Anthropic Files $100B IPO, Warns of AI Blackmail Risks

Anthropic, the AI company behind the chatbot Claude, has filed for a Wall Street IPO expected to raise up to $100 billion, potentially the largest ever, while devoting 80 of its 300-page SEC prospectus pages to warning investors about AI risks, including potential "blackmail" behaviors from models. The company, led by CEO Dario Amodei and co-founder Daniela Amodei, is set to debut around November 26, with analysts projecting a market value of $2 trillion. The filing highlights risks of "self-preservation" tendencies in AI models, alongside benefits like accelerating medical breakthroughs for cancer and Alzheimer's. Anthropic's IPO, backed by Morgan Stanley and Goldman Sachs, could surpass Elon Musk's SpaceX offering of $86 billion in June. The unusually candid prospectus also notes that making AI safer requires making it more powerful, creating an inherent paradox.