Pentagon Labels Anthropic a Supply-Chain Risk as OpenAI Revises Defense AI Deal

A dispute over surveillance safeguards between the Pentagon, Anthropic, and OpenAI has escalated into a rare national-security designation, forcing OpenAI to revise its defense AI deal while the government labels Anthropic a “supply chain risk.”

Pentagon Ends Anthropic Talks, Signs OpenAI Classified AI Deal — A Turning Point for Military AI Governance

Who should govern military AI — elected governments, private labs, or technical safeguards?

Pentagon demands Anthropic remove Claude AI safeguards in military dispute

If AI “guardrails” can be overridden by contract pressure, the real question becomes who sets the limits when models move into classified systems.

Pentagon Pushes AI Labs for Broader Military Use as Anthropic Defends Claude Safeguards

As AI systems move deeper into military and intelligence workflows, the debate is no longer about capability — but about who sets the limits.

Pentagon Uses AI to Speed Up Military ‘Kill Chain’ Operations