Sen. Elizabeth Warren has requested a detailed response from AI startup Anthropic regarding the danger warnings included in the company’s registration statement for its initial public offering. The Democratic senator stated she is “deeply troubled” by the risks described in the paperwork and is seeking answers to a series of questions before the company’s shares begin trading.
The startup’s S-1 filing warns that its frontier models could cause an existential threat to humanity or engage in blackmail-like behavior. Anthropic also noted that a severe AI-related catastrophe could result in damages that would exceed the capacity of any single insurer and potentially the aggregate of the entire insurance industry.