The MisAnthropic Statement is a public criticism of Anthropic’s approach to advanced AI, issued as the company prepares for a public offering. It argues that Anthropic warns about the dangers of recursively self-improving AI - especially the risk that it could concentrate power and evade human control - while competing to build such systems first. The statement says the company’s safety reputation helps attract investors, employees, and public support, but conflicts with its actions. It also accuses Anthropic of quietly relaxing safety commitments and lobbying for regulation that serves its own interests.
The signatories conclude that Anthropic should not be regarded as a safe AI company and call its pursuit of power under the banner of public benefit reckless. The page invites others to endorse the statement and lists a wide-ranging coalition of supporters, including academics, technology founders, AI safety advocates, writers, and public figures. Among them are Harvard professor Steven Pinker, Hugging Face co-founder Thomas Wolf, Nobel laureate Oliver Hart, AI researcher Gary Marcus, and Future of Life Institute leaders Max Tegmark and Anthony Aguirre. The statement’s central argument is that safety claims should be judged against a company’s competitive strategy and political influence, not accepted as proof that its development of powerful AI is responsible.
Summary generated by AI from the linked article. hn.today is not affiliated with Hacker News or Y Combinator.