Anthropic Won’t Let Pentagon Turn AI Into Big Brother Armed with Lasers
KEY POINTS
- •Secretary of War Pete Hegseth held a January 29, 2026 White House meeting to discuss AI military contracts.
- •Anthropic CEO Dario Amodei refused Pentagon demands to allow fully autonomous lethal AI and mass surveillance.
- •Pentagon CTO Emil Michael threatened to label Anthropic a 'supply chain risk' if they didn’t comply.
On January 29, 2026, at the White House Cabinet Room, Secretary of War Pete Hegseth, flanked by President Trump and Commerce Secretary Howard Lutnick, tried to charm AI firm Anthropic into loosening AI controls for the Pentagon’s 'any lawful use' clause—even if that means full-on mass surveillance and fully autonomous lethal robots. CEO Dario Amodei respectfully declined, despite ‘threats’ including Pentagon CTO Emil Michael threatening a 'supply chain risk' label usually reserved for national security nightmares. Meanwhile, OpenAI and Elon Musk’s xAI quietly signed on, proving you really can make AI do scary military stuff if you just say yes politely.
Share the Story
Source: Theverge | Published: 2/27/2026 | Author: Stevie Bonifield