Story perspectives

Pentagon adds ChatGPT Mil and Grok to GenAI.mil

9/3/2026

Summary

DoD Adds ChatGPT Mil

The U.S. Department of Defense adopted OpenAI’s ChatGPT Mil and xAI’s Grok for Government.
Both models received Impact Level 5 clearance for handling Controlled Unclassified Information.
They are offered through GenAI.mil, the secure portal launched in December 2025 with Google Gemini.
GenAI.mil now hosts three IL5-approved models: Gemini, ChatGPT Mil, and Grok for Government.
Over 1.7 million of roughly 3 million DoD personnel have accessed the platform.

Summary

Judge Lin Blocks

Judge Rita Lin struck down the Pentagon’s supply-chain risk label on Anthropic on Aug. 27 2026.
Defense Secretary Pete Hegseth labeled Anthropic a risk in February after a U.S. President Donald Trump directive to halt Claude.
Anthropic kept Claude off GenAI.mil refusing to waive safeguards against mass surveillance and autonomous weapons.
The Defense Counterintelligence and Security Agency warned in June that ‘shadow AI’ tools could leak Pentagon data.

History

Historical Context

The Department of Defense launched Project Maven in April 2017 to embed machine-learning video analysis in intelligence workflows, initially funded with $70 million. After Google pulled out in 2018 following a walkout by more than 2,000 engineers, Congress pressed for tighter oversight of DoD AI contracts. That early clash foreshadows today’s GenAI.mil platform, which now runs ChatGPT Mil, Gemini and other models under new legal constraints.

1 of 3

DoD Adds ChatGPT Mil

The U.S. Department of Defense adopted OpenAI’s ChatGPT Mil and xAI’s Grok for Government.
Both models received Impact Level 5 clearance for handling Controlled Unclassified Information.
They are offered through GenAI.mil, the secure portal launched in December 2025 with Google Gemini.
GenAI.mil now hosts three IL5-approved models: Gemini, ChatGPT Mil, and Grok for Government.
Over 1.7 million of roughly 3 million DoD personnel have accessed the platform.

1 / 3