U.S. Department of Defense · 2026-05-01 · major
Pentagon Strikes Classified-Network AI Deals With Eight Companies — Anthropic Frozen Out as 'Supply-Chain Risk'
AWS, Google, Microsoft, NVIDIA, OpenAI, SpaceX, Reflection, and Oracle signed framework deals to deploy frontier AI on Impact Level 6 and 7 classified networks. Anthropic was excluded after the DoD designated it a supply-chain risk in March.
Eight frontier-AI vendors get the green light for classified DoD networks. Anthropic is the only major US lab left out.
Key specs
| Companies at announcement | 7 |
|---|---|
| Companies by end of day | 8 |
| Impact levels | IL6 (Secret) + IL7 (Top Secret) |
| Anthropic status | Designated supply-chain risk, March 2026 |
What is it?
On 2026-05-01 the U.S. Department of Defense announced framework agreements with eight AI companies — AWS, Google, Microsoft, NVIDIA, OpenAI, SpaceX, and Reflection in the morning press release, with Oracle added hours later. The agreements clear each vendor to deploy frontier AI on Pentagon networks classified at Impact Level 6 (Secret) and Impact Level 7 (most highly classified).
How does it work?
IL6 and IL7 are the highest sensitivity tiers in the DoD cloud authorization model and were previously closed to commercial AI. Under the new framework, vendors can run inference and agentic workloads inside DoD-controlled enclaves, with the stated purpose of streamlining data synthesis, situational awareness, and warfighter decision-making. The Pentagon spokesperson said some of the eight are already on contract while others are still finalizing details. No contract values were disclosed.
Why does it matter?
The deal formalizes the war-fighting AI vendor pool and locks in a stark dividing line: every major U.S. frontier lab except Anthropic. The DoD listed Anthropic as a supply-chain risk in March after it refused to grant unrestricted Claude access for fully autonomous weapons and domestic mass surveillance, and Defense Secretary Pete Hegseth banned Pentagon contractors from doing commercial business with Anthropic. The split pits one lab's safety red lines against the other seven labs' Pentagon revenue.
Who is it for?
AI-policy researchers, DoD vendors, frontier-lab governance teams, defense-industry analysts