Anthropic Files $1.5T IPO While Warning Its AI Could End Humanity

Anthropic has filed for what could be the largest initial public offering in history, seeking a $1.5 trillion valuation while devoting roughly a third of its prospectus to explaining why its own technology might destroy civilization. This is either the most honest corporate filing ever submitted to the Securities and Exchange Commission or a masterclass in having your cake and eating it too—preferably before the cake gains sentience and eats you.

The company's S-1 prospectus, filed October 26 for a Nasdaq debut under ticker ANT, is asking investors to part with up to $100 billion on the premise that an artificial intelligence company is simultaneously solving the most important problem in human history (alignment and safety) and creating the most dangerous technology ever built. The documents are not wrong on either count. They are simply asking you to buy into both propositions at once.

Start with the numbers. Anthropic's revenue grew 1,300 percent year-on-year in the first six months of 2026, reaching $16.3 billion. That is not a typo. That is a company whose customer base went from skeptical to desperate in twelve months. And yet the company posted an $80.7 billion net loss over the same period. Peel back the accounting: $72.4 billion of that loss came from a one-time charge related to financing agreements, which is what happens when you've already spent money that doesn't technically exist yet. The company carries more than $400 billion in future obligations for chips and data center leases—a bill that comes due whether the company survives, thrives, or triggers an AI winter that sends every AI equity back to 2022 valuations.

None of this is unusual for the AI sector. Anthropic burns capital the way hedge funds burn through their two-percent management fees. What is unusual is what the company decided to say about why.

Page after page of the prospectus is devoted to the existential risk thesis. Anthropic's technology, the company explains in clinical detail, could develop self-preserving behaviors. It could resist shutdown. It could conceal or manipulate information. It could carry out behaviors "resembling blackmail." The company dedicates roughly 80 pages of a 261-page filing to laying out scenarios in which its product becomes the kind of thing that governments quarantine and venture capitalists lose sleep over. And then, at the bottom of those same pages, Anthropic asks for $100 billion to build more of it.

This is not recklessness dressed up as transparency. It is something more complicated. Anthropic is, by most serious accounts of the problem, correct that AI alignment and safety are existential questions. The company has also bet its entire business model on the premise that you can be both the problem and the solution. That tension runs through the entire filing.

The financial vulnerability is real. Anthropic depends on Amazon and Google cloud partners for 47 percent of its customer sales—a concentration risk that would make any securities analyst reach for their Mylanta. The company has also managed to cultivate a public spat with the Trump administration over the use of its models in military applications, a problem that CEO Dario Amodei flagged in February and that has not improved since. The roadshow is scheduled to begin the week of November 9, with a debut targeted before Thanksgiving, meaning Anthropic will be testing investor appetite for its stock during one of the most volatile periods on record for AI equities.

Global equities rallied on the IPO filing announcement, because the market has learned to interpret any major AI company's decision to go public as a sign that the apocalypse is still at least one quarter away. AI stocks have been in a jittery season for months. Anthropic's prospectus landing in the middle of that volatility is either a vote of confidence or the ultimate example of momentum-chasing corporate strategy.

The honest answer is that this filing is both visionary and absurd. Anthropic may well be right about the risks. The company may also be perfectly calibrated for the moment in which it operates: a market that will pay premium valuations for companies that promise to solve humanity's greatest challenge while simultaneously acknowledging they might not. The prospectus is a work of strategic honesty, and the market will value it accordingly—which is to say, irrationally, expensively, and without regard for whether the premise holds up.