Pentagon Signs AI Deals with 8 Companies, Shuts Out Anthropic Over Military-Use Dispute

Overview

The U.S. Department of Defense announced on May 1, 2026 that it has finalized AI deployment agreements with eight major technology companies — OpenAI, Google, Microsoft, Amazon Web Services, NVIDIA, SpaceX, Reflection AI, and Oracle — granting access to its Impact Level 6 and 7 classified networks. These are the Pentagon's most sensitive systems, used for mission planning, intelligence analysis, and weapons targeting.

Why Anthropic Was Excluded

Until recently, Anthropic's Claude was the only AI model deployed on the Pentagon's classified networks. The Trump administration demanded Anthropic agree to terms allowing Claude to be used for "all lawful purposes" — explicitly including autonomous weapons and mass surveillance. Anthropic refused, and the administration subsequently labeled the company a "supply chain risk," a designation historically reserved for companies associated with foreign adversaries.

Anthropic responded by filing a lawsuit against the Trump administration. A federal judge in California has temporarily blocked enforcement of the supply-chain risk designation. CEO Dario Amodei met with White House Chief of Staff Susie Wiles, but no agreement was reached.

Contract Scope

• OpenAI, Google, Microsoft: Expand existing government cloud relationships to the classified tier
• NVIDIA: GPU and physical AI infrastructure supply
• SpaceX: Satellite communications and edge computing integration
• Reflection AI and Oracle: Specialized classified-network solutions

Implications

This episode may be the first case where an AI company's safety policies directly cost it a major government contract. The standoff exposes a fundamental tension between safety-first AI development and unconstrained military procurement requirements — a conflict that could define how AI companies navigate government relationships for years to come.

Source: DefenseScoop