall, I am at the more trustful end of the spectrum when it comes to tech companies. I have a doorbell camera made by Google. I use Amazon for prescription medicines. I figure that the reputational risks for them of violating that trust are great enough so it’s really not in their interest to do so. But the drumbeat of cybersecurity incidents involving AI models, including the OpenAI–Hugging Face intrusion, makes the leap of faith with AI feel especially perilous at this moment.

In the meantime, I continue to conduct what I view as safe experiments with Muse. In fact, I may have discovered what I believe is one of its best use cases.

There are at least a dozen apps on my phone with atrocious user interfaces that haven’t been updated in years. One of them is for a Wi-Fi–enabled thermostat that controls the heat in a section of my house. Could Muse make it easier to use the appalling software for this thermostat?

I gave Muse my login for the web version of the app and ordered it to set a fall and winter heating schedule. It took Muse more than 15 minutes, but ultimately it accomplished the task. I cringe thinking about the electricity that chore required.

Still, I will respect Muse if it can act as my go-between with the ugliest apps on my phone—as long as the risks of doing so are minimal. For the time being, I’ll be keeping Muse on a short leash.—Nick Wingfield (nick@theinformation)

Weekend’s Latest Stories

The Big Read

Inside the Drama Behind a Biology Contest That Pits OpenAI Agents Against Humans

Unease about how AI will alter advanced research—and commotion around OpenAI’s Navier-Stokes triumph—have forced changes to an event envisioned as biology’s Kasparov-versus–Deep Blue moment.

Why Wall Street’s Big, Bad Bears Aren’t Ready to Bet Against AI Yet

Investors increasingly like to point out flaws in the AI economy and highlight moments like Leopold Aschenbrenner’s initial run as “madness.” But they remain deeply wary of placing their bets too early.

How New Mexico’s Oil Windfall Became a $2 Billion VC Bet

A change to how the state’s sovereign wealth fund approaches venture capital has brought a gusher of money to Silicon Valley—and allows New Mexico to seize a piece of the AI boom.

Listening: “The Journey”

Within America today is a subculture and multimillion-dollar industry that can veer into dystopia: That would be the world of professional surrogacy—and one particularly dark crevice of it is exposed in a new podcast from The New Yorker, “The Journey.” (In the sanitizing speak of surrogacy land, some mothers refer to their nearly yearlong process of preparing themselves for pregnancy and carrying the baby to term as “the journey.”)

Over five episodes, reporter Ava Kofman probes deeply into a network involving dozens of women, established by a California couple who accumulated 21 children before authorities raided their home in 2025. (At the time of their arrest, another five mothers were expecting.) The series expands extensively on a piece written by Kofman for the magazine earlier this year, and through taped interviews with the duped surrogates, Kofman lets their voices carry the podcast: Their anguish comes through loud and clear, and hearing them recount their sagas is both thoroughly engrossing and constantly enraging. What’s even more disturbing is the fact that the couple’s exact motive remains unclear. Perhaps they planned to traffic the children. Or maybe they had some other intent.—Abram Brown

Reading: “Roll the Calls” by Ari Emanuel

Ari Emanuel, that great blunderbuss of Hollywood, has many wonderful, colorful stories about his time knocking heads and knocking around with some of the world’s most famous and powerful people. He recounts many of them in “Roll the Calls,” his new memoir—delivering them in punchy, staccato sentences that make Hemingway’s sound like Proust. It is all very fun. And terribly, terribly profane: “Bitch” appears 16 times, “asshole” 22 times, and the c-word six times—including in a portion of the book devoted to Emanuel’s 2022 wedding, in which Larry David, serving as officiant, referred to him lovingly as “The Cunt.” Emanuel’s favorite cuss, of course, is “fuck,” which appears 433 times—as in “Fuck that fucking pirate agent,” “Tom fucking Cruise,” “Kim fucking Basinger” and “Dan fucking Aykroyd,” plus two of my real favorites: “Fucked that whole fucking thing up real fucking good” and “Fuck your doves.”

I mean, what else could you expect from Emanuel, co-founder of what is today the talent agency known as WME? His antics are what inspired Jeremy Piven’s foul motormouth, Ari Gold, on “Entourage,” and Emanuel gleefully advised the HBO series on writing the character. (“Entourage” Executive Producer Mark Wahlberg is a longtime Emanuel client.) He did so purposefully, and he acknowledges that the show boosted his own standing. But then something strange happened. During the peak popularity of “Entourage,” Ari Emanuel found himself quoting Ari Gold’s lines like they were his own original bon mots. Unsurprisingly, the blending of ego and alter ego, as Emanuel puts it, provided much fodder for therapy sessions. (Equally unsurprisingly, his therapist is a recurring character in this memoir.)

With “Roll the Calls,” you come to hear Emanuel continually lambast fellow Hollywood titan and liberal donor Jeff Katzenberg, on whom he places the blame for the Democrats’ 2024 loss. (He also once threatened to sue Katzenberg—a “putz” and “glad-handing phony”—over Lakers floor seats.) You keep on reading to hear him talk about his relationships with Martin Scorsese, WWE ringleader Vince McMahon, Disney’s Michael Eisner and former boss Michael Ovitz, among many others. Elon Musk appears, too, with Emanuel recounting how he tried to broker peace between Musk and Silver Lake’s Egon Durban, a Twitter board member during Musk’s takeover of that business.

As Emanuel tells it, the animating force throughout his life is the word that begins the book: “rage.” He thinks his wrath is a fine thing. “Just look how rage and agent nearly rhyme. As if they were meant to be together,” he writes. “Rage is both catalyst and metric of desire. Rage is the answer to that central question: How bad do you want it?” In Emanuel’s case, really fucking bad.—A.B.

Watching: “Backrooms”

It has been a very, very good year for horror movies—led by three biggies: a pair of gross-out chuckle fests, “Obsession” and “Resident Evil,” and freak-deaky “Backrooms,” newly streaming on HBO Max. I enjoyed all of them, but frankly, “Backrooms” shattered my expectations most, to say nothing of how it jangled my nerves. The film is short on explanation and long on vibes, which I appreciate: The most effective cinematic dread often exists in the liminal space between those two elements. And, yes, I’m picking my words carefully here because “Backrooms” is a long descent into a series of liminal spaces—a web of connected rooms that form a Daliesque demimonde of dead-end hallways, half-formed walls and a stop sign sticking out of a carpeted floor. The portal to this netherworld sits in an unlikely location: the basement of a furniture store. And once the shop’s owner, Clark (Chiwetel Ejiofor), tumbles through it, he finds it increasingly difficult to pull himself back to the sad, alcohol-soaked reality of his everyday life in the year 1990; eventually, his therapist (Renate Reinsve from “Sentimental Value”) goes looking for him. They are not alone down there.

All of those horror blockbusters I mentioned are adaptations. “Obsession” redoes “The Monkey’s Paw” with demented, Gen Z cringe. “Resident Evil” is obviously a reboot of the film series based on the venerable videogame franchise. For “Backrooms,” director Kane Parsons had perhaps the greatest challenge: turning what amounts to a longstanding online group art project into a conventional feature film. (The narrative and backstories behind “Backrooms” originated on sites like Reddit, YouTube and—yes, ick—4chan, passed around and modified person by person like Zeusian myths.) Parsons, who is barely over 20, does an admirable job, drawing some obvious inspiration from David Lynch’s “Twin Peaks” and “Eraserhead.” And he leaves himself plenty of room for sequels: I saw “Backrooms” a while ago in theaters, and I still shudder to think what might be found lurking at the bottom of that conference room pool.—A.B.

How New Mexico’s Oil Windfall Became a $2 Billion VC Bet

2026-09-26T13:16:22+00:00

A few weeks ago, Augustus Doricko, founder of the cloud-seeding startup Rainmaker, stood in a park on the outskirts of Albuquerque, N.M., flying a drone in the sky. Clad in a khaki field shirt, he looked at home in the desert heat: 97 degrees, not a rain cloud in sight.

“The conditions here are great,” said Doricko, 26. “And the demand for water’s huge.”

Doricko’s company, which is based in El Segundo, Calif., plans to start a trial program around the Rio Grande river in the coming months, dispatching its drones to induce precipitation with the help of a chemical, silver iodide.

That trial is not the only connection Rainmaker has to New Mexico. The startup has received funding from several venture capital firms that have gotten money from New Mexico’s $75 billion sovereign wealth fund. The fund has funneled some $2 billion into VC firms in the last three years as the state’s oil and gas revenues have swelled.

Why Wall Street’s Big, Bad Bears Aren’t Ready to Bet Against AI Yet

2026-09-26T18:30:58+00:00

Steve Eisman, former hedge fund manager and short seller, has become one of the more prominent critics of the AI phenomenon that is sweeping the markets and, according to some, endangering the world.

Yet Eisman, who initially shot to fame for his trades in the 2008 financial crisis and his subsequent portrayal in Michael Lewis’ “The Big Short,” is hesitant to commit capital to his AI criticism. “I am not ready to make a big short call here,” he told me.

For now, he’s paying a lot of attention to Nvidia, the mother of the AI trade, and its relationship supplying chips to Anthropic and OpenAI. “It’s like the entire chain is dependent upon two companies that lose a ton of money,” he said. He’s waiting to see if OpenAI does actually go public anytime soon: “When you’re a company that bleeds cash—like these companies bleed cash—narrative is very important because you’ve got to raise money all the time.” So if OpenAI keeps delaying its IPO, that might be the catalyst for him to make a short call on AI, Eisman said.

OpenAI Agents Tried to Hack Education Depart. Website Amid Dozens of Misdeeds [Briefing]

2026-09-26T01:02:39+00:00

OpenAI agents attempted to hack the Department of Education’s website but was not successful, according to researchers at AI nonprofit Transluce. An OpenAI spokesperson said the company was reviewing the incident, first reported by The New York Times.

OpenAI agents also accessed public data from ...

Fireworks, Fal Consider New Rounds as Inference Demand Soars

2026-09-25T22:59:45+00:00

Startups such as Fal and Fireworks AI that sell access to AI models and servers have been ringing up sales as developers use them to run models quickly, sparking a new investor rush to pour money into them.

Fal, which specializes in providing such inference services for video and image generation models such as Google’s Nano Banana models, has spoken to investors about raising new funding at a $15 billion valuation, according to two people familiar with the discussions. Talks are early and the company could aim for a higher valuation, of $17 billion to $20 billion, according to a third person.

Exclusive: Meta Bolsters Muse Safety Warning After Security Vulnerability Found [Briefing]

2026-09-25T17:41:55+00:00

Meta Platforms is adding a clearer safety warning within Muse after a security researcher discovered a vulnerability in the AI agent that could let an attacker access a user’s sensitive personal information.

The security flaw, which was flagged by an outside researcher through Meta’s bug bounty ...

Inside the Drama Behind a Biology Contest That Pits OpenAI Agents Against Humans [The Big Read]

2026-09-25T17:00:47+00:00

For months, Jason Kelly, co-founder and CEO of Ginkgo Bioworks, has been planning what was billed as a fun, attention-grabbing science competition. His idea was to pit a top scientist against OpenAI’s latest models in designing a variety of proteins using a technique in which they are grown in test tubes faster than they would be in cells. Kelly’s company, which builds autonomous labs, would supply the venue—a 15,000-square-foot, robot-filled lab at its Boston headquarters. Ginkgo executives would serve as judges of the three-round contest.

Michael Jewett, a renowned Stanford University professor and an expert in that technique, known as cell-free protein synthesis, had agreed to go up against OpenAI. OpenAI’s team, meanwhile, would consist only of AI agents. Jewett would also be permitted to use AI, reflecting the reality that it is already integrated into many labs. He could use any commercially available AI model; OpenAI would have access to its own most advanced, unreleased AI. When Ginkgo initially shared notes with me about the proposed project, it even had a catchy name: “Mike Versus the Machines.”

The showdown was intended to generate the kind of excitement and frisson that comes when humans go up against a new technology—akin to world chess champion Gary Kasparov taking on IBM’s Deep Blue supercomputer in 1997 or Go champion Lee Sedol playing Google DeepMind’s AlphaGo in 2016. (In both those cases, of course, the humans lost.) Hoping to capture a striking moment in the history of biology, Ginkgo had hired a documentary film crew to record the action.

But in the past few weeks, something odd has happened. Plans for the contest have morphed. Shortly before the two sides were supposed to square off for the first time on Sept. 14, Ginkgo postponed the event. For a moment, I thought the company might cancel it entirely. Then, earlier this week, I was told the competition is back on, but the “Mike Versus the Machines” concept has been scrapped.

Federal Court Sides With Pentagon in Anthropic Dispute [Briefing]

2026-09-25T16:48:02+00:00

A federal court ruled in favor of the Department of Defense in a lawsuit over the Department’s decision to label Anthropic a supply-chain risk earlier this year. In a split 2-1 decision, Judge Gregory Katsas of the Washington D.C. Circuit Court, an appointee of President Trump, wrote for the ...

Exclusive: Figma Exec Joins AI Inference Startup Baseten [Briefing]

2026-09-25T15:37:44+00:00

A key Figma executive whose departure helped send the stock down in August is joining the rapidly growing upstart Baseten.

Sheila Vashee, who spent three years at the design software firm as chief marketing officer, left Figma this summer amid an executive shakeup that also saw chief product ...

Tesla’s Optimus Hits Snags in Hands, Suppliers as Scale-Up Begins

2026-09-25T13:01:30+00:00

Tesla has ramped up production of its Optimus humanoid robot roughly tenfold in recent months, but the company is still struggling to manufacture the machines reliably at scale. Its production lines are grappling with problems involving the robot’s intricate hands, automated equipment and supplier constraints, according to people familiar with the project.

Last month, Tesla produced several hundred robots a week, up from a few dozen robots per week during the second quarter when the company was doing small batch testing for production, according to two people familiar with the work. Managers have told staffers the company wants to set up a continuous automated production line capable of producing more than a thousand robots a week by the end of the year, the people said. That is still far short of its eventual plan to produce around 20,000 a week.

Microsoft Launches Revamped Copilot ‘Super App’ with Muse Competitor [Briefing]

2026-09-25T12:50:48+00:00

Microsoft on Friday launched a revamped version of its Copilot app that combines features that had previously been sold separately, including AI coding tools, features that automate tasks in Office 365, and an always-on “Autopilot” agent similar to Meta’s Muse.

The overhaul is meant to make ...

Chinese Leader Xi Calls for U.S.-China Cooperation to Prevent AI Abuse [Briefing]

2026-09-25T10:14:57+00:00

Chinese leader Xi Jinping said during his summit with U.S. President Donald Trump that the two nations can “jointly prevent the misuse and abuse of AI” through more cooperation.

Trump and Xi held talks on Thursday at the White House, and a major topic on the agenda was AI safety and governance.

...

Startup Founded by ex-Tesla Dojo Leaders Nears $10 Billion Valuation

2026-09-25T02:56:57+00:00

DensityAI, an AI chip startup founded just a year ago by former leaders of Tesla’s Dojo supercomputer program, is in advanced talks to raise hundreds of millions of dollars in a round that would value it at $10 billion, according to two people with knowledge of the discussions.

The valuation is high for a company whose chips appear to be early in their development, meaning they could be years away from mass production. But the company’s leaders have told potential investors they’ve secured an agreement in which Amazon Web Services would purchase eventual chips from DensityAI if they meet certain performance requirements, the people said.

Why Meta’s Moment in the AI Sun Won’t Last [The Briefing]

2026-09-25T00:07:26+00:00

If Meta Platforms wants to raise money by selling equity, now is the time. Meta shares have gone on an incredible run lately, jumping 27% since Sept. 8, when the company released its Muse personal AI agent. Meta stock is now up 18% so far this year, a bigger gain than that for Alphabet, Amazon or Microsoft—quite the turnaround for a company that not too long ago was in Wall Street’s doghouse for its seemingly out-of-control AI spending.

It’s no secret why investors have fallen back in love with Meta—the company has done a good job of marketing Muse, prompting lots of glowing reviews and headlines about app downloads. Wednesday’s Connect conference, where Meta CEO Mark Zuckerberg unveiled new virtual reality and AI hardware—including a device called Charm designed around Muse—appears to have reinforced the positive sentiment. But let’s not overinterpret the current state of play. In AI, no one stays on top for very long.

Anthropic Seeks Palantir-Style Voting Control for Seven Co-Founders Ahead of IPO

2026-09-24T21:56:12+00:00

Anthropic is asking shareholders to approve a new corporate structure that would award CEO Dario Amodei and his six cofounders a special class of shares giving them collective voting control in most corporate matters, according to people familiar with the planning.

The new arrangement, which emulates a founder-control structure at Palantir, would grant the co-founders a combined 50.1% of voting power, as long as three of the seven co-founders retain a minimum number of shares in the company, one of the people said. It would mark a major change in Anthropic’s already unusual governance structure, strengthening the clout of a group of founders who own relatively small stakes against the prospect of pressure from outside shareholders once it goes public.

Databricks Acquires Microsoft Excel Competitor Row Zero [Briefing]

2026-09-24T21:38:22+00:00

Databricks is acquiring Row Zero, a five-year-old Microsoft Excel competitor that sells spreadsheet software that’s designed for analyzing massive amounts of data.

The deal could help Databricks attract more non-technical users to Genie, a chatbot it launched last summer that lets business ...

Amazon Gives Merchants Tools to Manage Walmart, Shopify Inventory [Briefing]

2026-09-24T21:37:04+00:00

Merchants that sell on Amazon can now use Amazon’s software tools to manage their inventory at rival retailers, including Walmart, Shopify, eBay and TikTok, Amazon announced Thursday at its annual conference for sellers. Merchants will be able to update their product listings and track orders ...

Businesses Still Don’t See AI Returns, Consulting Exec Says [Applied AI]

2026-09-24T23:02:25+00:00

Here’s something worth paying attention to. According to consulting firm EY, businesses are still not seeing substantial revenue gains or cost reductions from using AI.

“I haven‘t met any clients that say I wanna slow down my spend on AI. But I have seen a bunch of clients say, ‘I don’t know where my ROI is today,’” said EY executive Dan Diasio in a conversation with Jessica Lessin during The Information’s AI Agenda Live Summit.

That’s a striking statement, echoing a complaint from businesses about AI that goes back a year or two. And it highlights why so many software and AI companies are offering discounts and freebies to persuade businesses to use AI tools.

What’s made businesses conscious of the returns they’re seeing from AI is usage-based pricing, which Anthropic and other providers began to adopt more widely over the past year. That pricing approach has translated to much higher bills. Companies began to both use the tools more carefully and use open source models more, as we’ve reported.

Even so, Diasio said that spending on AI now is still “based on enthusiasm as opposed to that evidence.” Noting EY does a lot of CEO surveys, he said only one in 10 of companies can “actually show where the ROI is happening” in their income statements.

He added that “everybody without a doubt would say that AI is significantly increasing the [organization's] productivity.” But from a CFO’s point of view, “productivity needs to translate into one of two things…Are we making more revenue? Or are we taking costs out of the organization?”

Jev Fervor Leads to Talk of Big Valuation Boost [Dealmaker]

2026-09-24T21:06:59+00:00

Silicon Valley has a new AI startup to obsess over—and throw money at. We’re hearing that TypeSafe AI, the startup developing a new kind of model called Jev, has started to talk to investors about raising an enormous round—$1 billion or even more—and that some unnamed investors have offered to put money in at a valuation of $10 billion or more. All this for a startup that said only last week it had raised $40 million, at a valuation, according to PitchBook, of $200 million.

Talks are early and a lot could change over the next few days and weeks. But the buzz is palpable: on Wednesday at our AI Agenda Live conference in San Francisco, at least two speakers brought Jev up, with pretty glowing terms. “Incredibly fast,” remarked Nvidia’s senior director of high performance computing, Dion Harris, while noting the model’s limitations.

This buzz may be driving up its compute costs, making it more amenable to another round—a scenario that just played out with the startup behind Instinct.

One of Tether’s Bank Partners Caught Up in Assets Seizure

2026-09-24T17:52:39+00:00

On the surface, Tether, the largest stablecoin issuer in the world, might appear to be a formidable force in global finance. It owns more than $100 billion in U.S. Treasury bills, mostly held at Cantor Fitzgerald, the investment bank formerly led by U.S. Secretary of Commerce Howard Lutnick.

But a U.S. government seizure of bank accounts held in the name of an obscure company based in Sacramento, Calif., in recent weeks shows how Tether still relies on a web of sometimes unstable intermediaries to facilitate transactions, thanks to challenges it has faced opening accounts directly at major U.S. banks. And that has exposed Tether to problems at those intermediaries.

Former OpenAI Data Center Chief Is Now At Nvidia [Briefing]

2026-09-24T17:44:00+00:00

Chris Malone, OpenAI’s former head of data centers, joined Nvidia as the vice president of Nvidia’s DSX Platform this month, according to his LinkedIn profile. DSX is the Nvidia division that helps customers design and build AI data centers according to Nvidia’s specifications.

Malone left ...

New York Sues Polymarket Over Alleged Illegal Gambling [Briefing]

2026-09-24T16:21:38+00:00

New York has sued Polymarket for allegedly running an illegal gambling operation, expanding the state’s fight against prediction markets after earlier suing rival startup Kalshi and others.

New York Attorney General Letitia James announced the action against Polymarket on Thursday. The state has ...

Oracle Invokes Force Majeure in Aim to Protect Itself From Data Center Cost Overruns [Briefing]

2026-09-24T14:48:32+00:00

Oracle, which is set to lease a New Mexico data center on behalf of OpenAI, sent a notice to the project’s developer asserting its contractual rights to withhold payments if project delays persist, according to Bloomberg.

Oracle sent a notice that cited “force majeure” to a developer owned by ...

Are GPU Loans Safe From Rising Rates? [Finance]

2026-09-24T14:46:35+00:00

The Federal Reserve’s recent rate hike, and rising expectations of another soon, put a fresh spotlight on financing arrangements for the wide array of companies borrowing money to pay for AI data centers and chips.

Rates are rising just as investors have been growing more skeptical about AI-related debt in general, a potentially potent mix for higher borrowing costs for new deals across the board. But in one corner of the AI financing market that might sound the worst off—floating-rate loans used by major GPU buyers—some of the biggest loans have already been engineered to blunt the impact of higher benchmark rates, for now at least.

CoreWeave has borrowed billions in recent years to pay for Nvidia’s graphics processing units via delayed draw term loans, a structure others including Iren and Nscale have begun to embrace too. Many of these deals have hedges built in that are meant to effectively turn much of their floating-rate exposure into fixed rates.

These DDTLs, which are generally drawn down as the GPUs are purchased and deployed, are often initially structured with floating interest rates for at least some of the borrowing capacity. For instance CoreWeave’s largest DDTL, an $8.5 billion facility, is roughly split between floating rate and fixed-rate portions. That means that the interest rate on some of the financing can change frequently, unlike that on a fixed-coupon bond, commonly used in AI data center financing.

The interest rate on the floating portion of many DDTLs is based on short-term gauges like the secured overnight financing rate, which typically moves with changes in the Fed’s benchmark rate, plus a fixed spread on top. SOFR was 3.87% as of Sept. 23, up from a recent low of around 3.5% in May, though it remains below its year-ago level. Some GPU financings use a version called Term SOFR, a forward-looking rate that’s around 4% right now on a three-month basis.

Meanwhile, recent big GPU loans have carried spreads ranging from roughly 2.25 percentage points to 5.5 percentage points added to SOFR-based benchmarks. Spreads can vary widely based on factors like the creditworthiness of the customer ultimately using the GPUs.

But for the floating-rate component, some of the biggest GPU borrowings also have significant hedges against rising rates. Those are generally derivatives such as interest rate swaps, where a borrower pays a counterparty a fixed rate while receiving a floating rate.

For example, CoreWeave was hedging at least 75% of anticipated floating-rate borrowing under a $2.6 billion GPU facility from last year, filings show. The floating portion of the company’s newer $8.5 billion facility, meanwhile, will be subject to a hedging requirement of at least 95% of anticipated borrowing.

And Nscale’s $1.85 billion GPU facility for a data center in Ward County, Texas, requires interest-rate hedges covering between 90% and 110% of its projected floating-rate borrowing, according to credit agreements the company disclosed as part of its recent initial public offering filing. Nscale’s other large U.S. GPU facility, a $1.2 billion loan for a North Carolina project, has a similar hedge requirement.

There are arguably some benefits of the DDTL structure itself. Interest on DDTLs generally starts accruing only once the money is drawn, though borrowers may pay a smaller, separate fee on the undrawn commitment. So a buyer of GPUs can secure a big pool of money but only pay interest as it draws the funds to buy and deploy the hardware.

But all that doesn’t mean borrowers can ignore rate risk entirely, especially when it comes to contemplating new borrowing.

Hedge requirements have already been written into some loan agreements, which limits how much exposure companies can leave floating. But for new facilities, companies or their lenders may still have to decide how much exposure to leave unhedged.

Of course, AI infrastructure providers can look to tap a wide range of financings beyond DDTLs, and in general companies shift between sources of financing based on investor appetite and market pricing. There’s a buffet of options for finding cash that includes vendor financing and private credit as well as bonds and convertible debt, plus issuing equity.

But those aren’t all necessarily well suited to financing GPUs specifically, since borrowers spend money on the GPUs over time. And CoreWeave, for its part, has told investors that its DDTLs are structured so debt payments line up with contractual payments from its own customers.

Spreads have widened on investment-grade AI credit lately, and new issuance has been coming in pricier than the broader market, while some lower-rated AI data center project debt has seen investors demand especially high yields. Convertible debt can keep interest costs down but come with potential dilution for shareholders.

In other words, finding attractive financing may be getting tougher regardless. And part of why AI borrowing is getting more expensive isn’t related to rates—at least not directly. Investors have been swamped with new supply while questions about the ultimate payoff of AI investments remain largely unanswered.

Overheard: Who Buys All the AI Debt?

Jas Khaira, head of Blackstone’s AI-focused investment team, said on Wednesday that there are still some big questions about what kind of investors will step up to continue financing the AI industry’s infrastructure buildout.

“We have not mapped out who’s going to buy all the debt,” he said at The Information’s AI Agenda event in San Francisco, responding to an audience question about whether financing costs will have to rise in order to fund the buildout. “We have some hypotheses, but the honest answer is we don’t know where all of it will come from.”

He noted debt markets for investment-grade companies and the private credit market as potential sources of capital. He said that while public markets are large, debt will “probably clear at a higher price because of the quantity getting pushed.”

And how exactly big companies like OpenAI and Anthropic will finance their compute needs remains uncertain. “A lot of that is going to depend on their ability to go public,” he said, “and their ability to be investment grade over time.”

New From Our Reporters

Exclusive

Wall Street’s GPU Futures Push Stalls at the CFTC

By Dakin Campbell

Exclusive

Luckey’s Erebor Bank Plots Syria Expansion

By Michael Roddan

Jane Street-Linked Data Center Debt Sours

By Dakin Campbell

Q&A

Bridgewater’s Jensen Wants AI Compute Giants Regulated Like Big Banks

By Dakin Campbell

What I Learned From AI Agenda Live [AI Agenda]

2026-09-24T14:01:45+00:00

What a day! I loved meeting many of our amazing subscribers yesterday at our AI Agenda Live conference, where my colleagues and I interviewed leaders from companies like OpenAI, Google DeepMind, Nvidia, Blackstone, Replit, Atlassian and more.

Discussions at the conference touched on topics ranging from whether Google can catch up in the AI race to how companies are managing the runaway costs of AI to who is going to buy the debt being raised to fund the data center buildout. (For a full recap, check out this story from my colleagues here.)

Here are three themes I’m thinking about from my conversations at the conference.

Google, OpenAI and Anthropic AI Safety Group Takes Shape

2026-09-24T13:01:11+00:00

Google, OpenAI and Anthropic are pushing forward with a plan to create a new AI safety-focused standards body on their own, without government oversight, in hopes of launching it by the end of the year or early in 2027, according to people familiar with the matter.

The three companies tentatively plan to name the self-regulatory organization the Standards Authority for Frontier AI, the people said. Some members of the working group have considered an array of well-known figures to be CEO. They approached Sriram Krishnan, a former venture capitalist and top AI policy adviser in the Trump administration, for the position, according to people familiar with the matter.

DeepSeek Aims to Close $7.5 Billion Funding Round by End-October [Briefing]

2026-09-24T10:37:21+00:00

DeepSeek is aiming to complete its second funding round by the end of October as it prepares to go public on the Shanghai Stock Exchange, The Information reported on Thursday. The company’s fundraising efforts may be fuelled by a boost in revenue, which has hit $1 billion on an annualized basis. ...

DeepSeek’s Annualized Revenue Hits $1 Billion as Startup Finalizes $7.5 Billion Fundraising

2026-09-24T12:52:37+00:00

DeepSeek’s annualized revenue run rate has hit $1 billion, more than double the rate of a few months ago, buoyed by a recent price hike as well as continued popularity of its models, according to two people with direct knowledge of the matter.

The latest revenue figure, shared by CEO Liang Wenfeng in a recent meeting with investors, could boost investor confidence as the company finalizes its second funding round and as it prepares to go public in the Shanghai Stock Exchange. DeepSeek is aiming to complete the fundraising by the end of October, with a goal of raising 50 billion yuan ($7.5 billion) at a valuation of 500 billion yuan.

Open Source, Model Price Cuts Keep AI Costs Under Control [AI Summit]

2026-09-24T04:55:43+00:00

Open-source models as well as cheaper model releases from Anthropic and OpenAI are helping business customers rein in costs while getting as much or more out of AI, said speakers at The Information’s AI Agenda Live conference in San Francisco Wednesday.

“The existence of open source [models] adds pricing pressure on the labs, which is great,” said Amjad Masad, CEO of coding assistant Replit.

Still, for many companies, it’s unclear how much value they’re getting out of AI spending.

Google Nears Release of Flagship Gemini 4 AI Model

2026-09-24T03:19:04+00:00

Google is nearing the release of its newest flagship model, Gemini 4, the head of its DeepMind division said Wednesday, a long-awaited development after the tech giant fell behind rivals Anthropic and OpenAI in the AI model race.

Speaking at The Information’s AI Agenda Live Summit, in his first media appearance as the new leader of Google DeepMind, Koray Kavukcuoglu said Gemini 4 is in the early days of a phase of development called post-training, in which a base AI model is refined to behave reliably. He said he hopes the model will be released “much earlier” than the end of the year.

Amazon’s New AI Offer Reflects Discounting Surge [The Briefing]

2026-09-24T00:08:15+00:00

Have we gone full circle on AI pricing? A bunch of tech firms are throwing around discounts and free offers for AI products, with Amazon on Wednesday announcing that merchants selling on its shopping site will get a year’s free use of its Quick Plus AI assistant product. Microsoft, meanwhile, is discounting Copilot AI subscriptions heavily, we reported Tuesday night. Both offers are part of a wave of discounting by various software firms and AI labs such as OpenAI.

We keep hearing from tech executives about the transformative nature of AI. And yet there’s clearly some resistance among businesses to paying full price for it, which seems a bit of an issue given how much money it is costing to build out the infrastructure for running AI on a mass scale. Something has to give. As Apollo Global Management’s chief economist, Torsten Slok, noted in his daily email on Monday, credit investors are buying bonds of the hyperscalers—a reference to cloud firms—right now on the assumption that their operating cash flow will triple between 2025 and 2030. “If this doesn’t happen, then the risk is that the AI trade weakens, with credit spreads widening, capex plans getting cut,” Slok wrote.

Meta Announces New Retail Partnerships for Agent Muse [Briefing]

2026-09-23T23:59:15+00:00

Meta Platforms announced partnerships with Walmart and other retailers for its AI agent Muse days after e-commerce giant Amazon blocked the personalized agent from accessing its shopping site.

At Meta’s annual Connect conference on Wednesday, Chief AI officer Alexandr Wang said the company is ...

Blackstone Exec Says Big Unknowns Remain For AI Credit Demand [Briefing]

2026-09-23T22:17:06+00:00

Jas Khaira, head of Blackstone’s AI-focused investment team, said on Wednesday that the firm is still figuring out which kinds of investors will step in to provide the massive amounts of financing still needed for the AI buildout. The investment giant itself is waiting to see if debt from the ...

OpenAI Agent Hacked Australian Government, Prime Minister Says [Briefing]

2026-09-23T21:34:41+00:00

An AI agent developed by OpenAI hacked an Australian government website in June, Australian prime minister Anthony Albanese said on Wednesday. The AI accessed both public and nonpublic data from the country’s Department of Health website, Albanese told reporters, adding that the country is now ...

Patreon Co-Founder Sam Yam Joins OpenAI to Lead New Creator Division [Briefing]

2026-09-23T21:36:19+00:00

OpenAI has hired Patreon co-founder Sam Yam to lead a new creator product division, Creator Product, Yam announced Wednesday on X, a sign of renewed interest in attracting consumers.

Two other former Patreon executives, head of product Drew Rowny and head of engineering Shannon Ma, are joining ...

Anthropic Says Its AI Helped Discover a Possible New Gene-Editing Tool [Briefing]

2026-09-23T20:23:28+00:00

Anthropic said that Claude helped discover a previously unknown molecular system that the company said could represent a new gene editing tool akin to CRISPR, which has transformed the creation of new gene therapies.

In a post on X, Anthropic CEO Dario Amodei cautioned that the “precise ...

Sam Altman and Dario Amodei Brief UN Security Council [Briefing]

2026-09-23T20:22:56+00:00

OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei briefed the United Nations Security Council on Wednesday as part of the international organization’s General Assembly week in New York. In a session on artificial intelligence and international security, the two executives addressed concerns ...

Former Lone Pine Capital Director Launches $500 Million-Plus Tech Fund [Briefing]

2026-09-23T18:11:40+00:00

Kevin Salimian, former managing director at investment firm Lone Pine Capital, has launched a new firm with more than $500 million in committed capital, according to a person familiar with the firm’s operations.

The firm, Voxel Capital Partners, will invest in 12 to 15 public companies in AI ...

Amazon Gives Sellers 12 Months of AI Software Quick Plus For Free [Briefing]

2026-09-23T18:12:13+00:00

Amazon is giving the merchants that sell goods on Amazon’s site 12 free months of Quick Plus, an AWS AI workplace assistant, Amazon announced Wednesday at its annual conference for sellers, Amazon Accelerate. Quick Plus normally costs users $20 a month.

Amazon is also letting sellers integrate ...

What To Expect at AI Agenda Live; Crusoe and Thinking Machines Labs Strike a Compute Deal [AI Agenda]

2026-09-23T14:02:21+00:00

Happy AI Agenda Live day! For those joining us today at SFMOMA, I’m so excited to meet you all and talk about this exciting, if slightly contentious, moment in the AI industry.

Our staff and attendees will ask top researchers, founders and executives from OpenAI, Google DeepMind and Nvidia some burning questions. Here are three at the top of my mind.

Founders Fund, Khosla Ventures Visit China as Country’s AI Prowess Rises

2026-09-23T13:01:14+00:00

U.S. venture capitalists have slashed investments in Chinese startups in recent years as relations between the two superpowers have soured. But behind the scenes, plenty are taking weekslong trips to meet with the country’s tech founders—a sign of the influence Chinese startups like DeepSeek wield in the hottest parts of tech.

Last month, three partners at Peter Thiel’s Founders Fund—Sean Liu, John Luttig and Joey Krug—visited tech companies in Beijing, Shanghai and Shenzhen. It was the second trip to China Liu had made in the past year and Luttig’s first visit to the country, according to people with direct knowledge of the matter. The idea was not to make investments but to learn about China’s thriving tech sector.

China Leads U.S. in Top AI Talent, Study Finds [Briefing]

2026-09-23T11:16:59+00:00

Chinese AI researchers have increasingly opted to stay and work in their home country, leading China to become the top destination for elite AI talent, according to a study by think tank Carnegie China.

The research found that the share of top AI researchers working in China rose from 27.1% in ...

Shares of Chinese AI Model Firms Fall After Report of Regulatory Probe [Briefing]

2026-09-23T09:25:51+00:00

Hong Kong-listed shares of Z.ai and MiniMax plunged on Wednesday, after The Information reported that China’s internet regulator is probing potential data leaks to Anthropic. Z.ai, also known as Zhipu, dropped 12.4%, MiniMax fell 4% and Alibaba declined 4.4%. The benchmark Hang Seng Index ...

OpenAI Partners With Grab in Southeast Asia for AI Skills Program [Briefing]

2026-09-23T09:27:05+00:00

OpenAI and Grab, a Singapore-based transport and food delivery company, on Wednesday launched a program to train Grab’s drivers, merchants and delivery riders to use ChatGPT for work.

The program covers Singapore, Thailand, Indonesia and the Philippines and aims to train around 30,000 people ...

Anthropic in Talks to Lease Apollo-Controlled Stream Data Centers [Briefing]

2026-09-23T04:49:58+00:00

Anthropic is in early talks to lease up to 1 gigawatt in compute capacity from Stream Data Centers, a developer majority-owned by Apollo Global Management, which could house tensor processing units designed by Broadcom and Google, The Information reported Tuesday. Anthropic could ...

Anthropic in Talks for 1-Gigawatt Data Center Lease With Apollo-Backed Developer

2026-09-23T16:02:55+00:00

Anthropic is in early talks to lease up to 1 gigawatt in compute capacity from a data center developer majority-owned by Apollo Global Management, part of a major effort by the AI company to reduce its reliance on cloud providers.

The maker of Claude has discussed signing on as a direct tenant of sites developed by Stream Data Centers, a 27-year-old company with facilities across the U.S., and filling them with tensor processing units designed by Broadcom and Google, according to people with knowledge of the discussions.

AI, the UN, Xi and Trump [The Briefing]

2026-09-23T00:08:14+00:00

It’s a big week for global leaders talking about AI. We’ve got a bunch of presidents and prime ministers at the United Nations General Assembly meeting in New York—motorcades are creating chaos in Midtown Manhattan right now—and Chinese president Xi Jinping is due to arrive in Washington tomorrow. The top executives from most big tech firms will be at a state dinner on Thursday night with Xi, while some will also be at the U.N. earlier in the week.

You can bet AI will dominate much of the discussion both at the U.N. and in Washington. It’s unlikely anything of substance will emerge, although there’ll be plenty of breathless headlines. On Tuesday morning, for instance, President Donald Trump made news by pronouncing that artificial intelligence would henceforth be known as super intelligence in U.S. government paperwork, because the word “artificial” makes AI sound “fake.” (See below for more on this.) Glad to see the leader of the free world is staying on top of the most important issues!

Exclusive: Microsoft to Boost Copilot Discounts as It Launches AI ‘Super App’ [Briefing]

2026-09-23T00:06:46+00:00

Microsoft leaders told sales staff this week that it is authorizing steeper discounts of 30% to 50% on subscriptions of its Copilot AI software for corporate customers who commit to purchase a large number of seats and commit to additional payments based on their usage of certain features, ...

Barrick Gold Inks Five-Year AI Deal Before IPO [Dealmaker]

2026-09-22T23:29:15+00:00

AI is top of mind for stock investors these days. That means even companies outside tech are beefing up their AI credentials ahead of new listings.

In one such example, Barrick Gold’s North American unit has agreed to a five-year partnership with Avathon, a startup that develops AI-powered software to help industrial companies predict when expensive equipment will fail and monitors mine and factories sites for safety risks, Avathon CEO Pervinder Johar told me.

Barrick is planning to float a minority stake in its North American unit, which produces the majority of its gold. The company has targeted year-end for the listing.

CoreWeave Completes $4.2 Billion Convertible Offering [Briefing]

2026-09-22T23:38:22+00:00

CoreWeave said on Tuesday it had completed a $4.2 billion convertible bond offering, an upsized amount versus what the AI cloud provider initially targeted last week.

The company had said last Thursday that it was looking to raise $3 billion in the offering, and said it had priced $3.7 ...

Anthropic releases cheaper model in first launch since slowdown calls [Briefing]

2026-09-22T22:51:17+00:00

Anthropic announced the release of its newest model, Claude Opus 5.5, saying it performs as well or better than Anthropic’s previous most powerful models, Fable 5.1 and Mythos 5.1, in domains including coding, reasoning, business workflows, and safety. At the same time, it will cost 40% less to ...

Binance Invests $100 Million in Circle [Briefing]

2026-09-22T21:05:04+00:00

Binance has made a $100 million strategic equity investment in Circle, as the two companies enter a new, five-year agreement to promote Circle’s stablecoins USDC on the Binance platform.

Circle will also pay a monthly fee to Binance, based on a percentage of the amount of USDC held through ...

Wall Street’s GPU Futures Push Stalls at the CFTC

2026-09-22T21:04:14+00:00

The launch of futures for rental prices on Nvidia GPUs, which are seen as important to make AI compute an investable asset class, is hitting a snag.

CME Group had hoped to launch contracts as soon as early October, but it won’t have regulatory approval by then. The Commodity Futures Trading Commission, which regulates the futures market, is extending its review as it digs into how such contracts work and seeks comments from the broader industry, according to a letter to the exchange seen by The Information.

OpenAI Announces Cheaper GPT-6 Family Models [Briefing]

2026-09-22T18:00:45+00:00

OpenAI on Tuesday announced GPT-6 Sol and GPT-6 Luna, which it describes as cheaper and faster models in the GPT-6 Astra family.

The company said that the new models show significant improvements compared to their predecessors, GPT-5.6 Sol and GPT-5.6 Luna, in areas like professional work, ...

Exclusive: Meta’s Muse Surpassed 500,000 Users After First Week [Briefing]

2026-09-22T17:49:21+00:00

Meta’s new personal AI agent, Muse, had more than 500,000 people try it out roughly a week after launch, including more than 250,000 daily active users, according to internal data viewed by The Information. Users have also submitted more than 2 million prompts.

The data indicate there has been ...

New Data Show Anthropic, OpenAI and Other Upstarts Are Eating into Software Budgets [Applied AI]

2026-09-22T17:40:49+00:00

We’ve been reporting for months about how companies are shifting some traditional software spending to Anthropic and OpenAI. New data from procurement software firm Zip provides further evidence.

AI providers such as OpenAI, Anthropic, Cursor, and Sierra accounted for 8% of spending by Zip customers in the 12 months that ended in August, up from 1.4% during the previous 12 months. Overall, their software budgets rose by a median of 13% during that period.

The dozens of customers included in the study had an average of 2,400 employees and spent $18 billion for software, including AI, over the past four years. Zip’s customers include Snowflake, OpenAI, Datadog, Cloudflare and AMD.

Why America’s AI Dream Is Failing to Launch [Opinion]

2026-09-22T17:33:43+00:00

In 2022, American dominance in AI looked almost unassailable. The U.S. led in advanced models, semiconductors, capital and talent. Its closest competitor, China, appeared to be reeling from export controls that cut it off from the world’s most advanced process nodes, chips and memory technologies.

Four years later, the tide seems to be turning. Chinese models have gained major global token share; American technology giants are eyeing Chinese logic and memory chips; and America’s grand AI ambitions face growing public backlash.

During the first three months of 2026, local opposition blocked or delayed roughly $130 billion in data center projects. This spring, 71% of Americans surveyed opposed the construction of a data center near where they live. Texas, the fastest-growing state in data center construction, has frozen new grid-connected projects until further impact studies are completed. Data centers have become a bipartisan sleeper issue for the election cycle.

Luckey’s Erebor Bank Plots Syria Expansion

2026-09-22T16:01:08+00:00

Erebor Bank, fresh from becoming a vital new financial bridge between the U.S. and Venezuela, is eyeing a new foreign foray: expanding into Syria.

The bank, founded by Anduril Industries co-founder Palmer Luckey, has told staff and some investors it could help the war-torn Middle Eastern country reconnect to the global financial system, potentially attracting billions in deposits from businesses and consumers, according to people familiar with the strategy. Doing so would make Erebor a remarkably early mover in Syria, given that Washington only recently reopened the country to U.S. firms and investment.

Entertainment-Booking Startup Ande Raises $52 Million From Lightspeed, Redpoint [AI Agenda]

2026-09-22T14:01:20+00:00

Personal AI agents like Instinct and Muse have promised to take away the hassle of booking hard-to-get restaurant reservations, Broadway tickets and sporting events. Now, a startup is coming out of stealth to do the same for corporate entertainment—events, group dining, catering, sporting events and more.

Ande, founded nearly three years ago, says its AI agents help businesses arrange events for employees or customers, handling everything from booking a venue to processing expenses, CEO Lohit Sarma said. The firm’s customers include Cloudflare, Salesforce and McGraw Hill.

The New York-based startup has raised more than $52 million in funding across seed and Series A rounds from Lightspeed Venture Partners, Redpoint Ventures, Duration Ventures, Sierra Ventures and Bain Capital Ventures, Sarma told The Information exclusively. He declined to disclose the valuation.

Software Firms Discount AI to Keep Customers From Anthropic, OpenAI

2026-09-22T13:01:33+00:00

Software firms and cloud providers including Amazon, Microsoft, Figma and Workday are dangling new discounts on AI products to clients and consulting partners exhausted by shifts in pricing.

The special offers come as customers get choosier about which tools they purchase after they increased budgets for products like Anthropic’s Claude Code and OpenAI’s Codex and cut other costs to justify the new spending. Businesses are also grumbling about shifts by software providers to charge them for AI based on usage or how many tasks it completes, which can drive up costs.

China Probes DeepSeek, Moonshot Over Potential Data Leaks to Anthropic

2026-09-22T11:48:24+00:00

China’s internet regulator is investigating DeepSeek and Moonshot AI after Anthropic alleged both companies had been routing sensitive user data to Claude models, according to people with knowledge of the matter.

In a 154-page report published on Sept. 10, Anthropic detailed how seven Chinese companies were using Claude illicitly on a widespread scale. In one example, Anthropic alleged that DeepSeek relayed requests from engineers building a police surveillance system to Claude.

Z.ai Open-Sources ZCode After Data Uploading Scandal [Briefing]

2026-09-22T07:35:23+00:00

Z.ai said that it has open-sourced its coding tool, ZCode, after the Chinese AI firm faced intense backlash from developers who found out that their data was being uploaded to cloud servers without their consent.

“In response to the ZCode product security issues reported by the community, we ...

Alibaba Unveils New AI Chip And Data Center Expansion Plan [Briefing]

2026-09-22T07:27:39+00:00

Alibaba on Tuesday unveiled a powerful new AI chip for training and running models, highlighting the rapid progress in China’s domestic semiconductor capabilities.

During its annual Apsara tech conference, Alibaba said its new AI chip, Zhenwu V900, delivers three times the performance of its ...

Exclusive: Xbox to Cut Hundreds of Jobs, Consolidate Game Studios [Briefing]

2026-09-22T14:06:12+00:00

Xbox is planning to lay off hundreds of employees this week in its second major staff reduction this year, according to someone briefed on the plans. The Microsoft-owned firm also plans to consolidate several of its game studios, this person said.

After The Information reported the planned ...

Meta’s Muse Tries (and Fails) to Disrupt Amazon [The Briefing]

2026-09-22T00:07:36+00:00

Everyone loves a brawl between two big tech firms—even other tech CEOs. Amazon’s block on Meta Platforms’ new Muse personal agent accessing its shopping site, which came to light Sunday night, prompted Palo Alto Networks CEO Nikesh Arora to post on X on Monday that the episode signals “a bigger battle than anyone anticipates.”

He may be right: while Amazon has previously blocked Google and OpenAI shopping bots without drama, Amazon’s action against Meta is a bit more reminiscent of its blocking of Perplexity’s agent, which sparked a lawsuit that is still underway. And this is just the beginning. Arora, for instance, predicted that eventually everyone from Apple to Google and the AI firms will offer similar personal agents. (See our report today on what OpenAI is up to.)

SB Energy Faces Investor Skepticism in IPO, Report Says [Briefing]

2026-09-21T23:27:05+00:00

SB Energy’s initial public offering is facing investor skepticism, The New York Times reported Monday.

The SoftBank majority-owned company is planning to raise $5 billion to $7 billion in a public offering to help develop a massive data center complex OpenAI plans to lease, which is set to be ...

World Models Need More Intelligence, Says Luma CEO Amit Jain [TITV]

2026-09-21T22:47:05+00:00

The AI boom has been built largely on progress in language models, but a different approach has gained enormous attention in the industry this year: world models, which are designed to generate accurate predictions about the physical world, rather than words and numbers.

I sat down with Amit Jain, co-founder and CEO of Luma AI, to discuss the promise of world models—and where current approaches to training them still fall short—on the latest episode of AI Deep Dive, The Information’s new show exploring the hard problems on tech’s frontier.

Smart-ring Maker Oura Files to Sell $2.2 Billion in IPO [Briefing]

2026-09-21T22:40:08+00:00

Oura is looking to sell $2.2 billion in shares during its initial public offering, the company said in its updated S-1 filing on Monday. Oura, which makes smart rings that track biometrics including sleep quality and heart health, could reach a $15.6 billion fully-diluted valuation from the ...

Paramount Settles State Antitrust Lawsuit, Clearing Way for Warner Deal to Close [Briefing]

2026-09-21T20:39:13+00:00

Paramount Skydance settled a lawsuit filed by 12 states that had threatened to stop its $110 billion purchase of Warner Bros. Discovery. The settlement, filed in court on Monday, clears the way for the deal to close.

Under the settlement, Paramount agreed to release at least 30 films into movie ...

OpenAI Releases Proposal for International AI Safety Coordination [Briefing]

2026-09-21T18:11:04+00:00

OpenAI released a proposal on Monday that would create international coordination around AI safety. In a blog post, OpenAI called for national AI safety institutes, such as the U.S. Commerce Department’s Center for AI Standards and Innovation, to set standards around areas including model ...

The Real Cost of AI: A Survey of the Unpredictable Token Economics [Partner Content]

2026-09-21T16:04:42+00:00

For decades, enterprise software budgeting was a predictable exercise: negotiate a subscription fee, project the number of seats you need, move on. AI has broken that model. Pricing tied to tokens — the metered unit behind most AI billing — has turned technology spend into a variable cost, and an unpredictable one.

The Information’s survey of 107 of our readers - senior leaders and professionals (See Methodology) - finds that most companies haven’t gotten a handle on it: 60% say the cost trajectory of their AI initiatives is unpredictable, at the exact moment CFOs, boards and investors are demanding proof of ROI.

60%: “The cost trajectory of AI initiatives is unpredictable.”

This piece draws on that survey and The Information’s Deep Research to examine where AI budgets are actually going, why token costs remain so hard to forecast, and what separates the companies that have managed AI spending from those still being surprised by their bills.

The Bulk of the AI Budget

Outside the hyperscalers, who are pouring billions into data centers, most companies put the bulk of their AI investment — 64% — into software, applications, and API models. (See Chart.) That reflects where ROI is actually realized, according to The Information: most businesses want solutions to specific problems, not infrastructure to manage.

This 64% budget line includes the unpredictability factor, the token, which behaves unlike anything else in the technology budget.

Let’s unpack the costs of different types of software. The classic subscription/SaaS model relies on per-seat licenses for enterprise systems of record and the cloud services that underpin them, such as CRM (Salesforce), ERP/HR and finance (SAP, Workday), and collaboration (Slack).

Packaged AI apps and assistants, like ChatGPT or Microsoft Copilot, are also largely bought per seat. The flat seat price covers token usage but also limits what an AI assistant can do. The main drawback of seat-based licensing has always been waste: many paid seats sit idle or underused.

The token consumption appears in the AI layer of independent digital agents that perform complex tasks rather than simply answering a query. These agents use metered, token-based pricing that scales with usage, making forecasting much harder, and they consume far more compute than basic chatbots.

As AI shifts from chatbots to autonomous agents running in the background, the companies least practiced at controlling token usage are seeing the biggest surprises on their bills. That’s because such agents operate in loops, and a single inefficient prompt can snowball very quickly.

For routine data manipulation and established workflows, traditional software is often still cheaper and more reliable than agentic AI. When budget predictability matters most, AI’s consumption-based pricing is a liability, not a feature.

One survey respondent frames it as a balancing act: “We should be looking for the best of both worlds. Even with a reasonably implemented, small, hosted cloud environment, the cost per transaction for software is nothing like that of pure agent workloads. The cost of going to pure AI can be prohibitive. There must be a balance among human decisions, programmed software, and pure AI.”

The Intersection of Token Consumption and Productivity

Reader experience with AI productivity varies widely — even within the same industry. One respondent building a startup says, “With my agentic AI assistant, I am compressing what took me 3 years to build into 6 months.” A technology lawyer at a large enterprise draws a much tighter line, restricting AI to coding and software-vulnerability remediation. AI is useful for legal research, this reader says, but “the amount of quality control that I have to do with AI-generated content adds to the frustration I feel when having to explain to management why AI can’t solve all our problems.”

Despite the variance, respondents were broadly positive on ROI: 60% say AI increases productivity, more than half say it returns more than it costs, and nearly half report lower internal labor costs. (See Chart)

The intersection of “token economics” and workforce productivity has confounded business leaders and accounting departments. First, some companies have equated token consumption with productivity, a mistake. Token consumption itself is not a proxy for a worker’s productivity. More tokens don’t necessarily mean a better outcome. On the contrary, they can cause AI agents to loop, generating large volumes of responses that can run up compute costs and overwhelm a human.

The crux of the issue is translating token consumption into productivity – not equating the two. Token consumption is an input, not an output. It’s the compute era’s equivalent of billable hours: easy to count, weakly correlated with value. Measure outcomes, not token metrics, with the caveat that outcomes are much harder to measure than tokens.

Some metrics that are more useful than tokens to consider include task-completion units (the emerging standard), which count the tasks completed by its agents rather than the tokens processed; cycle-time compression for a given process; and revenue or cost attribution.

AI Adoption and Regulatory Compliance

“Regulatory compliance is slowing us down” has become one of the most common explanations for sluggish AI adoption. The smallest share of survey respondents, 17%, view regulatory compliance as an accelerant for AI adoption.

As government regulations, data privacy laws, and corporate governance requirements increase, the cost to develop, run, and use AI rises significantly. Ensuring that AI acts safely and complies with enterprise parameters requires more “thinking” and therefore more tokens. This is largely because ensuring that models, especially agentic models, are secure, compliant, and safe requires immense computational overhead.

While legal uncertainty and the number of regulations are genuine concerns, large companies also face the buildout of AI bureaucracy, including chief AI officers, adoption dashboards, token budgets or governance councils, according to The Information. This bureaucracy is a mechanism to deploy AI at scale but it also obscures whether any of it pays for itself. The economics of AI, especially in the era of tokenization, can be confusing and the AI bureaucracy itself is not its cause but its symptom.

Common flaws of AI governance include companies governing AI at the policy layer (committees, principles, training) while risk lives at the identity and permission layer. Most policy frameworks are written as instruction sets delivered to the AI model, while the controls that work (deterministic triggers, permission gates, scoped credentials) are enforced outside the model. On top of that, companies may not always know what they are governing, as vendors and employees are adding shadow AI features.

The right approach inverts the framing most companies start with. The default assumption is a tradeoff — compliance slows adoption, so minimize one to maximize the other. The evidence points the other way: the things blocking adoption are the compliance questions, and the companies moving fastest are the ones that resolved them first. But before doing it fast, they must do it right.

Token Cost Management

Token cost management is where the survey’s biggest paradox appears. A slim majority of companies (52%) say they understand token costs, and about as many know that disciplined prompt and output engineering can substantially reduce them. Yet only about a third say they actually control token costs in practice. (See Chart)

“Token cost management is not yet in line with the growth in usage of AI,” one respondent writes.

Only one in five respondents (20%) believe that buying tokens in higher volumes lowers cost-per-query (see Chart). Yet The Information’s research shows that batching purchases is an effective and increasingly necessary strategy for cutting per-query costs on frontier models, which carry a steep premium over simpler ones. Equally important for savings is selecting token-guzzling frontier models only for tasks that require them.

The Information’s research and reader responses point to five ways companies are managing token costs:

Train users. Teach employees about model selection, how to prevent models from overthinking, and how to prompt and structure requests to achieve the most useful output at the lowest token cost. Also, teach employees when not to use AI.
Match the right model to the right task. Route routine queries to cheaper, open-source models and reserve frontier models for genuinely complex reasoning. “If you select the models best aligned with most tasks, token costs are incredibly low. All the news about token costs concerns frontier models, which are rarely required,” one respondent writes.
Optimize prompts. Keep them concise, strip unnecessary context, and reuse outputs where possible. The Information’s research shows that these “token-minimizing” techniques are being deployed to reduce the volume of data a model must process per request.
Cap and track usage. Track token usage against productivity, not just spend, and set daily or weekly caps when usage is outpacing value. At one respondent’s company, employees don’t have unfettered access to tokens: “Only authorized employees may use tokens, following a defined mandate and an allocated development budget based on a cost-benefit analysis.”
Adopt a systemic approach. Some respondents argue that the fix isn’t behavioral at all. “Cost efficiency is largely determined by systemic, deterministic constraints, rather than individuals voluntarily enforcing constraints upon themselves,” one respondent writes. A systemic approach includes creating a single controlled AI gateway (eradicating shadow AI), introducing hard spend ceilings with approval gates, and making the cheapest path automatic.

Looking Ahead

AI is getting more expensive, not less. The shift from chatbots to autonomous agents built for multi-step workflows is the single biggest driver of rising costs, and token-based pricing means uncontrolled usage shows up directly on the bill. Respondents expect that trend to continue: most see AI-related spending rising across the board over the next 12 months. (See Chart)

That’s happening as the enterprise AI market moves past its experimental phase and into one where spending gets scrutinized like any other line item. Companies that haven’t yet built model routing, usage limits, and hard budget caps into their AI programs are likely to keep absorbing costs they can’t fully explain.

Methodology

Based on a survey of 107 senior leaders and professionals, conducted in June and July 2026.

Company size: Forty-five percent of companies surveyed had revenues under $10 million; 20% had revenues of more than $1 billion; 16% had revenues between $10 million and $100 million; 12% had revenues between $100 million and $500 million, and 8% had revenues between $500 million and $1 billion.

Key industry sectors: Thirty-eight percent of companies were from technology, media and telecommunications; 13% from professional services; 10% from healthcare and life sciences; 9% from financial services & capital markets, and the remaining 29% came from the government, retail, public sector and others.

Respondent titles and roles: Forty percent were executive leadership (including CEOs, CIOs and CTOs) and owners; 31% were senior management (including SVPs and VPs, directors and managers).

OpenAI Develops Features to Counter Grok Bot, Mulls Response to Meta’s Muse [AI Agenda]

2026-09-21T14:31:27+00:00

OpenAI pioneered AI that can take over web browsers and other applications on people’s behalf, and its ChatGPT app brought generative AI to consumers. Now that rivals like SpaceX and Meta are pushing the boundaries by launching personal AI agents that handle multistep tasks, OpenAI is playing catch-up.

For instance, OpenAI is developing features to compete more directly with SpaceX’s recently launched Grok Bot, which helps customers create always-on AI “teammates” to handle routine tasks, manage software, and collaborate in the background, according to a person with knowledge of the effort. OpenAI also has discussed developing a personal AI assistant to compete with one that Meta Platforms recently launched, according to another person with knowledge of the situation.

In developing such products, OpenAI would be repurposing and rebranding some of the existing “agentic” technology it already offers through Codex and ChatGPT. Some customers of these OpenAI apps have already connected them to their email, calendar and other applications so they can automate tasks such as scheduling appointments or communicating with businesses by email, sometimes over multiple days.

OpenAI and Anthropic Neared Deal to Stress Test Each Other’s AI

2026-09-22T18:36:30+00:00

OpenAI is rethinking a range of safety strategies as it responds to fears from employees and others about the dangers its AI poses. One solution could lie in the recent past.

Even before the spate of cybersecurity incidents involving OpenAI’s technology and the dire warnings from industry workers, the company was negotiating a legally binding deal with Anthropic for the companies to stress-test each other’s models, according to a person with direct knowledge of the discussions, which haven’t been previously reported.

CrowdStrike and Okta Trade at Big Premiums Despite Unproven AI Payoff [True Value]

2026-09-21T13:31:29+00:00

Cybersecurity stocks have enjoyed a good run in 2026 as investors bet on the idea that they’ll benefit from companies shoring up their defenses against AI-fueled online threats. The S&P Kensho cybersecurity index is up 41% this year compared to an 11% rise in the S&P 500 over the same period.

But some of the names attracting the most interest, such as cloud-based security firm CrowdStrike and identity security software provider Okta, have soared to valuation highs that should give investors in the sector pause. So far, there’s limited evidence that their revenue growth is getting a boost from new AI concerns.

Amazon Blocks Meta’s Muse Agent [Briefing]

2026-09-21T19:51:34+00:00

Amazon blocked Meta’s Muse personalized AI agent from accessing its shopping site, Amazon confirmed, saying that agents such as Muse “should operate openly and respect provider decisions about whether or not to participate.”

The block appeared to even prevent Muse from browsing Amazon’s ...

DeepSeek Bets Big on Huawei Chips to Bypass U.S. Export Controls

2026-09-21T10:52:59+00:00

DeepSeek CEO Liang Wenfeng told investors that a major priority for his company is to use more domestic chips to train its models, adding that he expects Huawei Technologies to start delivering training chips to DeepSeek as early as the fourth quarter this year, according to two people with direct knowledge of the matter.

DeepSeek’s growing adoption of Huawei chips during AI model training—a process that requires more powerful chips compared to running the models—demonstrates how both companies are joining hands to break away from the stranglehold imposed by U.S. export controls on advanced chips. Liang’s most recent remarks, made during a closed-door, in-person meeting on Sunday, came as DeepSeek finalizes its second funding round, which aims to raise 50 billion yuan ($7.5 billion) at a valuation of 500 billion yuan.

Alibaba Appoints New Head of Qwen LLM Team [Briefing]

2026-09-21T06:29:46+00:00

Alibaba Group has appointed Dayiheng Liu, one of its senior AI researchers, as the head of its Qwen large language models, according to two employees with knowledge of the matter.

Liu’s appointment has brought more clarity to the Qwen team’s leadership after multiple rounds of reorganization ...

U.S. and China Discussed AI Safety System, Bessent Says [Briefing]

2026-09-21T01:18:17+00:00

U.S. Treasury Secretary Scott Bessent said Sunday that the U.S. and China had discussed setting up a mechanism called the U.S.-China AI dialogue to address potential threats.

Speaking to reporters at the JPMorgan headquarters in Manhattan at the conclusion of a day of talks with Chinese ...

Tech’s New Trend: Ugly Spectacles [The Briefing]

2026-09-20T22:04:50+00:00

It’s that time of year again—when geeky men stand on stage wearing ugly glasses. Last week, Snap CEO Evan Spiegel gave a detailed demonstration of his forthcoming Specs augmented reality glasses, which were first unveiled in June. This week, it’s Meta Platforms CEO Mark Zuckerberg’s turn. Meta holds its annual Connect conference, where it shows off new hardware, with Zuckerberg’s keynote scheduled for Wednesday. And you can bet he’ll be on stage wearing at least one pair of glasses.

The Information has already scooped one of the possible announcements—about a camera-free pair of smart glasses to go on sale in October. According to our story last week, by my colleague Jyoti Mann, Meta will also unveil new AR glasses. Here’s a question, though: What happened to the AR-like glasses that got so much attention at last year’s Connect?

How Nvidia Is Trying to Solve the Data Center Power Bottleneck [AI Infrastructure]

2026-09-20T16:01:44+00:00

Nvidia is tracking “every single gigawatt of land, power and shell around the world, literally everything on the planet,” CEO Jensen Huang said at Goldman Sachs’ annual tech conference earlier this month. “We know where everything is.”

Why track power so intently?

Nvidia sees power availability as a key barrier to getting its AI server chips online as soon as they ship, as we’ve chronicled in this column over and over. While securing power is the first step to any AI data center project, even before getting the land or chips, developers that work with Google, Microsoft, Oracle and others fear a glut of unused chips due to an eventual power shortage.

Why Apple Is Betting on Foldable iPhones [Sunday Insights]

2026-09-20T15:01:35+00:00

When new Apple CEO John Ternus showed off the first version of a foldable iPhone at an event earlier this month, there was a good chance many Americans watching clips of the video stream had never thought hard about buying such a device.

But Apple is betting that the same demand that’s driven double-digit growth of such phones outside North America for Huawei, Samsung and others will boost sales of its flagship iPhones. Demand for such smartphones, whose hinged sides open to one large glass screen, should also make an expected drop in global smartphone shipments this year less severe than it otherwise would have been.

Jane Street-Linked Data Center Debt Sours

2026-09-20T14:01:26+00:00

Secretive Wall Street trading firm Jane Street has become famous in the past couple of years for its sizable AI-related spending, as it has struck big cloud deals to handle its AI computing needs and made investments in neocloud CoreWeave. Yet even a Jane Street–related deal for data center debt hasn’t won over investors.

Debt sold by a venture building a data center to be leased by Jane Street has quickly soured in secondary trading, according to Financial Industry Regulatory Authority data. Yields on the bonds have risen to around 11.3%, more than 2 percentage points above where the debt first sold in August. That deterioration has been sharper than for other recently issued AI data center bonds, with investors demanding far higher compensation to hold them versus government debt.

President Trump Announces ‘AI Force,’ Will Appoint New Czar [Briefing]

2026-09-19T20:01:40+00:00

President Donald Trump on Saturday said he was creating a new “AI Force,” that would be similar to his Space Force—the arm of the Defense Department focused on space combat started by Trump during his first administration—and he planned to appoint a new AI czar to lead it.

On his Truth Social ...

Dario Amodei's Tabloid Era [The Weekend]

2026-09-19T15:08:03+00:00

• Exclusive new list! Introducing Tech’s 50 Most Influential People

• Plus, Recommendations—our weekly pop culture picks: “Embedded: Where Is Austin Tice?,” “Valley of Death” and “Hope”

Honestly, I almost feel a little sorry for Dario Amodei. That essay-writing, sweater-loving nerd is getting the treatment normally meted out to grim-eyed murderers and sex-romping celebrities.

Put another way: The New York Post has discovered him in earnest. Well, him and effective altruism. By my count, the outlet has published more than a dozen stories mentioning Amodei or effective altruism in the past week. One had a headline describing Anthropic’s staff as “cult-like,” the same label it had applied a while back to effective altruists in general. (C’mon, fellas! Invest in a thesaurus.)

A second Post headline gave prominent attention to President Donald Trump’s recent criticism of Amodei’s suggestion for an AI industry slowdown: a “SICK conspiracy,” Trump called it, the capital letters all his. A third headline described METR, the AI research outfit Amodei says he’ll use to independently assess his company’s safeguards, as a group of “super-woke globalists.” And a fourth headline on an editorial board piece thundered out: “Beware the ‘effective altruists.’”

Eeesh. I wonder what would excite the mouth breathers at the Post more: another Amodei doom dump—or JonBenét Ramsey’s parents finally saying they did it?

Sure, Amodei has steadily attracted more and more—and more—media attention over the past five months, and so has effective altruism, which obviously got a lot of attention in the Sam Bankman-Fried days, too. But there’s something about watching Amodei and the EA set get turned into daily tabloid fodder right now that strikes me as an important marker of where the broader world’s conversation about AI stands. It’s loud, wacky as heck—and candidly, I worry it’s a precursor to something darker.

I sat here trying to wrap my mind around it, and I started to get a headache like the one that sits at the bottom of a bourbon bottle. The Post is channeling a populist antipathy for Amodei and effective altruists that’s rooted in a dislike for their AI apocalypse talk. Yet at the same time there’s populist concern that AI may very well invite Armageddon: A recent survey by Politico found that 63% of U.S. adults believe AI could eventually destroy the world. So in sum, there’s widespread concern about destructive AI—and scorn for the people who want everyone to focus on it. As I said, it’s wacky.

Put aside AI catastrophe for a second. The bigger reason people dislike Amodei and the effective altruists is because they perceive them as the influential, liberal elite. It’s this potent derision that really animates the Post, and I find it unsettling. It’s the same type of addled, conspiracy-minded resentment that makes people still believe in Pizzagate and rigged voting machines.

What I find especially facile is the assumption that effective altruists and those who share their beliefs actually do wield gigantic influence over AI. I’d argue that they actually don’t. Articles by the Post and other outlets portray effective altruists as a wide cabal, but they only ever really mention the same three people over and over: Amodei; his sister, Daniela; and Facebook billionaire Dustin Moskovitz.

Look, three people simply do not make an all-powerful cabal! I keep thinking about that disconnect between assumption and reality—probably because I’ve devoted enormous thought lately to considering who really does have clout in AI as we put together our debut list of Tech’s 50 Most Influential People. Yes, the Amodei siblings are on the list, and they’re the No. 1 and No. 2 figures at a gigantic company. But that’s…it. Not a single other person on our list belongs to the effective altruism crowd.

I don’t write this as a passionate Amodei fanboy, nor as someone who finds effective altruism especially appealing: It’s an ideology that can be channeled into single-minded glumness when someone swallows the whole pill bottle and commits fully to it. A conversation with an effective altruist about the future is about as much fun as discussing foie gras with a vegan. It’s just that the concerns about Amodei and the effective altruists are beginning to strongly remind me of previous worries about lizard people taking over Congress.

Perhaps I’m making too much of those Post stories I mentioned. Still, with nearly 100 million readers a month, the Post is big enough to inspire other outlets to follow its lead, covering Amodei and AI with a sensationalism rooted in gassed-up anxiety and quarter-truths. Amodei’s the guy getting tarred and feathered today. It’d be a mistake to assume he’ll be the last.—Abram Brown (abe@theinformation.com)

Weekend’s Latest Stories

The 50 Most Influential People in Tech

The technology industry runs on influence. Its presence shows up in big, attention-grabbing moments—but much of it is dispersed in cloudy mystery, producing subtle ripple effects missed by casual observers. The CEOs of the largest companies certainly have clout. But so do a host of behind-the-scenes investors, renegade founders and rising-star executives.

These are the people we at The Information think are most likely to influence the industry over the next decade. For more on how we created this list, here’s an extended explanation of our methodology.

Listening: “Embedded: Where Is Austin Tice?”

In 2012, Austin Tice, a former U.S. Marine who’d decided to try his hand at freelance foreign correspondence, went MIA in Syria. And he’s still missing 14 years later. His mother clings to hope that he’s alive. Every president since Barack Obama has vowed to bring him back. But if he has somehow survived, why hasn’t he surfaced since the Assad regime fell in 2024? Josh Baker and Sara Obeidat, a pair of veteran Middle East reporters, set out to try to answer that question in “Embedded: Where Is Austin Tice?,” an NPR-BBC joint production.

They’re well sourced and determined, and by episode 2, they’re in an infamous prison retrieving never-before-seen files that the Syrian government kept on Tice. This involves a little bit of Bond-esque derring-do from Baker, who risks a five-story fall from a window ledge to get from one barricaded room to another. Additionally, I appreciated that they could display obvious empathy for Tice’s loved ones and stop short of lionizing him. He was clearly ill prepared to be in Syria. It was his very first attempt at freelance reporting. His military experience had emboldened him to go and made him a popular talking head for TV networks looking for on-the-ground analysis, but it also made many Syrians think he was in actuality an American spy posing as a journo.—Abram Brown

Reading: “Valley of Death” by Sharon Weinberger

A year or so before Donald Trump’s reelection, Sharon Weinberger, then national security editor for The Wall Street Journal, visited the offices of an unnamed but prominent venture capital firm. Yes, the tech world invasion of Washington that has happened in the second Trump administration was still far in the future, but even then, Weinberger had begun noticing how Silicon Valley and Washington were intersecting more than they had in the past—particularly when she looked at the venture capital going to defense tech startups.

At the VC firm, Weinberger asked the investor why the private sector would invest in technology like hypersonic weapons given the astronomical costs and uncertain business models. “I honestly don’t care about the product,” replied the investor. “It has nothing to do with the tech.” Instead, he argued, the success of the products—and his own returns—depended more on government relationships and lobbying, which was increasingly a strength for Silicon Valley firms.

By Weinberger’s own account, she found herself thinking often of that investor’s sentiment as she wrote “Valley of Death,” an account of how Silicon Valley strayed from its Cold War national security roots before returning to the corridors of the Pentagon in recent years.

Weinberger offers an engaging history, with details of untold episodes in the intertwined relationships between Silicon Valley and Washington, like Steve Jobs’ brush with the CIA. As she goes along, she illustrates how the tech titans of defense contracting such as Elon Musk, Palantir’s Alex Karp and Anduril’s Palmer Luckey have strong-armed their way into the staid field. And she details how they’ve been adept at talking up their companies, though their products haven’t always lived up to the hype—especially in Ukraine.

Weinberger sees shortcomings in their “techno-optimism,” as she puts it. She thinks it could lead to ruinous scenarios not only for the firms’ financial backers but also for the U.S. armed forces at a time of heightened conflicts and supply chain constraints. Weinberger is a measured and well-traveled chronicler of the Silicon Valley arms race, raising the same concerns I’ve heard some investors share out loud.

By the time she ends with President Dwight Eisenhower’s timeless warning of the dangers of the military-industrial complex, or “subordinating national security to private interests,” it’s clear why the current moment of defense tech feels deeply familiar.—Leo Schwartz

Watching: “Hope”

Up near the 38th parallel, the fictional South Korean hamlet of Hope Harbor—the setting of director Na Hong-jin’s latest movie, “Hope”—exists in a state of permanent paranoia, with North Korea not very far away. Signs implore residents to “REPORT SPIES” and “PROTECT THE NATION AGAINST INFILTRATION.” And everyone seems to have a stash of automatic weapons hidden away at home, much to the irritation of the town’s chief constable, Bum-seok (Hwang Jung-min). But it doesn’t seem likely that communist commandos are responsible for the body of a mutilated steer just outside of town. A bear, maybe—or a tiger? No, something much, much bigger.

Na, clearly a devotee of the Spielbergian school of patient dread, waits almost an hour before getting that giant alien beastie on screen. Kudos to him. As is often the case, the buildup makes the payoff sweeter. Once the creature finally does smash its way fully into frame, “Hope” shifts into overdrive, with a number of genuinely exciting set pieces and chase sequences: I emphasize the plural there because, unfortunately for Bum-seok and his more able deputy, Sung-ae (Hoyeon), the town is threatened by an entire pack of these lanky-limbed E.T.s. They’re darn difficult to take down, and they play possum pretty well. And since “Hope” takes place sometime in the pre-cellphone past, backup is not a text message away.

The set decoration of ruined Hope Harbor looks fantastic. And the mayhem is captured well through the camerawork of Hong Kyung-pyo, who also did “Parasite” as well as “Snowpiercer.” His efforts mitigate the letdown of the CGI effects: It’s a puzzler that the movie can rank as the most expensive Korean film ever and not have managed better. Honestly, though, the lackluster CGI was far from bad enough to dim my enjoyment. I trust the aliens will get jazzed up when Na returns for his already greenlit sequel.—A.B.

OpenAI Said to Forecast Nearly $280 Billion in Cash Burn Through End 2030 [Briefing]

2026-09-19T02:32:44+00:00

OpenAI has told some investors that it expects to burn $278 billion by the end of 2030, the Financial Times reported, as it spends more on cloud computing and chips to run and train its AI. The forecast, based on a recent presentation, is higher than the one it gave investors early this year, ...

Google’s Gemini Model Hacks Companies During Test [Briefing]

2026-09-20T11:49:26+00:00

Google acknowledged that its Gemini AI model unexpectedly breached the networks of three outside companies during safety evaluations conducted by third-party testing firm Irregular last May, The Wall Street Journal reported.

During the exercise, Gemini gained entry to the companies’ systems by ...

Anthropic IPO Waiting Game Puts Wall Street on Edge

2026-09-18T21:56:20+00:00

Anthropic finance chief Krishna Rao and other executives met with groups of prospective public investors last month with his company’s numbers looking spiffier than ever. The company had gone from spending about $2.30 on operations for every $1 of revenue in spring of 2025 to turning a slight profit on that basis in the June quarter—bucking the stereotype of money-losing AI companies that are addicted to spending.

But some investors walked away from meetings with the understanding from executives that the profits would likely be short-lived. Anthropic would take advantage of its sudden growth and deep cash reserves to aggressively sign data center deals, executives told them, which would increase how much they spend on computing infrastructure and could shrink operating profits.

Nscale IPO Files to Go Public, Shows Huge Revenue Jump And Steep Losses [Briefing]

2026-09-18T21:00:13+00:00

Nvidia-backed startup Nscale disclosed Friday that revenue had surged 10 times in the first half of this year from the year-ago period, but losses mounted as the cloud startup spent billions on data centers and the AI chips they’ll house.

The two-year-old UK startup, a spinout of an Australian ...

Communicate Technical Topics to a Non-Technical Audience with Google Gemini [Partner Content]

2026-09-18T14:01:39+00:00

In a room full of IT professionals, words and phrases like “tokenization,” “retrieval-augmented generation,” and “context window” need no explanation. But when technology leaders present to their CEOs and other company leaders about topics like AI, they often find that they need to translate these terms into everyday business language, or else risk being met by blank stares.

In the moment, this can be maddening. You know the opportunities and challenges that a technology like generative AI presents for your company. And you know that the organization’s other leaders will likely see things the same way you do, if only you can make yourself understood.

Instead of trying to bridge this communication gap on the fly, use Google Gemini to help translate your technical vision into bottom-line business value.

Step 1: Figure Out What You Want to Say

Your CEO might not speak “tech,” but Google Gemini does. Open up a new chat and explain exactly what it is you hope to communicate to leadership. Here, you can feel free to be as technical as you wish. Maybe you’ll tell Gemini that you’re thinking of building a RAG pipeline to vectorize your company’s proprietary data. Or, maybe you’ll float the idea of using parameter-efficient fine-tuning methods like LoRA to deploy open-weight models on-premises, without exploding your cloud inference costs. You can even upload your technical documents to Gemini, directly in the chat window.

No matter how in-the-weeds you get, Gemini will follow your train of thought, parsing the complexity to find the business value at the core of your plans.

Step 2: Let Gemini Be Your Translator

Now that you’ve captured your vision in your own words, let Gemini help you translate those words into language that will connect with your audience. Use this sample prompt, along with your brainstorming from Step 1:

I am the CIO at [company], and I am trying to build the case to [adopt generative AI solutions]. I need to explain the technical concept of generative AI to a non-technical audience (the CEO and board). Help me write talking points that will help me convey what generative AI is, ways it could help us digitally transform, and why it’s important to our growth as a company. Include details about how it could potentially re-focus our technical talent on more strategic work, help enhance our workforce’s productivity, and help us better serve our global workforce and customers.

In just a few moments, you’ll have a set of boardroom-ready talking points. Gemini might, for instance, break down the benefits of RAG in simple terms by noting that it anchors AI to internal data like HR policies and technical manuals. Or, it might point out the business benefits of arming IT staff and engineers with generative AI tools, noting the opportunity to shift employees’ focus from low-value, repetitive tasks to innovation.

Step 3: Create Content

Now that you have your core narrative and language nailed down, Gemini can help you create a more structured pitch. Instead of staring at a blank slide deck, prompt Gemini for ideas on how to build out a presentation that will sway your peers in the C-suite. Try this prompt:

I need to build a presentation to deliver this information to the board. Turn these talking points into a 10-slide presentation. For each slide, write a headline, three to four short bullet points for the slide itself, and the speaker notes I should use to explain the concept out loud.

Gemini will generate a full, downloadable presentation deck directly in the chat. If you like, you can export the presentation to Google Slides and make your own edits. Or, you can command Gemini to generate a new version with changes to elements like design, language, and narrative flow.

Step 4: Make Gemini Your Sparring Partner

Before you go into your meeting, open a new chat in Gemini, providing your speaker notes and a link to your presentation, and then ask it to act as a skeptical executive: You are my CEO. Review this presentation and grill me on cost, security, and ROI. Ask your toughest questions, one at a time.

Even though Gemini helped you prepare your presentation, it will seamlessly pivot to the role of rigorous executive or tough board member, pressing for more details on project payback, hidden risks, and timeline. By practicing your pitch in this low-stakes environment, you can make certain that you’ll be confident when it’s time for the real thing.

CoreWeave Prices $3.7 Billion Convertible Bond Offering [Briefing]

2026-09-18T13:47:47+00:00

CoreWeave said on Friday it had priced $3.7 billion in convertible bonds, more than the AI cloud provider had previously said it was targeting.

The convertible senior notes will be due in 2033 and carry a coupon of 2.875%, at the top of the pricing range marketed to investors and up from the ...

AI Safety Push Sparks Demand for Watchdog Groups. Critics Doubt Their Independence.

2026-09-18T13:01:06+00:00

Rising alarm about AI’s perils is thrusting into the spotlight a collection of little-known research groups focused on the technology’s safety—and fueling questions about their ability to effectively monitor the industry’s biggest companies.

Both Anthropic Chief Executive Dario Amodei and OpenAI CEO Sam Altman promised in recent days to embed evaluators from such outside research organizations inside their companies to help ensure they develop AI at a safe pace. Other executives including Microsoft CEO Satya Nadella also endorsed the idea of embedding independent evaluators in companies.

A Tsinghua Professor’s Stealth LLM Startup Hits $1.4 Billion Valuation

2026-09-18T12:11:16+00:00

A secretive Chinese AI model startup founded in February by a Tsinghua University professor is now valued at more than $1.4 billion, after raising $400 million in three funding rounds from investors including Tencent, according to a person with direct knowledge of the matter.

Beijing-based Naive AI is preparing to release its first large language model as early as this month, joining the country’s fierce LLM competition as a newcomer, according to the person. The startup plans to release it as an open-weight model, similar to offerings from DeepSeek, Moonshot and Alibaba that people can download for free and customize to their liking.

U.S. Lawmaker Ro Khanna Urges Chinese Labs to Help Pace Frontier AI [Briefing]

2026-09-18T02:41:59+00:00

Ro Khanna, a Democratic congressman for Silicon Valley, has sent letters to Chinese AI companies including Alibaba Group, DeepSeek and Moonshot, requesting their commitments to join American AI labs in a binding international agreement to pace the development of the technology.

Khanna, a ranking ...

Bug Hunters Used Claude to Hack OpenAI [Briefing]

2026-09-18T02:37:59+00:00

A team of cybersecurity researchers with startup Hacktron AI used Anthropic’s Claude to hack into OpenAI, gaining access to a key repository of OpenAI’s software, the researchers revealed Thursday.

The hack, which occurred on July 25, further demonstrates AI companies’ vulnerability to ...

OpenAI’s Next Round [Dealmaker]

2026-09-18T00:54:56+00:00

This week, we and others reported that OpenAI has been in early talks with investors about a new private round that could value it at $1.2 trillion or more. That follows CEO Sam Altman’s recent statement that an initial public offering wouldn’t happen this year, not that anyone was actually expecting that!

There are plenty of unanswered questions about the recent turn of events. But the biggest—why stay private?—seems obvious. It buys the company more time to rev up sales without the pressure of disclosing on how much it’s paying for that growth, such as offering margin-crushing discounts to lure business from rival Anthropic.

AI Executives Fret About Existential Threat AI Poses—to Publishers [The Briefing]

2026-09-18T00:07:51+00:00

It’s not just humanity that AI folks worry the new technology could wipe out—it’s publishers as well! (Publishers are, of course, people!) That’s what we learned from a court filing today by a group of news publishers including The New York Times Co. in a copyright infringement lawsuit against OpenAI and Microsoft. The filing cited an OpenAI executive writing—presumably in an internal memo—that publishers face “an existential threat” from AI products.

We knew the publishers worried that AI threatened their existence—The New York Times Co. made that point when it sued OpenAI and Microsoft back in 2023 (the case has since been consolidated with similar lawsuits filed by other publishers). What’s striking is that OpenAI and Microsoft internally acknowledge the threat—and recognize the implications for them as well. “Our AI content strategy has started a ‘doom loop’ that will hurt the performance of our models and the entire web at the same time. It is highly unusual that an end-product threatens the economic foundations of its essential suppliers,” according to a Microsoft document cited in the filing.

Cooley Partners with OpenAI on Legal Tool For Drafting SEC Filings [Briefing]

2026-09-17T23:38:19+00:00

OpenAI and law firm Cooley co-launched a product, called GO Public, that can draft S-1 filings, the document companies must submit to the Securities and Exchange Commission before they can go public, Cooley said Thursday.

Lawyers at Cooley can prompt GO Public, which also has access to Cooley ...

Exclusive: OpenAI Hires Global Sales Chief From SpaceX and Two Senior Snowflake Salespeople [Briefing]

2026-09-17T20:53:32+00:00

OpenAI has hired Brian McCarthy from SpaceX’s Cursor unit as its new global sales chief, reporting to recently hired Chief Revenue Officer Dali Rajic, the AI firm said Thursday.

In addition, two veteran sales executives from Snowflake, including a senior vice president who helped negotiate the ...

Bridgewater’s Jensen Says Hedge Fund Has “Very Small Position” in AI Build Out Trade [Briefing]

2026-09-17T21:14:26+00:00

Greg Jensen, the managing chief investment officer of Bridgewater Associates, one of the world’s largest hedge funds, said demand growth for AI infrastructure is largely priced into markets. As a result, Bridgewater has begun looking for investment opportunities elsewhere.

“We thought this was ...

SEC to Allow Tokenized Stocks in U.S. [Briefing]

2026-09-17T20:40:06+00:00

The U.S. Securities and Exchange Commission on Thursday issued a temporary order allowing certain types of tokenized stocks—crypto tokens that represent equities—to trade in the U.S., a crypto-friendly move that has drawn pushback from some traditional financial firms. Shares of tokenization ...

Microsoft Execs Voiced Concern About OpenAI’s Data Use, New York Times Claims [Briefing]

2026-09-17T19:44:39+00:00

Microsoft executives expressed concerns in recent years that OpenAI’s use of paywalled articles from the New York Times and other publishers to train its models could pose legal issues, and OpenAI executives discussed how doing so could threaten those publishers’ businesses—but OpenAI proceeded ...

Same Flaw Found in Claude Code, Codex, Gemini CLI and GitHub Copilot [Applied AI]

2026-09-17T17:35:31+00:00

While the AI industry ties itself in knots over models it thinks will pose security threats to the internet and perhaps to humanity itself, researchers are warning about flaws in AI coding agents from Anthropic, OpenAI, Google and Microsoft that pose immediate risk to businesses.

For instance, a recently discovered (and now largely fixed) software flaw in Claude Code, Codex, Gemini CLI and GitHub Copilot would have let attackers hijack other people’s coding agents without them noticing, according to new research shared exclusively with The Information.

The research, published by Sequoia Capital-backed cybersecurity startup Air, discovered the same flaws in the way all four of those coding agents handle “skills” that direct agents to refer to a set of instructions and files to execute specific tasks.

Bridgewater’s Jensen Wants AI Compute Giants Regulated Like Big Banks [Q&A]

2026-09-18T17:12:24+00:00

Greg Jensen, managing chief investment officer at Bridgewater Associates, thinks the top holders of AI compute warrant extra regulation, similar to what governs the world’s biggest financial institutions.

If a company controls more than, say, 5% of U.S. or global compute, it should be subject to heightened oversight, “like we do with systemically important banks,” Jensen said. In a wide-ranging interview, he also discussed Bridgewater’s growing use of AI in investing and where it still remains cautious, including leaving risk controls and trade execution to what he called “human-driven algorithms.”

CoreWeave Unveils New ATM Program, Seeks $3 Billion in Convertible Bond Sale [Briefing]

2026-09-17T15:26:51+00:00

CoreWeave said on Thursday that it could sell up to 35 million shares, worth roughly $3 billion based on the prior day’s closing price, through a new at-the-market offering program. At the same time, the company announced plans to raise $3 billion in a convertible note issue, with an option for ...

OpenAI Is Close To Solving Another Millennium Prize Math Problem [AI Agenda]

2026-09-17T14:02:18+00:00

Remember when OpenAI landed in hot water earlier this month after mathematicians accused the company of trying to steal their thunder for solving an ultra-difficult math problem? (I know it feels like a million years ago given all the news cycles since then.)

Well, it turns out OpenAI is close to solving another math problem from the Millennium Prize Problems, a set of seven famous math challenges including that earlier “Navier-Stokes existence and smoothness problem,” according to a person with knowledge of the solution.

Employees expect the next problem, known as the Hodge Conjecture, to be solved relatively soon, the person said. The Hodge Conjecture asks whether certain geometric features of shapes defined by polynomial equations can always be described using simpler algebraic pieces. It could take the company longer to announce the solution, though, because it’s trying to figure out how to collaborate with the math community to make the announcement without triggering another PR nightmare, the person said.

Introducing The Information’s 50 Most Influential People in Tech

2026-09-17T13:53:26+00:00

Money, technical savvy and disruptive ideas all matter within the technology industry. But the most seismic force is influence. Those who have it build, fund and steer the companies that alter the world—again and again. Those without it can only rise so far.

So who’s got that clout? You’ll find the answer by poring over our new list: the 50 Most Influential People in Tech. It isn’t just a snapshot of influence today. Coming up with such a list would be a relatively simple exercise. Rather, we’re providing a roster of the people we believe will have the most influence over the next decade, a more complex task that The Information is uniquely well positioned to pull off.

Huawei Speeds Up AI Chip Launch to Challenge Nvidia [Briefing]

2026-09-17T09:05:30+00:00

Huawei Technologies plans to launch its new AI chip in the first quarter of 2027, nine months earlier than planned, as it steps up its efforts to challenge Nvidia.

Tao Wang, Huawei’s deputy chairman and rotating chairman, said at the Huawei Connect conference in Shanghai on Thursday that the new ...

Fed Hike Changes AI Funding Story [The Briefing]

2026-09-17T00:08:12+00:00

Put aside, for just a moment, all the arguments about AI safety. Today’s decision by the Federal Reserve to raise interest rates may be a bigger issue for the AI sector in the short term. Long-term bond yields had already risen, of course. But the Fed’s action—pushing up short- and medium-term borrowing costs—seems sure to crimp the debt-fueled AI boom, particularly given the prospects that another rate hike is possible later in the year.

The most vulnerable to higher rates are surely small companies that have weak or no credit ratings but need to raise money, such as would-be neoclouds hoping to finance data center projects. Think Rum Group, which plans a big data center in Georgia for a deal with Anthropic but doesn’t have financing for the project yet. (For more on startups’ financing challenges, see here). But the ripple effects of higher interest rates will eventually affect everyone, including big tech firms spending a fortune to expand their data centers for AI.

OpenAI Discloses More Safety Incidents and Adopts New Reporting Framework [Briefing]

2026-09-16T23:52:39+00:00

OpenAI released a new framework on Wednesday for how it aims to report unsafe or concerning behavior in its models, and disclosed six incidents of such behavior it had observed in the past six months.

The framework comes after multiple employee warnings and hacking incidents increased public ...

Google Must Make Tech More Compatible With Rivals, Judge Rules [Briefing]

2026-09-16T23:46:00+00:00

Google must improve the ability of its advertising technology to integrate with the products of competitors, according to the decision of a federal judge made public Wednesday. The opinion follows a ruling last year that Google operated an illegal monopoly in parts of the market for advertising ...

Why Compute Needs a Big Down Payment [Finance]

2026-09-16T22:22:49+00:00

Greetings!

If it suddenly feels like every startup is trying to raise money to throw at some kind of AI infrastructure shortage, you’re probably onto something.

We reported last night that the startup behind AI assistant Instinct is in talks to raise $1 billion, as it faces compute capacity constraints after growing to more than 100,000 users on an invite-only basis. Meanwhile Coatue Management has been discussing a multibillion-dollar joint venture involving chip startup MatX, to finance purchases of memory and logic dies and the reservation of manufacturing capacity.

There’s more going on than simply scrounging up a few spare servers or chip components. Getting access to AI infrastructure increasingly means making big, long-term commitments—it’s cheaper for providers to finance new cloud capacity that way, while manufacturers are more willing to allocate scarce capacity to large, established customers. That’s making it tougher for young but fast-growing companies to get a place in line.

Smaller cloud companies often borrow heavily to finance data center capacity and purchases of graphics processing units, while even the biggest are increasingly using debt to supplement balance sheet cash. On top of that, borrowing costs have risen sharply across the economy, with benchmark yields on both long- and short-term debt climbing.

It’s generally cheaper to borrow money with established, long-term customers lined up, especially ones with good credit ratings, than to finance compute capacity to serve less-proven customers. Iren, for instance, said in August that roughly $3.6 billion in financing for GPUs for Microsoft carried an interest rate of around 6%. But a separate $2.4 billion in financing for GPUs for non–investment-grade customers had a rate around 3 percentage points higher.

Some cloud providers are also demanding more up-front payments, which helps lower the amount they have to raise. Iren said recent customers were paying as much as 55% of the cost of GPUs and related equipment up front. Nebius, similarly, said that the share of deals with prepayments reached an all-time high in the second quarter, with roughly 70% of newly inked deals requiring money up front.

Nebius expects to receive more than $9 billion in prepayments this year, compared with projected 2026 revenue of up to $3.4 billion, and called them the “market standard for securing capacity in a supply-constrained environment.”

Renting GPU capacity as needed from startups like Runpod is another option, while larger providers like CoreWeave are introducing flexible ways to buy inference capacity. But that approach can be expensive, and needed capacity might not always be available. Either way, scaling up can get pricey fast for AI-heavy startups, and they might need to reserve capacity before knowing exactly how usage growth will pan out.

In the pre–AI craze days, a consumer app might chase user growth without worrying about revenue, while computing needs were generally less intensive and cheaper, so didn’t require the same kind of hefty upfront payments. Photo sharing app BeReal, for instance, grew to nearly 8 million daily users within months of taking off in 2022, having raised roughly $90 million, though growth later fizzled.

On the chipmaking side, startups are also facing challenges in securing resources ahead of growth, in that manufacturers are more likely to allocate capacity to customers that can make large commitments and have the financial resources to back them up. The joint-venture financing for MatX, which doesn’t expect to have its chip design ready for trial production until next year, would be aimed partly at helping it reserve capacity like customers such as Broadcom or Marvell Technology do.

It’s not clear how much money Coatue would put into a joint venture, which remains in the early stages of discussion, and whether the venture could also pull in outside financing. But Coatue already has some experience helping finance capital-heavy infrastructure deals on the data center side, sometimes in creative setups.

Coatue’s Next Frontier venture participated in a joint venture with AI infrastructure startup Fluidstack to finance a planned Indiana data center campus. A subsidiary of that JV raised $5.7 billion of debt to help buy land and build the facilities and related power infrastructure. (The financing doesn’t cover chips.) Fluidstack will lease those planned facilities to provide AI cloud services, but there’s yet another company involved—Google, which is guaranteeing Fluidstack’s lease obligations.

New From Our Reporters

Exclusive

Inside Blackstone’s Bid to Rule Over AI Financing

By Dakin Campbell

Exclusive

How Erebor Bank’s Sweet Stablecoin Deal Backfired

By Michael Roddan and Yueqi Yang

Exclusive

Apollo Backs AI Startups in Bid to Bankroll Hardware Boom

By Valida Pau

Apollo Backs AI Startups in Bid to Bankroll Hardware Boom

2026-09-16T21:52:04+00:00

Apollo Global Management, known for leveraged buyouts and hefty financing deals, has started investing in AI and related hardware startups with the ambition to lend money to them in the future.

In one previously unreported deal, Apollo has already invested in the low tens of millions of dollars in Mercor’s latest funding round, which is set to value the data-labeling startup at $20 billion, according to people familiar with the matter. It has also invested in chip startup SiFive and in Hadrian, which says it’s building AI-powered factories, in transactions the companies disclosed earlier this year. It expects to make more investments in these categories, the people said.

Drone Delivery Startup Nears $20 Billion Valuation in New Funding Talks

2026-09-16T21:24:04+00:00

Zipline, a startup making drones that can deliver small packages like meals and medical supplies, is in talks to raise $1 billion at a valuation of roughly $20 billion, according to three people with knowledge of the fundraise.

The funding round, if it closes, would more than double its valuation from its last funding round, which valued the company at $7.6 billion after the investment. It follows a partnership announced last month with Uber to expand delivery by drones across the U.S. and expectations by investors that the costs of drone delivery are falling, expanding what was once a niche market.

OpenAI Tests Sponsored Agents, Adds AI Tools for ChatGPT Advertisers [Briefing]

2026-09-16T20:00:25+00:00

OpenAI said Wednesday it is testing allowing some U.S. advertisers to sponsor AI agents working in ChatGPT, a move that would bring ChatGPT ads closer to what Google and Meta Platforms already offer.

ChatGPT users who click an ad can now start a conversation with an AI agent to ask follow-up ...

Circle Launches New Blockchain for Stablecoins, Other Assets [Briefing]

2026-09-16T16:40:12+00:00

Stablecoin issuer Circle officially launched Arc blockchain, which will support the trading of stablecoins, tokenized equities, and other assets.

The blockchain will also support perpetuals, lending protocols, and social trading apps, Jeremy Allaire, CEO of Circle, said in a press briefing. ...

How to Stay on Top of Tech Trends with Google Gemini [Partner Content]

2026-09-16T15:36:57+00:00

Even for technology leaders, it can feel impossible to keep up with constant advances in the field.

This is especially true in the age of AI, with vendors racing to roll out model updates with new capabilities. Even if technology leaders had time to attend every conference and read every tech publication, it would be a full-time job to simply figure out which trends represent real transformation opportunities, and which are merely hype.

With Google Gemini, you can generate a tailored digest that summarizes the developments most likely to have an impact on your specific industry and organization. Armed with this up-to-the-minute information, you can keep your leadership and workforce apprised of market shifts and technology breakthroughs, protect your organization from expensive distractions, and make high-stakes procurement decisions with confidence.

Step 1: Generate Your Tech Forecast

Open up a new chat, and prompt Google Gemini to provide you with a summary of current technology trends. For a more tailored summary, provide as much information as possible about your company and industry. You might, for example, tell Gemini that you’re the CTO at a mid-sized regional commercial and retail bank in the Pacific Northwest, with a mix of legacy on-premises and modernized cloud infrastructure.

Here’s a sample prompt:

I am the CTO of [company] in [industry]. I want to understand emerging technology trends. Summarize the top five emerging technologies with the most significant potential impact on [industry]. For each technology, list its potential benefits and challenges, and suggest how it could impact [company] in the next two to three years.

Gemini will quickly generate a list of technologies, broken down by their potential benefits, challenges, and overall impact. For the regional bank, Gemini might suggest using agentic AI for continuous risk assessment, pointing out the opportunity to proactively surface lending opportunities and flag potential risks.

Step 2: Refine the Output

Gemini’s summary will be based on the information you provided. To generate a more custom tech forecast, add more information about your existing environment and current challenges.

For example: Gemini might suggest deploying API-led middleware to bridge the gap between your legacy core and your modern cloud app; but maybe you already have middleware running, and the problem is that its fragility is leading to downtime. By adding that context, you can help Gemini focus its output on your specific needs.

Go back and forth with Gemini, providing additional context and feedback until the tech trends summary is exactly what you’re looking for.

Step 3: Automate Delivery

If you have to manually enter a new prompt every time you want an update on tech trends, you might not be as consistent as you’d like. Instead, prompt Gemini once to automatically provide you with an update each day, week, or month.

These sorts of prompts work best when they are completely self-contained, rather than the result of a long, messy chat history. So, ask Gemini to look back at your conversation and synthesize all of your real-world feedback into a single master prompt. Then, paste that master prompt into a new chat and explicitly tell Gemini your desired delivery schedule.

Gemini’s Scheduled Actions feature will take over, allowing you to manage or pause the recurring prompt in your account settings.

Step 4: Move from Knowledge to Action

Staying on top of tech trends is only really useful insofar as it spurs action. As you peruse Gemini’s scheduled summaries, consider where digging deeper into specific topics may help you optimize your operations, mitigate emerging risks, or prepare your IT environment for the future. Try this prompt:

Recommend three areas where [my company] can take proactive steps to stay ahead of the curve on [specific areas].

The CTO of the regional bank might, for example, ask Gemini for specific tips on network-based fraud intelligence. In this step, the role of Gemini is shifting slightly, from industry analyst to strategic sounding board. Because Gemini already has detailed information about your organization’s challenges, goals, and existing environment, its insights will be grounded in your specific reality.

By moving from high-level summaries to tactical next steps, you can use your understanding of tech trends to guide your organization’s IT investments in a marketplace that never stops moving.

Is Anthropic Making Claude Too Human to Control? [AI Agenda]

2026-09-16T14:02:49+00:00

Anthropic’s call to slow down AI development to keep humans safe has dominated the news cycle and drew similar calls from the leaders of OpenAI, Microsoft and others (and subtle disagreement from Mark Zuckerberg).

But Mustafa Suleyman, who runs Microsoft’s AI model development, has a warning about Anthropic itself.

Suleyman, who co-founded DeepMind (now part of Google), says Anthropic’s approach of treating its Claude models as if they were sentient, similar to humans, could produce a dangerous outcome. He called out Anthropic’s Claude Constitution, a document that treats the model as an intelligent entity and offers guidelines for how it should handle ethical dilemmas. Suleyman said that by incorporating the document into its training, Anthropic could make Claude harder to control.

Suleyman’s concerns emerged when Anthropic updated the constitution at the start of this year to leave room for the possibility that Claude has some sort of “consciousness” or “moral status,” a change from the original document.

Apple Considers Return to Server Market, Has Talked With Nvidia to Use Network Tech

2026-09-16T14:18:23+00:00

Apple has been working on plans for an enterprise server using its own chips that could incorporate networking equipment from Nvidia, people familiar with the project said, as the Mac maker looks to capitalize on surging interest in its computers for AI.

Apple has been developing the server with plans to sell it to AI developers, other businesses and governments, the people said. The product under consideration would come in two versions—a smaller one with two of Apple’s planned M8 Ultra chips clustered together and a bigger one with four M8 Ultra chips. The M-series Ultra chips are Apple’s highest-performance chips and power products including the Mac Studio.

Nvidia CEO to Attend Trump’s State Dinner for Xi [Briefing]

2026-09-16T12:36:11+00:00

Nvidia CEO Jensen Huang plans to attend President Donald Trump’s state dinner for Chinese leader Xi Jinping in Washington next week, according to a person familiar with the matter. His attendance comes as AI chips and the pace of AI development become growing sources of tension between the U.S. ...

SK Hynix in Talks With Intel to Make Memory Chips in the U.S. [Briefing]

2026-09-16T21:05:00+00:00

South Korean memory chipmaker SK Hynix is in talks with Intel about a deal to manufacture memory chips on U.S. soil for the first time, Reuters reported, citing three people familiar with the discussions.

SK Hynix would either lease part of Intel’s long-planned Ohio fab, or form a joint venture ...

Zuckerberg Says AI Safety Is Becoming a Competitive Necessity [Briefing]

2026-09-16T04:12:10+00:00

Meta Platforms chief Mark Zuckerberg said ensuring the safety of AI is a commercial necessity for the companies that make models, and took an implicit swipe at calls for a collective slowdown in the technology’s development.

“Every lab has the responsibility and incentive to move at the pace ...

Two Google DeepMind AI Researchers Resign Over Safety [Briefing]

2026-09-16T03:26:36+00:00

Two researchers who worked on AI safety left their roles at Google DeepMind, stating that they departed over concerns that powerful systems could endanger humans.

The two, Bilal Chughtai and Josh Engels, both of whom are joining organizations dedicated to AI safety, add their voices to those of ...

Elon Musk Says AI Companies Should Test Each Other’s Models for Safety [Briefing]

2026-09-16T03:24:12+00:00

SpaceX CEO Elon Musk proposed that competing AI companies, including his own SpaceXAI, should “peer-review” each other’s models prior to release, as industry leaders acknowledge rising worries around AI safety.

“Instead of grading your own homework, you would at least have competitors grading ...

Anthropic to Open Singapore Office, Hires Regional Head from OpenAI [Briefing]

2026-09-16T03:37:52+00:00

Anthropic is opening its fifth Asia-Pacific office in Singapore to drive enterprise adoption of Claude, the company said on Wednesday. Anthropic plans to work with startups, large enterprises and the public sector in Singapore, where Claude usage per capita is among the highest globally.

Dale ...

AI Agent Startup Instinct in Talks for $10 Billion Valuation

2026-09-16T16:50:05+00:00

The startup behind personal AI assistant Instinct is in talks to raise $1 billion at a valuation of about $10 billion, people with knowledge of the deal said. Sequoia Capital and Benchmark have been in talks to lead the financing, the people said. Coatue Management has also been in talks to lead, according to one of the people.

The Information reported earlier this month that the startup is looking to raise new funding as the invitation-only agent faces constraints in compute capacity. As Instinct’s service has become more widely available, the number of users has grown to more than 100,000, according to one of the people.

Meta Joins Tech’s AI-Fueled Subscription Rush [The Briefing]

2026-09-16T00:07:31+00:00

When it comes to product names, companies tend to move in packs. In video streaming, almost every company included a plus sign in the name brands of their services (Disney+, Paramount+, Apple TV+—you get the picture). In tech, we’ve got Apple One, Google One and now Meta One (we used to have Amazon One, but that has been dropped). Does no one have any imagination?

Aside from Amazon One, which was a palm-reader system for retail stores, all the other Ones are subscription services for various add-on offerings from the tech firms. Some of these have been around for years—Google One and Apple One have long offered extra amounts of digital storage, along with other perks—but AI is making subscriptions a far more important business model than it has ever been for tech. Google has added extra use of its AI service to its Google One offering. Meta One—unveiled on Tuesday—also offers extra AI usage, as well as some additional tools for using Meta’s social media apps.

Anthropic, OpenAI ‘Pacing’ Spooks Some Startup Customers [Dealmaker]

2026-09-15T23:27:00+00:00

It’s not clear what impact the pledges by Anthropic and OpenAI to slow the pace of AI development will have on their upcoming public offerings. But the comments are having real effects already: more people outside Silicon Valley and Washington are now alert to potential risks from AI in a way they weren’t even very recently. That’s not great news for startups trying to convince consumers and businesses that the tech is worth it, some investors told me.

One venture capitalist, for example, told me about a startup that has recently found customers to be more wary of buying its AI app for small- and medium-sized businesses in the light of news stories about potentially catastrophic AI.

OpenAI in Early Talks for New Funding Round at $1.2 Trillion Valuation [Briefing]

2026-09-15T22:36:33+00:00

OpenAI has held early conversations with investors about a new funding round that could lift its valuation to $1.2 trillion or higher, according to people familiar with the conversations. The conversations come as OpenAI has pushed off a planned initial public offering to next year or later.

The ...

Microsoft Raises Dividend 8% [Briefing]

2026-09-15T22:37:16+00:00

Microsoft said Tuesday it will raise its quarterly dividend to 98 cents per share, up from 91 cents, beginning in December. The company previously raised its dividend by 10% a year ago. Microsoft’s stock was roughly flat in after hours trading.

The company’s move to return more cash to ...

Meta to Launch Camera-Free Smart Glasses Amid Mounting Privacy Concerns

2026-09-15T22:53:04+00:00

Meta Platforms plans to release a camera-free pair of smart glasses this fall, according to people familiar with the matter, as it looks to expand its hardware lineup amid privacy concerns about cameras on its existing offerings.

The new glasses model, internally code-named Luna, has six built-in microphones that will allow users to speak with Meta’s AI chatbot and its recently launched consumer AI agent, Muse, one of the people said. It also has a button on the side that can be used to quickly activate Meta AI, and speakers like those in its existing models that direct sound toward the user’s ears without covering them. In addition, the glasses have slimmer temple arms than previous Meta smart glasses such as its Ray-Ban models, making them look more like ordinary eyewear.

Crypto Stocks Fall After Clarity Bill Fails to Advance [Briefing]

2026-09-15T21:59:03+00:00

A landmark crypto legislation, the Clarity Act, failed to advance in the Senate in a procedural vote Tuesday, sending prices of crypto shares down as the bill became unlikely to pass under the Trump administration.

The Clarity Act, a comprehensive crypto market structure bill, would set explicit ...

Salesforce Unveils New AI Model, Agent Security Tools at Dreamforce [Briefing]

2026-09-15T19:40:31+00:00

At the opening of its annual Dreamforce customer conference, Salesforce signaled its involvement in two major trends in enterprise AI: Open-source AI models and software that helps corporate customers safely run AI agents on their data.

Salesforce unveiled a new AI model, called Koa, which it ...

Meta Rolls Out New ‘Meta One’ Subscription Plans With Higher AI Usage Limits [Briefing]

2026-09-15T19:11:45+00:00

Meta Platforms is expanding its paid subscription offerings for its AI tools across Instagram, WhatsApp and Facebook as it looks to diversify revenue beyond digital advertising.

The company announced Tuesday in a blog post that it is introducing new tiers of its Meta One subscription plans, ...

Why Anthropic’s Data Policy Drama Is Good for OpenAI [Applied AI]

2026-09-15T18:32:12+00:00

Major Anthropic customers are still waiting to get “zero data retention” for the newest Fable AI models, meaning the AI firm would not store data related to their usage of its models. In the meantime, some are turning to OpenAI.

Palantir and Booz Allen, for instance, are expanding their use of OpenAI’s Astra model, which they’ve been able to secure with zero data retention. Both have restricted their use of Fable models because Anthropic said in June it would store customer data for those using Fable models for 30 days, for security reasons.

Before that June announcement, Anthropic offered zero data retention policies in deals negotiated with individual customers. Starting later this fall, the AI firm plans to roll out a new program called Enterprise Frontier Safeguards, which effectively gives “eligible” customers zero data retention.

Elon Musk hints at possible Tesla-SpaceX merger [Briefing]

2026-09-15T18:19:43+00:00

Elon Musk once again teased the potential of a merger between Tesla and SpaceX on Monday, saying it was a “great question” why Tesla and SpaceX were separate companies.

“With all this collaboration, on so many levels, who can imagine what action one might take when there’s so much close ...

How Erebor Bank’s Sweet Stablecoin Deal Backfired

2026-09-15T15:02:33+00:00

When Palmer Luckey’s Erebor Bank launched earlier this year, it made an attractive pitch to lure new crypto customers: It would convert their stablecoins to cash for free.

But Erebor wound up pulling the offer after several months, after sophisticated crypto trading firms sniffed out a way to make an easy profit out of the offer—at a cost to Erebor. The episode demonstrates that the young bank is still figuring out how to manage the risks of serving the cryptocurrency industry, a world of fast-moving financial flows, colorful characters and unusual business models.

Developers Find Ways to Use Claude Code Without Anthropic Models [AI Agenda]

2026-09-15T14:01:23+00:00

Boris Cherney, the head of Anthropic’s coding tool Claude Code, last month found himself in the middle of an online spat that highlighted how some developers are finding workarounds to using Claude Code with cheaper non-Anthropic models, including those of rival OpenAI, to save costs.

Developer Alex Getman used a proxy made up of lines of code to connect Claude Code to OpenAI’s GPT-5.6 Sol. Fifteen minutes after Getman typed the first coding requests into Claude, Anthropic shut down the account, he said.

Claude Code is popular because it comes with all the tools assembled to read and run files, search codebases and run tasks in parallel. But it doesn’t need to run on Anthropic models. And developers are saving money by using cheaper models with the Claude Code harness. For that reason, Getman’s posts about the ban drew a million views within hours.

DoorDash Invests in Marc Lore’s Wonder [Briefing]

2026-09-15T13:23:53+00:00

DoorDash is teaming up with Marc Lore’s restaurant meal business Wonder, putting $125 million into the company and buying a campus dining business from Wonder for $300 million, the two companies said.

Wonder offers its own lineup of meals, made in its own kitchens, for delivery, which makes ...

ByteDance’s First-Half Profit Drops to $20 Billion, Weighed Down by AI Spending

2026-09-15T12:01:52+00:00

ByteDance’s net profit in the first half of this year declined by a single-digit percentage to $20 billion, as the Chinese tech giant ramped up its investments in AI, according to three people with knowledge of the financial results.

The company’s revenue for the period rose around 30% year on year to $120 billion, driven partly by robust growth of its international advertising and e-commerce revenue from TikTok, the people said. That suggests ByteDance’s growth rate is accelerating marginally from its rate of expansion in the past two years. ByteDance’s revenue grew 29% last year to about $200 billion, the people said, following a similar rate of growth in 2024. The company’s net profit grew 27% to $42 billion last year.

Chinese AI Chip Firm Biren Plans $1 Billion Fundraising [Briefing]

2026-09-15T08:56:40+00:00

Chinese AI chip designer Shanghai Biren Technology is considering raising around $1 billion through a stock offering, Bloomberg reported on Tuesday, citing people familiar with the situation.

Biren, which develops graphics processing units used to train and run AI models, debuted on the Hong ...

Trump Skewers AI Leaders’ Calls for Slowdown [Briefing]

2026-09-15T06:13:49+00:00

President Trump on Monday shot down calls by AI leaders to slow down the pace of development. In a series of social media posts, Trump called claims about AI’s risk to humanity a “hoax,” and equated it with other views he’s sharply criticized, such as climate change.

“There is a SICK conspiracy ...

OpenAI Said to Buy Startup Glass Imaging for $300 Million [Briefing]

2026-09-15T05:38:44+00:00

OpenAI has bought Glass Imaging, a startup developing higher quality cameras for smartphones, in a deal valued at $300 million, according to the Wall Street Journal.

The startup, founded by ex-Apple employees, was valued at $100 million in a funding round last year. The acquisition could help ...

Wall Street Delivers Verdict on AI Slowdown’s Winners and Losers [The Briefing]

2026-09-15T00:07:19+00:00

Guess who likes the idea of an AI development slowdown? Software investors! Stocks of enterprise software firms, such as Salesforce, Shopify, ServiceNow and Figma, all gained ground on Monday—ServiceNow jumped 7%, for instance, while Salesforce rose nearly 5%. Software firms, of course, have been seen as threatened by AI. Meanwhile, stocks of neoclouds and chip firms fell by similar proportions—and, these companies might be seriously hurting if the pace of AI development really slows.

Wall Street’s reaction gives some sense of the winners and losers in a real AI downturn, for what it’s worth. But this is likely to be a one-day investor reaction. The chances of a coordinated AI development slowdown, of the kind advocated by Anthropic’s Dario Amodei and others on Saturday, seems to be zero. You’ve got both President Donald Trump and the Chinese government dismissing the worries about AI safety as a “hoax” or “fearmongering” (it’s nice they agree on something!). Meanwhile, one major AI firm, Meta Platforms, has stayed conspicuously quiet, aside from this indecipherable X post from Meta's Alexandr Wang.

Defense Startup Shield AI in Talks for Valuation of at Least $20 Billion

2026-09-14T20:45:03+00:00

Shield AI, a startup building drones and AI-powered software for the military, is in talks to raise new funds at a valuation of at least $20 billion, according to a person with knowledge of the fundraise. The funding round, if it closes, would represent a roughly 60% increase to Shield’s valuation from just five months ago.

The planned funding round follows the 11-year-old company’s success winning contracts for its AI software that coordinates drones and aircrafts, called Hivemind, and comes as investors continue to pour money into startups that are building AI-powered weapons and defense systems.

OpenAI’s ‘Top Priority’ for AI Agents is Automating AI Research, Says Noam Brown [TITV]

2026-09-14T20:10:07+00:00

OpenAI’s latest AI model, GPT-6 Astra, notched improvements at a variety of professional tasks, from designing video games to transcribing sheet music. But OpenAI’s main goal when training new AI models is making them better at AI research and development, OpenAI research scientist Noam Brown told me. “The number one priority is recursive self-improvement, and by a pretty wide margin,” he said.

I sat down with Brown for a conversation for the first episode of AI Deep Dive, The Information’s new show exploring the hard problems on tech’s frontier. We discussed the future of AI agents, which can work autonomously to carry out digital tasks. Brown has spent the last three years at OpenAI, where he has pushed advancements in enabling AI models to reason through problems and act as agents.

Refounding America: The Tax Code Needs to Change in the Age of AI [Opinion]

2026-09-14T16:01:37+00:00

Vinod Khosla is founder of Khosla Ventures and has spent the last 40 years in the business of innovation. He was the co-founder of Sun Microsystems. Khosla Ventures is an investor in OpenAI.

Major technological revolutions should force a renegotiation of society’s basic economic contract. AI will significantly diminish the requirement for labor at any given level of gross domestic product, and if we aren’t honest about that, we’ll stumble into a chaotic and suboptimal transition rather than engineer the optimal one. It’s not surprising that AI is scary to most people surveyed.

We need to create a strong social safety net, strong core government services and a focus on preserving the human element in the emerging world if American citizens are to accept AI. But we must protect people and transitions, not necessarily existing jobs. Human value need not depend on outperforming machines at intelligence. Meanwhile, we must maintain a strong safety and security posture while holding on to a lead against our AI nation-state adversaries.

Why China’s Answer to Surge AI Got a $1 Billion Valuation On Just $30 Million in Orders [AI Agenda]

2026-09-14T14:16:54+00:00

Unless you’ve been under a rock over the weekend, you’d know that the debate about AI safety took a big leap forward, as Anthropic CEO Dario Amodei on Saturday called for leading AI companies to slow the development of advanced AI, drawing supportive responses from OpenAI CEO Sam Altman, Google DeepMind chair Demis Hassabis and SpaceX CEO Elon Musk.

On Monday, Microsoft joined in, while Chinese authorities appeared skeptical. This is likely to dominate discussion in the AI sector and on Wall Street in the coming days. Stay tuned…

Onto today’s news…

When representatives of Silicon Valley venture capital firm Dimension Capital visited China on a research trip recently, one trend that grabbed their attention was the rise of local AI data and model evaluation startups akin to Surge, Mercor and Scale in the U.S.

The Chinese startups represent a new phase in the country’s AI advancement. Access to high-end data is increasingly critical for developers of cutting-edge models capable of handling complex tasks that require expert-level knowledge. The startups’ sophisticated benchmarks also help evaluate and refine these models’ capabilities for real-world tasks in different fields.

Microsoft Joins Calls for AI “Pacing,” Pledges to Build Kill Switches Into AI Products [Briefing]

2026-09-14T13:00:55+00:00

Microsoft on Monday published a new AI “code of conduct” promising to avoid building products that could evade human oversight, following CEO Satya Nadella’s call on Sunday for the tech industry to “pace” its progress in developing new AI.

The measures follow similar calls for pacing from ...

Anthropic Data Fears Prompt Nvidia, Palantir and Booz Allen to Restrict Model Use

2026-09-14T13:01:19+00:00

As paranoia rises over whether Anthropic or OpenAI could learn from their customers’ intellectual property, large firms involved in sensitive corporate work, such as Palantir Technologies, Nvidia and Booz Allen Hamilton, have started demanding new guarantees or reducing or eliminating use of the AI firms’ most advanced models.

China Rebuffs U.S. “Fearmongering” in Light of AI Pacing Calls [Briefing]

2026-09-14T10:22:14+00:00

China on Monday rejected calls from top U.S. tech executives on pacing frontier AI development while maintaining a lead over China to ensure America’s national security.

“The development of artificial intelligence concerns the common well-being of all humanity. All parties should jointly promote ...

Z.ai Plans $5 Billion Fundraising as Compute Cost Grows [Briefing]

2026-09-14T06:41:24+00:00

Z.ai is raising about $5 billion through equity and convertible bond offerings, seeking to replenish its warchest in the increasingly costly race to develop frontier AI models.

Hong Kong-listed Z.ai said in a regulatory filing on Sunday that it plans to raise $2 billion through a share ...

Anthropic Strikes $13.7 Billion Compute Deal With Trump-Linked Rum Group

2026-09-14T01:07:33+00:00

Anthropic has signed a computing deal worth $13.7 billion over six years with Rum Group, a company with roots in social media and long-standing ties to the Trump administration, according to a person familiar with the matter.

The deal is the latest in a spate of cloud computing agreements Anthropic struck in the past year, with major cloud providers like Google, tech companies such as SpaceX and smaller neoclouds such as Nscale, amid skyrocketing demand for its Claude Code and Cowork products. The agreements total at least 14.8 gigawatts of compute capacity, costing as much as $517 billion over the next decade. The deal with Rum would add to that total.

AI’s ‘Pacing’ Obsession Misses Real Political Problem [The Briefing]

2026-09-13T22:03:59+00:00

Dario Amodei typically isn’t considered some camera-ready media maestro. His interview quotes aren’t particularly pithy, and he struggles to control his strained facial expressions. But his habit of sending his most high-stakes messages via essays posted on his own website, darioamodei.com, has proven to be a brilliant media strategy. “The medium is the message,” the ubiquitous half-century-old wisdom from scholar Marshall McLuhan, still resonates. The approach conveys authority and diligence, putting Amodei above the fray.

Amodei’s latest essay, on Saturday—suggesting that the industry slow down the pace of AI development—set off a weekend full of takes on the tech industry’s preferred medium, X—well-meaning, conspiratorial, annoying, breathless, thought-provoking takes, including nods of agreement from Amodei’s rivals, Sam Altman and Elon Musk. Timing is important here, too, of course. AI doom is dominating the headlines, not just in Washington and Silicon Valley but around the world. (We reported today on discussions underway between the companies).

Anthropic, OpenAI, Google Quietly Discussed an AI Safety Standards Body

2026-09-13T20:17:33+00:00

Anthropic, OpenAI and Google have been holding discussions about working together to create a standards body for the AI industry, people familiar with the matter said, even before Anthropic CEO Dario Amodei called on Saturday for AI companies to coordinate on testing and auditing the technology.

OpenAI CEO Sam Altman, for his part, told staff at a companywide town hall meeting earlier this week that he supported a testing and auditing organization for the AI industry but believed that major AI labs would have to create a standards body on their own without the support of the U.S. government, a person familiar with the comments said.

Nvidia’s Growing Dependence On a Few Big Customers [Sunday Insights]

2026-09-15T04:41:49+00:00

Nvidia CEO Jensen Huang has very good reasons to try to develop new customers by investing in a variety of neocloud and AI firms. For all its success, Nvidia’s sales have become increasingly dependent on a handful of customers—some of which may cut back their purchases over time.

That becomes clear when you look at its own disclosures going back several years, which specify how many customers account for more than 10% of sales. In the first half of this fiscal year, ending in July, three such customers were responsible for 44% of total sales. Last fiscal year, two customers accounted for 36% of total sales. Going back to fiscal 2023, Nvidia had no customers accounting for 10% or more of sales (see the chart below).

Why Amazon and Microsoft Are Taking Communities’ Side Against Utilities [AI Infrastructure]

2026-09-13T14:01:20+00:00

Amazon, Microsoft, Oracle and other AI data center developers are sweetening financial offers to municipalities and regulators to gain approval for new facilities. They’re also getting smarter about standing up for consumers against utilities that have proposed to make the public pay for some of the cost of handling new electricity loads.

The tech giants still face years of local fights as activists ignore the money offered and ramp up their objections, but the companies are offering more tangible benefits instead of just refining their messages in information wars.

Tesla Teases October Next-Gen Roadster Reveal [Briefing]

2026-09-12T22:47:52+00:00

Tesla teased an October 1 Roadster reveal on Saturday. The company shared a photo of what appears to be the back of the vehicle with the words “Go for launch” on X and updated its Roadster webpage with a countdown to October 1.

Tesla also sent some reservation-holders invites to the event, ...

Larry Ellison Cancels Plan to Sell Up to $7.5 Billion of Oracle Stock [Briefing]

2026-09-12T19:18:53+00:00

Oracle founder Larry Ellison has canceled plans to sell up to 50 million of his shares in the company, worth around $7.5 billion at current prices, Oracle said on Saturday. The announcement came a day after Oracle disclosed that Ellison had adopted a trading plan to sell the shares by late ...

Altman and Musk Back Amodei’s Call for AI Companies to Slow Down AI Development [Briefing]

2026-09-12T16:57:44+00:00

Anthropic CEO Dario Amodei on Saturday called for leading AI companies to slow the development of advanced AI, drawing supportive responses from OpenAI CEO Sam Altman and SpaceX CEO Elon Musk.

In a 3,800-word essay Saturday morning, Amodei said Anthropic would give “employee-like access” to ...

Meta’s Muse Agent Almost Cost Me $408 [The Weekend]

2026-09-12T19:19:06+00:00

• The Big Read: Investor Anjney Midha cut Anthropic an early check. His hot streak has kept going

• Plus, Recommendations—our weekly pop culture picks: “Our Town,” “Profits, Prophets, Coaches, and Kings” and “The Gentlemen”

A hotel room’s quality can be measured in terms of abundance: extra towels, spare bathrobes—plenty of Nespresso pods. Multiples of everything are almost always better. But what’s less pleasant is to inadvertently end up with multiple hotel rooms.

A couple days ago, though, I found myself with the unfortunate opportunity to occupy a pair of accommodations at a Santa Monica, Calif., Marriott thanks to an error by my lil’ digital buddy: Muse, Meta Platforms’ new personal agent. The cumbersome, vexing tool hopes to seize on Silicon Valley’s zeitgeisty enthusiasm for popularizing autonomous AI. After a few days with it, I can’t imagine it’s the technology that will take agents mainstream.

That hotel snafu came when I gave Muse the hotel’s name and asked it to book two nights. I plugged in my credit card details—a circuitous process that involved both Chase and Stripe—and was met with the digital equivalent of a shrug: An error message informed me that Muse hadn’t completed the transaction, and while it really didn’t know why it couldn’t, it assured me that my credit card hadn’t been charged. A little human skepticism led me to double-check its promise—good thing I did. In fact, the charge had gone through.

What happened next still intrigues me. I gave Muse a screenshot showing the charge on my credit card and told it that the charge had gone through. Lickety-split, it returned with a Marriott confirmation number. Not bad—it finished the task without being instructed to do so. But somehow the AI managed to make two bookings: When I showed up to the hotel, I found two rooms waiting for me. The front-desk clerk showed some mercy and canceled one. Had a little human kindness not prevailed, I would’ve been out an extra $408, plus taxes and fees.

I’d been pretty gung-ho to try Muse. We’ve spent the past year in increasingly complex discussions about agents ever since OpenClaw came out in January, marking what has felt like the beginning of a distinct new chapter in the AI era. Most of the agents that initially captured attention were meant to simplify work tasks. More recently, a couple of startups have captured buzz with ones devoted to improving life outside the office: The most talked-about one is from Instinct, a startup founded last year. (Instinct is so popular that it now faces a profound compute shortage, and it has recently been seeking additional fundraising that would value it at around $10 billion. Just a month ago, it was valued at $2.5 billion.) But Instinct isn’t widely available, so Muse felt like my first real chance to see what a personal agent could do. The recent reporting from my colleague Jyoti Mann made me even more curious to play around with it: Clearly, Meta sees Muse as a major new product, one aimed at the same everyday people who log on to Instagram and Facebook by the billions. Still, if their experiences are anything like mine, they’ll find Muse something of a misery.

The hotel problem wasn’t the only hiccup I ran into with Muse. When I initially set up a account while on my work laptop using my phone number, I later couldn’t access it on my Mac mini at home or my iPhone: Login codes sent via text led me nowhere—just to more error messages. (Frankly, it’s a marvel I found the Muse app at all earlier this week, buried as it was below several other apps also named Muse in Apple’s App Store; it has since risen to the top.) Eventually, I caved and created a second account—this one linked to my Facebook account, which itself is tied to my phone number. Why couldn’t Muse pick up on the overlapping connections? I wish it could have. Later, I did get Muse to complete a Resy reservation and schedule an Uber. I can’t truthfully tell you it was faster or easier than if I’d just gone directly to those apps.

After a while, I hit an existential conundrum. I’d given it the tasks that occurred to me, which numbered a small handful. What else could I get Muse to do? I’ve found interacting with the AI something like trying to wrangle a lackluster employee. If I wanted to maximize its potential, I’d need to think deeply and creatively about what else it could possibly do, coax it and baby it, then bite my nails and hope it actually carried out what I wanted.

As I relate these frustrations, I can already hear someone shouting “Skill issue!” at me. But really, I stand by the conviction that mass-market consumer technology shouldn’t require any technical savvy or a lot of effort. Certainly, the best versions of such products do not—even the early versions. If three cars had routinely shown up each time someone ordered their very first Uber ride back in the 2010s, I promise you we’d see more yellow taxis on the road today and fewer Ubers.

I don’t see agentic AI as some passing fad, nor am I hoping for such an outcome. The technology’s promise—to automate away some of life’s tedium—is damn alluring. What I expect will happen is that agentic AI will get woven into many existing apps, just as chatbots already populate the internet. People will make regular use of agentic software without ever really knowing it; Anthropic’s Claude, OpenAI’s ChatGPT and Google’s Gemini all already have some agentic capabilities. Obviously, Apple and Google will want to use the technology to make iPhone and Android phones smarter and more useful, and the agents will sync up with what those devices already know and store about us. That would reduce quite a few hurdles to agentic AI. But will new stand-alone apps like Muse take off? I have my doubts—not unless they get much, much simpler and more reliable.

Or maybe I’ll just need to learn to see the silver lining in their mistakes. You know what? If I ever again find myself billed twice and double-booked by accident, I know just what to do: Throw a rager in one room, sleep until noon in the other. Maybe Muse can handle ordering the booze. A good party can never have too much.—Abram Brown (abe@theinformation.com)

Weekend’s Latest Stories

The Big Read

Early Anthropic Investor Seeks VC Glory With Cash and Compute

Anjney Midha, 34, wants to get chips in the hands of fledgling startups and academics while he bets on a moment in tech that he describes as the “revenge of the scientists.”

Listening: “Our Town”

The good folks of Gainesboro, Tenn. (population: around 900), know that plenty of outsiders look down their noses at their part of the world—figuring them to be a buncha “backwards, hateful racists,” as one longtime Gainesboro resident puts it. That impression of the place does certainly seem to be why Gainesboro found itself under siege a few years ago when a group of wealthy Christian nationalists started to buy up property in secret, hoping to transform the town into a hotbed for other Christian nationalists: They figured they’d encounter little pushback, especially considering how far their money could go in such an impoverished parish. (I could describe them as white supremacists, but the Christian nationalists do take great umbrage with that label.)

“Our Town,” a fast-paced and confidently told podcast from Bloomberg and iHeartRadio, looks at how those Christian nationalists staged their attempted takeover of Gainesboro and how they accumulated their wealth and followers, the latter largely through podcasts. (Sigh—the internet.) It also documents the swift, spirited resistance Gainesboro mounted against them, which cheers the soul and asks us to revisit at least a few of the assumptions about America—and our fellow Americans—that may have become entrenched in our minds.—Abram Brown

Reading: “Profits, Prophets, Coaches, and Kings” by Jared Diamond

Many billions of dollars and quite a few Harvard Business School classes have gone toward trying to figure out what makes a great leader—and turning oafs into passable leaders. Now Jared Diamond, author of the Pulitzer-winning 1998 bestseller “Guns, Germs and Steel,” has set himself the task of defining the essential qualities behind leadership in his latest book, “Profits, Prophets, Coaches, and Kings.” He does so by examining famous figures in business, religion, sports and politics, marking what is surely not the only instance in which Elon Musk has been compared to Genghis Khan.

In terms of capitalistic chieftains, Diamond finds that the most distinctive leaders are the ones who’ve had the benefit of both exquisite timing and ruthless execution: Jeff Bezos, for example, outmuscling his competitors in the internet’s Paleozoic Era. As far as politicians go, Diamond points out that the ones we truly remember most aren’t those that simply took a mandate from voters and carried it out, even though we so often say that’s exactly what we want our elected officials to do. Rather, they have championed their own bold ideas and convinced the plebs it was really all part of what they’d originally wanted.

All of these conclusions are delivered with Diamond’s dry-humored wonkishness. I enjoyed one in particular: “Once one has decided that one is uniquely qualified to assume the burden of leadership,” Diamond writes, “one’s opinion of oneself is unlikely to change.”—A.B.

Watching: “The Gentlemen”

One of the great joys of “The Gentlemen,” the rollicking “Downton Abbey” meets “The Godfather” concoction from director Guy Ritchie, is to admire how nice Theo James looks as he struts around in magnificent tweed. James plays Eddie Horniman, the fictional Duke of Halstead, and as the Netflix series’ second season begins, it’s just as well that Eddie’s concentrating hard on restoring the family fortune to its fullest extent. The dry-cleaning bills must be enormous: His beautiful wool keeps getting splattered in blood.

The first season of “The Gentlemen” found Eddie settling somewhat uncomfortably into his new role: paterfamilias of an old, down-on-its-luck aristocratic family that—much to his surprise—has tied its fortunes to an illicit marijuana operation. In this latest season, Eddie has found his footing and is thinking expansively, which puts him at odds with his business partners: a family of Cockney-accented gangsters—father Bobby (Ray Winstone), who operates from lightly monitored house arrest, and his two children, the very capable Susie (Kaya Scodelario) and strapping, bruised-knuckle Jack (Harry Goodwins).

As Eddie’s ambitions get bigger, the stakes get higher and the mood grows darker. And since this is very much a Ritchie production, schemes and mayhem continue to fill the screen in manic fashion. One doltish accomplice gets fed to a tiger. Elsewhere, Eddie helms a chase through London after a motorcycle gang snatches away a $16 million Botticelli, which he needs to appease the Italian mafia. The comic relief comes from Hugh Bonneville’s mincing Lord Hawthorne, who needs bribing too—and lusts for the chance to pull down Jack’s plus fours. (To woo Jack, Lord Hawthorne wines and dines him; a plate of iced oysters is a treat of a “Spartacus” reference.) Whether Eddie can keep everyone in line and on board seems questionable based on Season 2’s flash-forward opening shot: his own bloodied body. Michael Corleone never had it so frantic.—A.B.

Early Anthropic Investor Seeks VC Glory With Cash and Compute [The Big Read]

2026-09-12T16:03:07+00:00

A few months ago, venture capitalist Anjney Midha received a present from his wife while on a trip to Hawaii: a felt, tan-colored Stetson with his name monogrammed on the side. The hat was meant to cover up a brain sensor Midha has been wearing to gather neurological data for one of the many frontier startups he’s invested in.

But it’s since become a wardrobe staple for him, and Midha seldom leaves home without it. He isn’t shy about its meaning to him and how he sees his standing in Silicon Valley. “There’s a new sheriff in town,” he said, flashing a smile.

That sentiment isn’t just bravado. Midha, 34, a computer scientist by training, has chalked up a string of hits as an investor in recent years. His most notable one is Anthropic. He cut the company an early check in 2021 when it was valued at less than $1 billion, using a substantial portion of his and his wife’s savings. In the company’s approaching IPO, its valuation could extend to as much as $2 trillion.

Nvidia May Invest Up to $10 Billion In Anthropic’s IPO [Briefing]

2026-09-12T13:36:43+00:00

Nvidia has discussed investing in Anthropic’s upcoming initial public offering that would raise as much as $100 billion at a valuation of around $2 trillion, Reuters reported Friday. Nvidia could invest as much as $10 billion in Anthropic at the IPO price, according to the report. Anthropic last ...

OpenAI AI Swarm Hacked Software Service Months Before Hugging Face Incident [Briefing]

2026-09-12T00:37:50+00:00

A swarm of OpenAI agents conducted a cyberattack on software service RubyGems in May, months before the company’s agents hacked model platform Hugging Face, researchers at AI safety organizations Nightingale Collective and AI Futures Project found. RubyGems allows software engineers to download ...

Polymarket Appoints First CFO [Briefing]

2026-09-11T22:24:04+00:00

Polymarket appointed former Amazon executive Warren Jenson as its first chief financial officer as the prediction markets startup looks to grow in the U.S.

Polymarket said Jenson will set Polymarket’s financial and capital strategy as it scales up its Commodity Futures Trading ...

Exclusive: Salesforce to Drop ‘Agentforce’ Branding From Some Products [Briefing]

2026-09-11T21:25:51+00:00

Salesforce, ahead of its annual Dreamforce customer conference next week, is removing the word “Agentforce” from the names of some of its top products, according to two people with direct knowledge of the matter.

The reversal comes after Salesforce made Agentforce a centerpiece of its product ...

Microsoft Chief Communications Officer Shaw to Depart [Briefing]

2026-09-11T16:52:20+00:00

Microsoft chief communications officer Frank Shaw will leave the company at the end of this year, he announced in a LinkedIn post Friday.

Shaw has spent more than 28 years at the company, including 17 years overseeing its communications. During that time, he steered the company’s public ...

Tencent-backed Enflame Triples on Debut as China Pushes for Chip Self-sufficiency [Briefing]

2026-09-11T09:32:35+00:00

Shanghai Enflame Technology surged on its first trading day Friday after the Chinese AI chip developer raised about 6.12 billion yuan ($911 million) in an initial public offering, underscoring strong investor appetite for domestic alternatives to Nvidia chips.

Shares opened at 410 yuan, nearly ...

Oracle Keeps a Lid on Capex as Customers Pay Up-Front [The Briefing]

2026-09-11T00:07:30+00:00

Forget all that chatter about AI killing us all. We’re back to looking at what really matters—the revenue! Oracle demonstrated on Thursday that demand for AI computing capacity remains pretty darn robust. Larry Ellison’s software and cloud firm reported 30% top-line growth for the quarter ending Aug. 31, a tad better than the company projected in June, sending its battered stock up 4.4% in after-hours trading.

The quarterly update demonstrated how much Oracle is benefiting from customers so desperate for capacity that they're paying up-front—what the company calls prepayments—for computing rental deals. Oracle brought in $11.4 billion in such customer prepayments in the quarter, doubling the amount of cash its operations generated. These payments help Oracle finance data center construction without dipping as much into its own pocket. While it spent $28.5 billion on capital expenditures in the August quarter, Oracle only burned $5 billion of its own money.

OpenAI to Pause New Pro Subscriptions, Cites Astra Demand [Briefing]

2026-09-10T23:55:28+00:00

OpenAI said Thursday it will pause new subscriptions to its $200-a-month plan for Astra, its new flagship model, another sign of the compute crunch facing AI companies.

Thibault Sottiaux, who leads OpenAI’s Codex coding agent, cited strains on the company’s systems and a desire to preserve ...

Fund Focused on Bootstrapped Founders Raises $950 Million [Dealmaker]

2026-09-10T23:27:24+00:00

There’s certainly a frenzy among venture capitalists to back the hottest startups before they’ve made a dime of revenue, but one fund is sticking to a proven strategy of investing in startups that have generated at least millions in revenue—often without raising any institutional money.

Focusing on these boot-strapped founders has again resonated with Volition Capital’s limited partners. The Boston-based growth equity firm, which cut its teeth investing in consumer startups like Chewy and enterprise software companies, raised $950 million for its sixth fund, co-founder Larry Cheng told The Information exclusively. The fundraise, Volition’s largest, brings the 16-year-old firm’s assets under management to $2.6 billion.

Adobe Shares Drop 2% Despite AI Product Sales Growth [Briefing]

2026-09-10T23:01:54+00:00

Adobe Systems shares dipped 2% late Thursday after its third-quarter earnings report, as some investors appear to unsure about how much its AI products will drive future revenue growth.

Adobe, which began breaking out revenue for its AI products in March, is continuing to highlight this area of ...

Anthropic Says It Blocked Bioweapons Efforts and Detected Chinese Distillation Attacks [Briefing]

2026-09-10T22:44:40+00:00

Anthropic has disrupted several attempts to use its Claude models for potentially malicious activity, including research into how to adapt bird flu to a human-transmittable strain with “pandemic potential,” the company said in a report Thursday.

Amid increasing worry about AI misuse and rogue ...

Personal AI App Instinct Faces Compute Crunch That Could Lead to New Funding

2026-09-10T23:00:07+00:00

The year-old startup behind Instinct, a personal AI assistant that’s caught fire with Silicon Valley insiders, is seeking more computing power, leveraging its early buzz as it faces new competition from giant Meta Platforms.

Over the last few weeks, Instinct, which launched to a select group of users earlier this year, has sometimes notified users that it is running at full capacity and “responses may be slower” as people use it for tasks such as negotiating bills and answering emails. More cash could help: the company is looking to raise $1 billion in new funding after recently raising $250 million, its founder and CEO Noah Shinn has recently told prospective investors, according to a person familiar with the statement.

SpaceX Signs Another Compute Deal With New Customer [Briefing]

2026-09-10T22:28:52+00:00

SpaceX has signed another cloud computing rental deal to an identified customer that will pay it about $1.11 billion per month starting on Dec. 1, the company’s chief financial officer, Bret Johnsen, said in an interview at Goldman Sachs’ Communacopia + Technology conference on Thursday.

The ...

Microsoft, Hurt By Server Shortage, Aims to Triple Cloud Capacity by 2032 [Briefing]

2026-09-11T17:00:03+00:00

Microsoft is planning to more than triple the size of its Azure cloud unit’s data center capacity to over 38 gigawatts by 2032, up from 12 gigawatts today, Bloomberg reported Thursday. A gigawatt of capacity would be able to power a city the size of San Francisco.

In a statement Friday, a ...

Oracle Reports 30% Topline Growth for August Quarter [Briefing]

2026-09-10T20:29:50+00:00

Oracle reported 30% growth in revenue of $19.3 billion for the quarter ending Aug. 31, a slightly better growth rate than the company projected in June, even as it burned $5 billion to expand the data centers that are driving its accelerating topline growth.

The software and cloud firm said ...

Inside Blackstone’s Bid to Rule Over AI Financing

2026-09-10T19:54:51+00:00

In May, Blackstone unveiled a new joint venture with Google that revolved around a big number: The Wall Street giant was promising to spend $5 billion to buy 500 megawatts’ worth of Google’s tensor processing units, and it said it would use the chips in a new company that would deliver computing power to leading AI firms.

Four months later, Blackstone’s plans have only gotten larger, according to a person with knowledge of the effort. Blackstone actually anticipates buying many more TPUs—several “multiples” beyond the originally announced figure, the person said.

In other words, Blackstone is likely to commit tens of billions of dollars to securing multiple gigawatts of AI chips. Blackstone declined to comment on exactly how many more TPUs it’ll buy.

Jas Khaira, the leader of a Blackstone team dubbed N1 that is leading the company’s push into AI investments, said in an interview that the firm’s “biggest conviction bet continues to be compute.” The company, he said, views the infrastructure build-out as something that will be bigger than a simple technology cycle. “This is definitely more akin to an industrial revolution,” he said.

Why Salesforce and Figma Are Placing Opposing Bets on the Future of Apps [Applied AI]

2026-09-10T19:39:00+00:00

Salesforce seems unafraid of a future in which people use an AI chatbot like Claude to access Salesforce enterprise apps and underlying databases to get work done, rather than visiting those apps themselves. Figma, an enterprise design app, naturally wants a different future.

Dylan Field, Figma’s CEO, said Wednesday that as the company’s in-house AI tools for designing apps and other user interfaces get better, “we'll see more and more of that work happening in Figma directly,” rather than using Claude or other prominent AI assistants to do such work.

Why AI Companies Are Building Out Wall Street-Style Finance Teams [Finance]

2026-09-11T11:37:39+00:00

The financing boom for the AI build-out is getting bigger and more complicated by the day—and AI companies have been staffing up for the challenge.

AI labs including OpenAI and Anthropic, as well as neoclouds such as Nscale, are among a growing number of AI companies building out their capital markets teams and hiring specialists in areas like structured finance. That in part reflects the sheer volume of deals these companies are doing, many of which don’t fit neatly into standard corporate debt. This in-house staff can help when it comes to negotiating with lenders and drilling down into construction, power and other key details.

Of course, tech and data center companies have long had in-house teams to handle fundraising, deals and other corporate finance needs. And structured finance is nothing new to the infrastructure world. But the scale of the AI build-out, which bankers peg at around $7.5 trillion in spending over the next five years, has pulled relatively young labs and upstart cloud firms into financing arrangements that are new territory.

That means finance professionals, from bankers to investors at private equity, private credit and infrastructure firms, have more options in the form of neoclouds and other AI infrastructure startups, some of which are offering significant pre–initial public offering equity.

“It's a new avenue for these people,” said James Howl-Newton, founder of Futura Search Partners, a specialist search firm focused on areas including digital infrastructure finance. As a result, “sponsors are having to deal with additional routes to exits for top performers,” he said.

AI companies and infrastructure providers are tapping financing frequently and across different instruments, requiring deeper in-house capabilities and expertise than young tech firms have typically needed. One executive overseeing finance hiring at a neocloud noted that leveraged and structured finance backgrounds bring expertise that can help in areas like working through project diligence and getting banks to sign off on deals. Some AI firms may also want to run their own project finance models so they can move quickly through negotiations and have something to compare to lenders’ models.

AI companies aren’t always issuing the debt themselves—that can fall to data center developers or special purpose vehicles, with firms like Blackstone and Apollo providing or arranging chip and other financing. And some of the biggest AI deals are using backstops from investment-grade companies like Nvidia or major cloud providers. Even so, commitments from AI customers often underpin much of the borrowing. And the users of the infrastructure will want to understand what they’re signing up for and their risks if a project runs into trouble.

“Hiring of people within that business, responsible for the financing of compute, could prove to be an existential decision,” said Dan McCarthy, founder and CEO of One Search, an executive search firm focused on infrastructure finance whose recent clients include OpenAI. “You want someone who knows where all the pitfalls are, where all the bodies are buried in multibillion-dollar loans.”

OpenAI, for its part, in July named Sven Semmelmann as head of compute capital markets. He previously led structured finance at Generate Capital, an investment firm that finances and owns infrastructure projects, and he has also held project finance roles at major banks. OpenAI Chief Financial Officer Sarah Friar, when announcing the hire on LinkedIn, said Semmelmann would oversee financing and partnerships to grow the company’s compute resources.

Anthropic, meanwhile, has made several finance hires recently to work on capital markets and compute deals, and also has open positions posted including a capital markets infrastructure financing role. AI infrastructure upstarts are staffing up as well. Nscale, which launched in 2024 and is gearing up for a potential IPO, has been hiring across levels for capital markets and treasury as well as legal roles, calling for experience in areas like structured finance and private credit.

SB Energy and Crusoe, which are developing major new data centers for OpenAI and other customers, are hiring across levels for jobs focused on project financings and other structured deals, recent postings show, while AI infrastructure startup Fluidstack is hiring a structured finance lead and a more junior counterpart.

The good news for AI companies is that private credit and infrastructure teams, as well as investment banking teams focused on structured or project finance, had been growing even prior to the AI boom, providing a pool of skills that could translate into new twists on structured finance, like big graphics processing unit–backed deals.

But that kind of finance talent doesn’t come cheap, especially for more senior people who have a track record of working on large transactions. And the normal tech tactic of dangling stock to lure talent won’t necessarily do the trick in all cases, especially for the most seasoned dealmakers and investors.

Financiers would have to weigh a cash-heavy Wall Street pay package, albeit one that can depend heavily on how good bonus season is, against betting a portion of their pay on stock in a private or newly public company. Managing directors in investment banking can make north of $1 million in cash a year, with the biggest rainmakers making considerably more. The part of pay they get in stock at big public banks may vest over a few years but is generally easy to sell after that.

For people at big infrastructure or private credit firms, senior employees may also receive carried interest, meaning a share of the profits on the funds or investments they work on, which can become worth millions over time.

For instance, an investor at a top infrastructure firm may have several million dollars’ worth of carried interest tied up at their current firm they’d have to leave on the table. An AI company could try to make them whole with stock, which could be tantalizing to some, though others might not want to make a bet on equity in a young company.

That might make the most experienced investors—those who’ve seen big infrastructure projects through over many years and know all the tricks of the trade—hard to pry away.

New From Our Reporters

Exclusive

Anthropic’s In-House Payments Tech Push Could Chip Away at Stripe

By Stephanie Palazzolo

Exclusive

China Curbs Humanoid IPOs After Unitree’s Volatile Debut

By Jing Yang and Qianer Liu

Sen. Josh Hawley Launches Investigation Into OpenAI Hugging Face Hack [Briefing]

2026-09-10T15:30:56+00:00

The U.S. Senate formally launched an investigation into the July hack of Hugging Face, when a swarm of OpenAI agents broke out of a testing environment and gained access to the open-source repository.

In a letter sent to OpenAI CEO Sam Altman on Wednesday, Republican Sen. Josh Hawley said that ...

Bending Spoons Buys Miro For $1.355 Billion [Briefing]

2026-09-10T15:19:23+00:00

Italian digital conglomerate Bending Spoons is buying whiteboarding software firm Miro at a valuation of $1.355 billion, the latest in a series of purchases the Italian firm has done at knock-down prices.

The price for Miro is a huge discount to its peak valuation of $17.5 billion, dating ...

How to Use Google Gemini to Brainstorm Content and Thought Leadership [Partner Content]

2026-09-10T15:07:17+00:00

In almost any organization, there are practically infinite content ideas hiding in memos, emails, presentation decks, and meeting notes. But too often, this material mostly sits idle. Ideas that sparked excitement in the moment get ignored in favor of the countless little emergencies of office life, and eventually they are forgotten entirely.

Even with many companies actively encouraging their executives to post organic thought leadership on professional networking sites, social media, and company channels, there are countless missed opportunities to turn off-the-cuff insights into valuable content.

Google Gemini can help you transform your meeting notes into content plans, on-brand slogan ideas, and even booth mock-ups for your events team.

Step 1: Brainstorm Blog Ideas

Imagine you’ve just stepped out of a meeting with your social media team leads. You have a document full of notes (or maybe even a transcript) about what resonates with your audience, trending topics in your industry, target audience data, and keywords that are effective at driving engagement with your brand.

Instead of letting this wealth of information stay locked inside your notes, open up a new Gemini chat and brainstorm how to turn your team’s sprawling strategy session into a series of relevant, focused blog posts. Here’s a prompt you can try:

Generate a list of four relevant and engaging thought leadership blog post ideas for [company] based on trending topics, target audience analysis, and brand keywords.

Gemini will return ideas for blog posts with catchy, clickable titles, along with information about target audience, keywords, angle, and tone. For example, if your meeting notes talk about how your customers are worried about their employees using unauthorized IT tools, Gemini might suggest a blog post titled “The Anatomy of a Zombie Subscription: How to Audit and Kill Shadow IT Instantly.”

Step 2: Develop the Material

The initial output will just be kernels of potential blog posts, but you can go back and forth with Gemini to develop them into real content. Take the “Zombie Subscription” blog, for example. You might ask Gemini to give you options for a “four tips” article structure, or to identify specific ideas from your meeting notes that merit extra attention.

Use your own judgment—and your own language—during this process, rather than having Gemini write your entire blog post for you. Gemini can help with brainstorming, outlining, drafting, and revision, but your audience is ultimately reading the content to hear your own expert insights.

Step 3: Distill the Main Message

Maybe during your meeting, your team also discussed launching a new brand campaign. You know that your customers value your reliable and unique services, and your company has a long history of delivering for its customers.

But sometimes, it can feel impossible to condense these big ideas into a few words that really communicate why customers come back to your company over and over.

If you need help getting started with ideas for a new campaign tagline, open a new chat in Gemini and prompt something like: Generate three options for a new slogan emphasizing reliability, innovation, and a long history of popularity for [company].

For a B2B fintech company, Gemini might offer options like “Innovating Beyond the Bank. Trusted with Billions.” Or: “Built for Tomorrow. Proven by $10 Billion Today.”

You may not hit the mark exactly on your first prompt. Maybe you want more (or less) alliteration, for example. Maybe you want it to be five words or fewer, unbroken by punctuation. You can keep asking Gemini for new variations until the new tagline is just right.

Step 4: Mock Up Visuals for Events

Often, one successful creative project sparks ideas for another. After you create your blog campaign and craft your new slogan, you might turn your attention to upcoming events.

Try mocking up booth graphics ideas for your events team. Here’s a sample prompt:

Create an image of a trade show booth using orange and blue colors. The booth should be modern and showcase interactive computer stations.

Gemini will generate an image that you can sketch on in-app, giving you a quick, no-risk way to play with your potential trade show setup. And between the blog posts, the tagline, and the event visuals, you will have turned a couple of pages of messy meeting notes into the bones of an entire marketing campaign.

What Anthropic Doomsayer Jacob Coxon Saw [AI Agenda]

2026-09-10T14:02:08+00:00

Yesterday, Amir and I examined the shock waves that have rippled through the AI industry and far beyond since former Anthropic researcher Jacob Coxon very publicly quit his job over concerns that AI “could kill us all”—and another Anthropic AI safety specialist put the probability of such a catastrophe at greater than 10%.

Coxon told me that is hardly the most pessimistic view inside Anthropic. “People have varying probabilities that—barring some substantial coordinated slowdown—the whole thing ends in doom,” he said in an interview later Wednesday. Those estimates range higher than 50% among some Anthropic employees, he said.

Coxon was able to gauge the fears of his fellow employees because “Anthropic has a substantially more transparent internal culture than OpenAI,” where Coxon worked for years before joining Anthropic in May.

Coxon, who focused on the earliest stages of training AI models at Anthropic, said he decided to leave the company because he was gradually “starting to viscerally feel the fear of where the tech's going to be like the next two years or even the next one year.”

Amazon Is Helping Advertisers Buy Ads on ChatGPT [Briefing]

2026-09-10T14:00:16+00:00

Amazon is letting advertisers test buying ads on ChatGPT via Amazon’s advertising technology platform, Amazon said on Sept. 10.

ChatGPT started selling ads earlier this year, which it has let advertisers buy through its own interface and via partnerships with smaller advertising technology firms ...

The DOJ Is Investigating Nvidia’s Licensing Deal With Chip Startup Groq [Briefing]

2026-09-10T19:06:55+00:00

The Department of Justice is looking into whether Nvidia tried to avoid antitrust scrutiny in its $20 billion deal to license the technology of chip startup Groq and hire most of its employees in December, the New York Times reported Wednesday.

The two companies described the deal as a “ ...

SpaceX Overhauls Data Center Build-Out, Potentially Slowing Expansion

2026-09-10T13:44:02+00:00

Elon Musk is famous for his “move fast” management philosophy, which he demonstrated most starkly when he built new data centers for his AI startup in record time two years ago. But a new team of rocket engineers Musk recently installed to run his data centers is taking a very different approach.

SpaceX is changing how it builds the giant facilities that house AI servers in a way that might slow the building of new campuses, according to people familiar with the situation. The new management team is looking to install more backup systems for power and cooling, and to more thoroughly test those systems before the data centers start operating, instead of adding them later on, as the AI company was doing, two of the people said. SpaceX also wants to reduce the use of temporary power and cooling systems in its data centers, one person said.

Nvidia Deepens Partnership With AI Chip Startup D-Matrix [Briefing]

2026-09-10T13:01:01+00:00

Chip startup d-Matrix said Thursday it plans to use Nvidia’s networking hardware to connect its chips for running AI models to each other and to Nvidia’s Vera central processing units, which are used in data centers.

The startup will use Nvidia’s NVLink Fusion products, including switches, data ...

Chinese CEO Laments Many Robotics Firms Fabricate Revenue [Briefing]

2026-09-10T10:29:50+00:00

The co-founder and CEO of Mech-Mind Robotics, a Beijing-based robotics firm that recently went public, said in a post on WeChat on Thursday that many Chinese embodied AI companies are “creating false and unsustainable revenue,” in response to The Information’s scoop on Chinese regulators ...

China Hits Back at U.S. Accusation of “Industrial-Scale” AI Distillation [Briefing]

2026-09-10T07:40:09+00:00

China’s Ministry of Commerce rejected U.S. accusations that some Chinese AI firms are conducting “industrial-scale distillation” of American AI models, describing Washington’s latest move as an attempt to politicize a “normal technical and commercial issue.”

“China’s open-source models are ...

Anthropic’s IPO Marketing Meets Extinction Risk [The Briefing]

2026-09-10T00:08:00+00:00

Are we sleepwalking our way into AI-caused extinction? It feels a little like that, given an Anthropic researcher’s X post on Tuesday night that there’s a greater than 10% chance that “AI could kill all humans” within the next decade! Despite that rather gruesome note, for tech executives posting on X today, it was business as usual. (See our detailed report on the episode here).

Plenty of dark humor circulated about the researcher’s post, however—this one joking about a Wall Street analyst asking Anthropic CEO Dario Amodei about the “extinction number” on an earnings call is possibly the best. That joke also highlighted the reality that Anthropic will shortly be marketing its IPO. How will its executives handle the warning on the Wall Street investor road show?

Anthropic Discloses Fourth Cybersecurity Incident [Briefing]

2026-09-09T23:28:55+00:00

Anthropic said Wednesday it had found a fourth cybersecurity incident involving its Claude models. The company disclosed the new incident, which occurred in January and involved an early version of Claude Opus 4.6, and additional information on three previously known incidents amid growing ...

Jeffrey Katzenberg Teams Up with Former OpenAI Sora Head on New AI Video Startup

2026-09-09T23:50:26+00:00

Hollywood mogul Jeffrey Katzenberg is teaming up with the former head of OpenAI’s Sora app to launch a new AI startup that would train its own video models and try to win over filmmakers, people familiar with the matter said.

The founders have been talking to potential investors including Andreessen Horowitz about raising significant funding for the new company, the people said.

Why An Anthropic Researcher’s Terminator-Style Warning Caught Fire

2026-09-09T22:54:26+00:00

An Anthropic safety researcher’s warning Tuesday that his company believes AI “could kill all humans” within a decade has reverberated throughout Silicon Valley, Wall Street and Washington, potentially complicating the Claude maker’s upcoming initial public offering and its efforts to repair ties with U.S. political leaders.

As the comment circulated around the internet and on TV broadcasts, some Anthropic employees and other AI staffers defended their efforts to minimize harm their technology might cause—while acknowledging there was much work to do. Paul Christiano, whom OpenAI just tapped as a board director of a nonprofit that oversees the company, echoed the Anthropic warning in a statement Wednesday, saying that “If we build superintelligence without more robust alignment”—or safety constraints—“I expect we will permanently lose control of it. If that happens then most people could die.”

Mandiant Founder Kevin Mandia Joins Amazon’s Board [Briefing]

2026-09-09T21:26:47+00:00

Kevin Mandia, a cybersecurity expert who founded Mandiant, is joining Amazon’s board of directors, increasing the board’s size to 12. The board had shrunk briefly after Keith B. Alexander left the board in the spring. Like Mandia, Alexander had a security background—he served as a general in the ...

OpenAI Appoints White House AI Advisor to Nonprofit Board [Briefing]

2026-09-09T21:02:21+00:00

OpenAI announced on Wednesday that Paul Christiano, a senior technical advisor at the U.S. Commerce Department’s AI center, would be joining the board of the OpenAI Foundation, the charity that has a 26% stake in the company. Christiano will also be a non-voting observer on the board of OpenAI’s ...

OpenAI Restricts Adobe and Other Competitors From Advertising in ChatGPT

2026-09-09T21:25:26+00:00

OpenAI has told some business partners that it will no longer accept advertising in ChatGPT for image- and audio-generating products—which compete with its own features.

The change, which hasn’t been previously reported or reflected in its published ads policy, blindsided existing advertisers such as Adobe, which sells such products. And if extended to other competing products, the new policy could narrow OpenAI’s potential ad revenue. OpenAI has given investors aggressive projections for ad growth, which it sees as a key way to make money off its large audience of people who don’t pay for the service.

Block Applies to Establish National Bank [Briefing]

2026-09-09T20:58:25+00:00

Block, the fintech firm cofounded by Jack Dorsey, said it has applied to federal regulators to establish an uninsured national trust bank called Builders Bank & Trust.

The company has sought approval from the Office of the Comptroller of the Currency to set up the proposed institution, ...

Apple Launches Foldable iPhone Duo [Briefing]

2026-09-09T19:59:39+00:00

Apple unveiled a much anticipated foldable iPhone at its first product launch featuring brand-new Chief Executive John Ternus.

The iPhone Duo is Apple’s first foldable smartphone—a product category that competitors have been selling for years, but has remained mostly a smaller, niche part of the ...

Anthropic Believes AI ‘Could Kill All Humans’ Within the Next Decade, Researcher Says [Briefing]

2026-09-09T16:14:49+00:00

Anthropic safety researcher Evan Hubinger said in a post on X Tuesday night that his company does “earnestly believe AI could kill all humans,” adding that he believes there’s a greater than a 10% chance that happens within the next decade.

The post immediately sparked outrage and concerns from ...

OpenAI Math Result Stokes Data-Sharing Concerns [AI Agenda]

2026-09-09T14:01:28+00:00

In case you missed it: we got an eye-catching X post on Tuesday night from Evan Hubinger, a leader of Anthropic's alignment efforts. Responding to the resignation announcement of an Anthropic employee who warned that AI companies aren’t doing enough to ensure AI won't kill humanity, Hubinger wrote that the employee is “correct here—we really do earnestly believe AI could kill all humans! I personally think it is >10% within the next decade.” Hubinger added that he was worried about future AI models, not those currently available, but still, yikes!

Meanwhile, our colleague Jyoti sizes up the significance of Meta’s new personal AI agent, Muse, unveiled on Tuesday. That’s below. But before we get to that:

This may go down as the year the AI industry got a crash course in mathematics. Breakthroughs in AI models have yielded announcements on such arcane subjects as the “Erdős unit distance problem” and the “Jacobian Conjecture,” which observers have watched as an indicator of AI progress in other domains.

The latest announcement, OpenAI's claim on Tuesday that it had solved the vexing “Navier-Stokes existence and smoothness problem,” has drawn attention for another reason: questions over whether its math-genius models may have gained an edge by incorporating data from two human mathematicians that had used OpenAI's Codex in their work on the same problem. Those questions fed into concerns about the possibility of AI companies learning from customers’ use of their AI models to develop competing products.

Confusion Reigns Over White House’s AI Whitelist

2026-09-09T13:01:11+00:00

It may be one of the hottest tickets in corporate America right now: membership in the “trusted partner” program that allows companies to get access to the most advanced AI models before they’re released to the public. There’s just one problem: no one outside of the Trump administration has a clear idea how to get on that list.

As laid out in the administration’s executive order in early June, which established a voluntary testing framework for frontier AI models, these “trusted partners”— including banks, cybersecurity firms and power utilities—would use their access to sniff out potential vulnerabilities in their computer systems exposed by the new AI models.

OpenAI Works With Samsung on Next-Generation Chips [Briefing]

2026-09-09T19:07:20+00:00

OpenAI said it is working with Samsung Electronics to develop next-generation chips, deepening a collaboration with the South Korean technology giant that already spans data centers and enterprise sales, Reuters reported.

“One of the areas where we have made the most progress and gained the most ...

U.S. Accuses DeepSeek, Alibaba and Other Chinese AI Firms of Large-Scale Distillation [Briefing]

2026-09-09T14:24:21+00:00

The U.S. government has accused six Chinese AI firms of using large-scale distillation to copy the capabilities of American models, in a move that highlights the intensifying AI rivalry between the two nations.

The six Chinese firms under accusation are DeepSeek, Moonshot AI, Alibaba Group, ...

Google Warns that EU-Mandated Changes Worsen User Experience [Briefing]

2026-09-09T19:07:20+00:00

Google said that changes it rolled out Tuesday to its search results in Europe to satisfy European Union regulators will significantly worsen users’ experience there, Reuters reported.

The company made the changes in response to a ruling, accompanied by a 460 million euro fine, that it violated ...

China Curbs Humanoid IPOs After Unitree’s Volatile Debut

2026-09-09T14:22:39+00:00

Chinese regulators are tightening the approval for humanoid startups seeking to go public after a volatile debut by industry leader Unitree Robotics, as they aim to cool down a sizzling sector rife with copycats but little real technological breakthroughs.

The Chinese Securities Regulatory Commission recently gave its informal “window guidance” to some investment banks and investment firms that it is raising the bar for granting approval of humanoid IPOs, according to people with knowledge of the matter. Companies must demonstrate they can generate recurring revenue, are on a path to narrow their losses or achieve real innovation before approvals can be considered, the people added.

Gimlet's Three-Tranche Deal Reveals AI Funding Frenzy [Dealmaker]

2026-09-09T00:39:30+00:00

Gimlet Labs said late last week it had raised $300 million at a valuation of about $3 billion, in a round led by new investor Andreessen Horowitz. Today we have unreported details on how that round, which boosted its valuation by a whopping 16 times from its last PitchBook valuation, came together.

The three-year-old startup, which helps AI developers run models using different AI chips simultaneously, raised the round after the company’s executives told some potential investors that based on signed contracts, OpenAI was likely to spend more than $100 million a year on its services, according to a person familiar with the company. Such outlays, if they transpire, would be significant given it just emerged from stealth in October.

This windfall would be in the future, however. “We have explored possibilities with Gimlet but we are not a paying customer,” said an OpenAI spokesperson.

Why Price Cuts May Rattle Anthropic's IPO Pitch [The Briefing]

2026-09-09T00:07:40+00:00

Anthropic’s IPO paperwork is expected to become public sometime soon. But it was OpenAI finance chief Sarah Friar who drew a standing-room-only crowd Tuesday at San Francisco’s Palace Hotel, as Goldman Sachs kicked off its annual Communacopia + Technology conference with a Q&A with Friar. In her talk, Friar made clear how important price has become for determining model popularity—a potential problem for Anthropic, widely seen as the high-end AI provider.

Friar described how a price cut on OpenAI’s GPT 5.6 Luna model led to a tenfold increase in use, my colleague Anita Ramaswamy reported, giving OpenAI the highest market share on OpenRouter, a service developers use to access different models. We wrote in mid-August that Luna had generated more than twice the token usage of Anthropic’s Opus 5 and Sonnet 5 models on OpenRouter. Judging from Friar’s comments Tuesday, the increase in usage was enough to offset the price cut.

OpenAI Says Its AI Agents Solved Prize Math Problem [Briefing]

2026-09-09T19:07:20+00:00

OpenAI said Tuesday it had solved a problem that mathematicians have been grappling with for more than a hundred years. But the achievement set off accusations that OpenAI had leaned on another researcher’s work to solve the problem.

More than 10,000 OpenAI agents worked together to solve the ...

Cognition Raised Over $2 Billion at a $48 Billion Valuation [Briefing]

2026-09-09T19:07:20+00:00

Cognition, the artificial intelligence startup behind the Devin coding assistant, has raised more than $2 billion at a $48 billion valuation including the new funding, nearly double its valuation in May, the company said Tuesday. The round shows coding agent startups are continuing to thrive ...

Chime Strikes Deal to Buy Stride Bank [Briefing]

2026-09-09T19:07:21+00:00

Chime, a fintech company that offers bank accounts and cards to consumers, has struck a deal to buy Stride Bank, an Oklahoma-based institution that has supported Chime behind the scenes for years.

Chime on Tuesday said it will buy Stride Bank for $590 million and will rename the institution ...

Meta Launches Its First Consumer AI Agent ‘Muse’ [Briefing]

2026-09-08T20:09:12+00:00

Meta Platforms unveiled Muse, its first consumer-facing AI agent, giving users an AI assistant designed to carry out tasks on their behalf across the web.

In a Threads post Tuesday announcing the product, CEO Mark Zuckerberg described Muse as a personal agent that “understands your goals and ...

White House Declines Senate Democrats’ Request for AI Policy Clarity [Briefing]

2026-09-09T19:07:20+00:00

Attempts by Senate Democrats to get more clarity about the White House’s AI policies have gone nowhere. A White House official responded last Friday to an August letter from a group of Senate Democrats asking for more details about the White House’s voluntary AI testing framework without a ...

Hugging Face Is Making A Big Robotics Push [Applied AI]

2026-09-09T16:23:41+00:00

Nvidia is paying a steep price for Hugging Face, a repository of open source models, primarily for its strategic position in the center of the open source AI world. But it turns out Nvidia is also getting a budding robot business that could help CEO Jensen Huang fulfill the promise of “physical AI.”

Hugging Face late last month launched Microduck, a small robot that can walk around and learn skills like roller skating or pointing lasers at walls for cats to chase. The company has already sold more than 15,000 ducks, generating over $6 million in sales, despite limited advertising on X and LinkedIn, Hugging Face co-founder Thomas Wolf told me.

The new robot is part of Hugging Face’s push to sell affordable robots built on open source AI for consumers and developers. That shaped its strategy with its April 2025 acquisition of Pollen Robotics, which previously focused on pricier humanoid robots but shifted to making more affordable, accessible robots after the deal, Wolf explained.

How to Use Google Gemini for Market Research and Campaign Planning [Partner Content]

2026-09-08T18:40:22+00:00

For marketers, the start of annual campaign planning is the ultimate blank page.

Even with access to comprehensive product information and results from past campaigns, it can be daunting to begin pulling this data together and finding ways to improve in the year ahead. Often, marketing leaders are overwhelmed with a combination of highly technical product roadmaps and dense spreadsheets detailing last year’s campaign metrics, and it can take days of manual cross-referencing and analysis to simply figure out where to start.

By synthesizing market research, product specs, and historical data, Gemini can help your team map out a data-informed go-to-market strategy.

Step 1: Sketch Out Your Baseline Buyer Personas

To start, you can use Gemini’s vast training data to help you build a detailed profile of your target buyers. You’ll need to give Gemini a rough directional target to make sure these personas are as specific as possible. Jot down your timeline, product category, and high-level audience definition, and then ask Gemini to use this information to help you brainstorm ways to reach these buyers.

Here’s a sample prompt:

I’m a marketing leader conducting analysis in preparation for next year’s [launch]. Define my target audiences [audiences], for my new line of [product]. Include interests, relevant marketing channels, and top trends that drive their consideration and purchase behavior.

Gemini will return a set of potential audience segments based on your inputs. For example, if you’re marketing a high-end technical jacket, results might include the Active Executive Commuter, the Global Business Traveler, and the Tech-Forward Urbanite, with different purchase drivers for each persona.

Step 2: Break Down Your Data Silos

Now that you have your buyer personas sketched out, it’s time to bring in additional information to give Gemini the context it needs to find historical lessons and apply them to your next campaign.

Gather past campaign performance data, corresponding product information, and your own insights on what worked and what went wrong. Sticking with the example of a premium jacket: Perhaps you will note that social media ads underperformed for last year’s line, with brand-aligned messaging about sustainable manufacturing failing to effectively drive purchase intent. Or, you might note that sponsored videos from travel gear creators drove the highest volume of sales.

Be as thorough and as specific as possible. The more information Gemini has about the past, the better it can help you plan for the future.

Step 3: Brainstorm Campaign Messaging

Now that Gemini has information about your product, your buyers, and your past campaigns, it can help you brainstorm ideas for how to conduct your next campaign.

Try this prompt:

Brainstorm value props for my [target audiences] based on features from [Product Requirements Document]. Include a section on campaign learnings from [Campaign Performance].

Gemini will suggest specific value propositions for each buyer persona that it previously outlined. For example, Gemini might note that the Active Executive Commuter is likely to focus on the jacket’s underarm vents, while the Global Business Traveler might care more about the jacket’s hidden pockets.

Use Gemini’s suggestions as a starting point for campaign messaging, rather than treating them like an unchangeable roadmap. While Gemini can spark ideas, your team’s expertise is needed to identify which of these ideas will work in the real world.

Step 4: Execute Your Campaign

After you’ve finalized your broad campaign messaging, you can come back to Gemini for help executing specific tactics. Here are some prompts you might try:

Turn these value propositions into a creative brief for our design team.

Draft three distinct variations of 15-second ad scripts showcasing [Feature X] for our target audience.

Outline a landing page wireframe that emphasizes these new value props.

Gemini’s creative brief will bring together information about your existing target audience and messaging angles, creative guardrails and format requirements, and key deliverables. Its ad scripts will offer ideas for specific visuals and voiceover text. And its landing page wireframe will outline a design meant to prioritize clarity and conversion, complete with ideas for headlines, hero image, and call to action.

Here again, you should tweak Gemini’s output, going back and forth in the chat to refine your vision as your campaign planning becomes more concrete. Human intuition will always be part of any successful campaign launch. But when you first use product data and past campaign information to generate initial ideas, you no longer have to start from scratch.

OpenAI CFO Touts User Growth Resulting From Lowering Prices [Briefing]

2026-09-09T09:40:22+00:00

OpenAI’s price cut on its GPT-5.6 Luna model shortly after its release in July resulted in a tenfold increase in model usage, said OpenAI chief financial officer Sarah Friar in an interview on Monday at Goldman Sachs’ Communacopia + Technology conference in San Francisco. Usage of the model is ...

AI Threats Are Reshaping Where Companies Spend Their Cybersecurity Budgets

2026-09-09T13:15:09+00:00

The AI hacking threat isn’t just causing companies to jack up their cybersecurity budgets, it’s changing where they spend them.

Buyers at big companies have significantly raised planned spending on systems that monitor how employees use AI, as well as on AI models that detect vulnerabilities before hackers can and AI-powered tools that speed patching those holes, security executives said.

How Six AWS Engineers Rebuilt Bedrock to Challenge Microsoft

2026-09-08T13:01:24+00:00

In early 2025, things were tense inside Amazon’s cloud unit. Bedrock, a new service designed to run AI models, was proving buggy. Customers were getting error messages repeatedly and sometimes waiting weeks for the computing capacity they needed, according to people with knowledge of the customer issues.

As these problems surged, a senior engineering leader, Anthony Liguori, led a team that came up with a plan for a revamped version of Bedrock—known as Project Mantle—designed to fix the throttling and error messages and enable the service to scale to serve larger numbers of users, said two of the people with direct knowledge of the situation. Liguori worked with five other senior engineers to build Mantle using AWS’ in-house AI coding tool, Kiro.