The Pentagon just canceled a $200 million contract with AI company Anthropic after they couldn’t agree on military control over the AI technology. The Department of Defense even labeled Anthropic a “supply-chain risk” when talks broke down.
The dispute centered on how much say the military would have over Anthropic’s AI models, including potential use in autonomous weapons and domestic surveillance. Anthropic pushed back on giving the Pentagon unrestricted access to their technology.
OpenAI Steps In, Users Step Out
When Anthropic walked away, the Pentagon quickly signed with OpenAI instead. OpenAI agreed to the military’s terms, but there was immediate backlash. ChatGPT uninstalls jumped 295% as users protested the deal.
This situation highlights a growing tension in Silicon Valley. AI companies need big contracts to fund their expensive research, but military deals come with strings attached. Some startups are discovering that federal money isn’t worth the controversy.
The Pentagon wants AI partners who won’t question how the technology gets used. Meanwhile, AI companies are realizing that saying yes to everything might cost them their regular customers. As more startups chase government contracts, they’ll face the same choice Anthropic did: take the money or keep control.