Anthropic Warns AI Risks Amid Industry Divide

Anthropic and other Silicon Valley firms are publicly clashing over how to develop and regulate AI, splitting between 'accelerationists' who prioritize rapid capability gains and 'doomers' who call for radical constraints. Anthropic CEO Dario Amodei warns of existential risks—including model misalignment, bioengineered threats, and authoritarian surveillance—and has launched a super PAC and resisted Pentagon requests, while rivals like OpenAI pursue faster deployment.