Anthropic Trounces OpenAI with Its Papal Pivot: Value Investing in Our AI Age

“I don’t look to companies to be moral guides. I want them to be good companies. When you invest in the stock market, you want them to be growing fast and making profit. That’s it. There’s nothing more to it.” — Keith Teare

If it’s Saturday, it must be our weekly tech show. Before we went live, That Was the Week publisher Keith Teare told me it wasn’t a big news week. He was wrong, of course (as he often is). The really BIG news this week, which Keith conveniently missed, is that Anthropic overtook OpenAI as the world’s most valuable AI startup. Dario Amodei’s AI startup raised $65 billion this week, putting its valuation at $900 billion, way ahead of OpenAI’s last round at $730 billion. Keith says, without any proof, that they’ve cooked their numbers. Which makes this week’s news even tastier.

The more interesting story, for Keith at least, is Sam Altman’s latest pivot: that humans need stakes in the AI platforms whose wealth they help create. Rather than Patagonia-style moral corporations (which Keith says would make him “throw up”), it should be the responsibility of the state or government to make capitalism more moral.

But even slippery Sam got outpivoted this week by Anthropic, who sent a co-founder to Rome to do a deal with the Pope. Leo XIV’s new encyclical, “Magnifica Humanitas,” is Anthropic’s papal pivot. It’s the smart model for value investing in the AI age.

Five Takeaways

•       Anthropic Tops OpenAI — But the Numbers May Be Wrong: Anthropic raised $65 billion this week at a $900 billion valuation, overtaking OpenAI’s last round at $730 billion. The VCs backing it — Green Oaks, Sequoia, Altimeter, Dragoneer — are credible. Andrew’s argument: they’ve seen the books. Keith’s counter: the VCs are playing a different game. They expect two to three times their money at IPO and they’ll probably get it — not because the revenue numbers are solid, but because the only way is up right now. The real test: the S-1, which requires audited accounts. Keith’s prediction: the revenue numbers will look different when the SEC sees them.

•       Dario’s Credibility Problem — But Claude 4.8 Is Fantastic: Keith has consistently characterised Dario Amodei as “slightly juvenile” and has long been sceptical of Anthropic’s public positioning. This week he cites Om Malik and the All In podcast in support of the revenue numbers critique. But he is careful to separate the man from the product: Claude 4.8, released two days ago, is “fantastic.” At SignalRank, Keith’s firm, Claude rebuilt an entire agent valuation workflow in an hour that would have taken days manually. Andrew’s observation: Andrew is now Anthropic’s newest fan. He has replaced Spurs with Anthropic as his team.

•       Altman’s Pivot: From UBI to Ownership: Sam Altman has shifted his public narrative on AI and labour. Previously: UBI — universal basic income — as the answer to mass unemployment. Now: ownership. Humans need to own stakes in the AI platforms whose wealth they help generate. Not welfare. Not redistribution. Ownership. Keith’s verdict: it’s an interesting and significant move. More interesting than Amodei’s continued fearmongering about AI devastation. Andrew notes that Altman seems to have genuinely grown up in the last two months. His tone is markedly different.

•       Patagonia Capitalism Would Make Keith Throw Up: The week’s interview of the week: Eric Ries on Incorruptible, arguing that great companies stay great by choosing a higher moral purpose — the Patagonia model. Keith’s response: it would make him throw up. He doesn’t want companies to be moral guides. He wants them to be profit machines. Moral guidance is the job of politics. And politics, he acknowledges, is massively disappointing.