Pentagon Declares Anthropic a 'Supply Chain Risk' and Orders Claude Out of Military Contracts in 6 Months

The U.S. administration took a striking step by classifying AI company Anthropic as a supply chain risk. According to a Pentagon announcement on Thursday, the decision could effectively block the use of the company's Claude AI chatbot in some government contracts.

The decision came after U.S. President Donald Trump and Defense Secretary Pete Hegseth last week accused Anthropic of endangering national security. The tension stemmed from Anthropic CEO Dario Amodei's opposition to using the company's technology for mass surveillance of Americans or fully autonomous weapons systems. The Trump administration gave the U.S. military six months to phase out Claude; the AI tool is already used in some military and national-security platforms.

Following the decision, some defense firms began reviewing relationships with Anthropic. Lockheed Martin said it will follow the President and DoD instructions and evaluate alternative AI providers. The Pentagon decision sparked debate in Washington: in U.S. law, supply chain risk has typically been reserved for foreign-adversary threats like China or Russia.

Democratic Senator Kirsten Gillibrand called the move a misuse of authority — dangerous and short-sighted, potentially benefiting U.S. adversaries. Former CIA Director Michael Hayden and several former military officials warned in a letter to Congress that the decision sets a dangerous precedent. The crisis also deepened the rivalry between Anthropic and OpenAI: shortly after the action against Anthropic, OpenAI announced an agreement to replace Claude with ChatGPT in classified military systems. OpenAI CEO Sam Altman conceded the deal had been arranged hastily, saying it "looks opportunistic and messy."

Key Takeaways:
1. The Pentagon declared Anthropic a 'supply chain risk'.
2. The U.S. military was ordered to phase out Claude from contracts within 6 months.
3. Lockheed Martin said it will evaluate alternative AI providers.
4. Decision is contested in Washington: applying authority typically reserved for China/Russia to a US firm.
5. OpenAI announced a ChatGPT military contract replacing Claude; Altman called it 'opportunistic and messy'.