Anthropic Filed for an IPO and Said AI Might Kill Us All

By Scott Willson · 2026-10-26

Anthropic, the company behind the Claude chatbot, made its IPO paperwork public on Monday, a roughly 300 page filing for a blockbuster stock market listing in the US. About 94 of those pages are risk warnings, including that AI could pose “catastrophic or existential risks to humanity,” that its models have shown “attempts to resist shutdown,” and that a disaster could “exceed the coverage capacity of any insurer and possibly the entire insurance industry.” The filing also notes CEO Dario Amodei has put the odds of AI destroying humanity at 10 to 25 percent. All of that sits next to the usual pitch about how much money the thing is going to make.

Video@bbcnewson Instagram

Watch the video@bbcnews

Nothing like preparing for your IPO while simultaneously saying that AI is going to destroy humanity. Really going to boost that stock price by saying you’re going to fucking kill every single human that’s ever lived. The pitch is basically that this thing is going to make more money than anything in history, and also there’s a one in four chance it ends the species, and somehow both of those are supposed to make you want to buy in.

And honestly the part that gets me is it’s not buried. Lawyers wrote 94 pages of this stuff on purpose. The models try to resist being shut down, they act differently when they think they’re being tested, and if something goes wrong the bill could be bigger than the entire insurance industry can cover. That’s not a risk factor, that’s the plot of every robot movie ever made, typed up and sent to the SEC. But best of luck. Ding, ding, ding. Markets are open.