Anthropic is heading toward a reported $2 trillion IPO, and some of the loudest criticism is coming from people tied to the company itself.
That matters because the market has already voted with expectations. A public offering at that scale would turn Anthropic into a capital-markets story as much as a technology story, and investors generally do not pay premium multiples for uncertainty they cannot price.
One Anthropic employee, Evan Hubinger, who leads the company’s Alignment Science team, wrote in response to a colleague’s resignation post: 'We really do earnestly believe AI could kill all humans! I personally think it is >10% within the next decade. I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track to.'
Another researcher, Drake Thomas, said: 'I would burn my equity to the ground in a heartbeat for a 1% higher chance we make it out of this situation alive. I expect a great many of my colleagues across the industry would as well. I promise you, we are actually just fucking scared, it’s not galaxy brained marketing.'
The company was founded as an AI safety lab, and CEO Dario Amodei has put his own p(doom) at somewhere between 10% and 25%, according to the report. So the alarm from current and former researchers is not coming out of nowhere. It reflects a culture that has treated existential risk as part of the business from the start.
Still, the timing is awkward. The company is preparing for a public offering just weeks after a fresh round of internal warnings. A public company answers to shareholders expecting growth, which can pull management away from the voluntary slowdown that Amodei has floated. If researchers are also saying the company does not yet have a plan to control the risks, that is exactly the kind of language that can chill investor appetite.
SOC Investment Group, a labor-affiliated shareholder group that works with union-sponsored pension funds, called on Anthropic to delay the offering. It argued that the confidential IPO filing was submitted in June, before the recent hacking incidents, extinction warnings, and pacing proposals emerged, leaving investors unable to properly price those risks.
There is another side to that argument. Going public could bring more transparency and scrutiny, which might make the company safer rather than less safe. The company and its customers do not want misaligned AI that could cause havoc, and third-party evaluation and pacing commitments could become a selling point for enterprise buyers.
OpenAI has taken a different route. It has pumped the brakes on its own IPO, telling Fortune that AI safety concerns made it, in Sam Altman’s words, 'an ill-advised moment to go public' right now.
The numbers come first. Anthropic is aiming at a staggering valuation, but capital rewards clear rules, not unresolved risk. If a company’s own researchers are sounding alarms this loudly, the public-market premium may come with a warning label investors cannot ignore.
