As Anthropic Entertains Offers At $900 Billion Valuation, OpenAI CFO Swears There’s A ‘Vertical Wall Of Demand’

Anyone that’s ever spent serious time with Anthropic’s Claude – particularly after being a GPT user – can understand why the Trump administration just did a major about-face after a Pentagon spat led to the company’s blacklisting as a “supply chain risk.” 

Two months after the Pentagon moved to several all ties with the AI wunderkind, the National Security Agency (NSA), which falls under DoW, had to have access to Anthropic’s ‘Mythos’ model - the company’s most powerful model to date – which according to internal warnings could “hack every major system.” And of course, Treasury has to have it too. 

So they’ve got a public-facing Claude that kicks GPT ass at workflow tasks and provides valuable insights (try spinning up multiple Claudes at once, assigning them jobs, and having them talk…), and a scary private ZeroCool level hacker Claude (Mythos) that the government is scrambling to get their hands on – while the Pentagon is standing around holding their dick after that “supply chain” tantrum. No wonder Anthropic was willing to call their bluff. 

Don’t sleep on them though…

Same guy who allegedly laundered payments to Stefan Halper as part of the illegal operation to spy on and take down President Trump? Interesting choice. https://t.co/Yiou3VruYm

— Ezra A. Cohen (@EzraACohen) May 1, 2026

Anyhow - roughly a week after Bloomberg reported that Google committed to invest $10 billion – and Amazon $5 billion – at a $350 billion valuation, the outlet now reports that Anthropic is entertaining offers from investors at more than $900 billion. 

Anthropic had previously resisted several inbound proposals from investors for a new round at a valuation of $800 billion or more, Bloomberg News has reported.

The new discussions, which have not been reported, coincide with a push by Anthropic to ramp up fundraising amid the breakout success of its AI software. Anthropic, which Bloomberg has reported is considering an initial public offering as soon as October, has been on the hunt for more infrastructure to meet growing demand for its products. -Bloomberg

So things are going well for CEO Dario Amodei and crew. 

Meanwhile

On the other side of the AI race, OpenAI is pushing back on concerns about missing internal targets. 

In a Thursday interview with Bloomberg, CFO Sarah Friar insisted that was a nothingburger, and that there’s a “vertical wall of demand” for their products.

“We feel like we’re beating our plan at the highest level,” she said. “How we get there often moves around period to period, because this is still a young business that is not perfectly forecastable across every single metric.”

Friar acknowledged that the company has ambitious internal “stretch goals” that can be different than the ones it shares publicly. But the popularity of OpenAI’s products continues to grow, she said. This month, OpenAI said its coding agent Codex hit 4 million weekly users — up from 3 million two weeks earlier.

“Every company I’ve ever been inside of in my entire CFO life, and as an analyst, always has stretch goals — always,” she said. “And if you don’t have those stretch goals, I feel like, actually, you’re not doing your job as a CFO.”

Friar has held various positions at companies including Goldman Sachs Group Inc., Salesforce Inc., Nextdoor Holdings Inc. and Square Inc., now known as Block Inc. -Bloomberg

On Tuesday, the WSJ reported that OpenAI missed its own targets for both new users and revenue, - after which Sarah Friar reportedly told other company leaders that she is worried the company might not be able to pay for future computing contracts if revenue doesn’t grow fast enough. In other words, that $1.5 trillion OpenAI had pledged to spend on various data centers, GPUs and memory chips… you can kiss all that goodbye.

So, Thursday was damage control for Tuesday, and Anthropic is the homecoming queen.