By various reports, Anthropic’s IPO is being delayed until mid October or later.
The delay comes amidst a firestorm, if not hysteria, set off by Anthropic’s Jacob Coxon, who just quit to call attention to what he considers the threat of human extinction posed by fast developing and ever more powerful AI models.
To fan the flames of concern, Evan Hubinger, Anthropic’s own Lead for Alignment Science, agreed with Coxon but went further:
Former AI Czar David Sacks even said Anthropic must delay its IPO to sort out Coxon’s comments.
Anthropic also just issued a safety report, detailing disturbing threats and incidents that Anthropic detected and prevented:
Does anthropic need to amend its form S-1 filed with Sec?
Every IPO requires the filing a Form S-1 with the SEC. In that Form, a company must disclose risk factors to the securities being offered to the public, see below. A 10% chance that AI will lead to human extinction within the next decade, according to the company’s own Lead for Alignment Science seems to make any investment in Anthropic “speculative or risky,” does it not?
As far as I can tell, Cerebras Systems, which had an IPO earlier this year, did not list human extinction caused by AI as one of the risks in its Form S-1.
The SEC is apparently still reviewing Anthropic’s filing. It will be interesting to see if this issue is flagged. Regardless, such a disclosure could be helpful to stave off future securities lawsuits.