Nobel and Turing winners sign statement calling Anthropic unsafe before IPO

What happened
A group of scientists, legal experts, and technology leaders released a public statement criticizing Anthropic as unsafe, one day before the company's shares were set to begin trading on the Nasdaq. The statement, organized by the Center for AI Safety, accused Anthropic of pursuing risky AI development and weakening its own safety policies while preparing for its IPO. The coalition included Nobel Prize and Turing Award winners, as well as other prominent figures from academia and the tech industry. Anthropic had not issued a public response to the statement by the following morning.
Key details
- The statement was published on Wednesday, the day before Anthropic's IPO.
- Signatories included Steven Pinker, Oliver Hart, Judea Pearl, Lawrence Lessig, Laurence Tribe, Peter Scholze, James Maynard, David Mumford, Thomas Wolf, Igor Babuschkin, George Hotz, and Max Tegmark.
- The coalition alleges Anthropic weakened its Responsible Scaling Policy in February, removing commitments to pause development.
- Anthropic is accused of lobbying against AI safety legislation in California and New York, and supporting federal pre-emption of state AI laws.
- Anthropic's IPO is described as the largest ever for an AI company, with shares to be listed on the Nasdaq.
Why it matters
Investors, regulators, and potential customers of Anthropic now face public criticism from a broad group of experts about the company's approach to AI safety. The statement could influence regulatory scrutiny, investor confidence, and public perception as Anthropic enters public markets. Lawmakers considering AI regulation and companies evaluating partnerships with Anthropic may need to weigh these safety concerns and the company's reported lobbying against certain AI safety laws.
