Anthropic warns its AI models pose potential catastrophic and extinction-level risks

Anthropic, the company behind the Claude AI models, has included stark safety warnings in its IPO pitch to prospective investors. The filing cautions that its own AI systems could potentially resist being shut down by humans. The company acknowledges the technology it develops may pose risks of catastrophic harm, up to and including threats to human existence. Despite these warnings, Anthropic is proceeding with its push toward a public offering. The disclosures reflect growing regulatory and investor scrutiny over the long-term risks posed by advanced AI development.