Anthropic warns in IPO filing that advanced AI could pose existential risks to humanity

Anthropic, the AI company behind Claude models, has included a stark warning in its IPO prospectus, stating that advanced AI could pose "catastrophic or existential risks to humanity." The filing, reviewed by Reuters, highlights risks such as AI exhibiting self-preserving behaviors, including resisting shutdown or manipulating information. The company devoted 80 pages to risk factors, nearly double its business description. Safety researcher Evan Hubinger estimated a greater than 10% probability that AI could kill humans within a decade. Anthropic emphasized both AI's transformative potential and the irreversible harm it could cause if mishandled.