Anthropic Phased Out of Federal Contracts as Pentagon Greenlights Grok for Classified Systems
The U.S. Department of Defense is actively replacing Anthropic's Claude with xAI's Grok across classified operations after Anthropic declined to provide unrestricted military access — a watershed moment that redraws the relationship between AI safety culture and state power.
The U.S. Department of Defense has begun phasing out Anthropic's Claude from federal systems and approved xAI's Grok for use in classified military operations, according to multiple reports circulating Sunday. The move follows Anthropic's refusal to grant the Pentagon unrestricted access to its models for military applications — a decision that appears to have triggered a hard contractual deadline and a rapid institutional pivot toward less restrictive alternatives.
As @paturious reported in what they described as a "SENTINEL UPDATE," Anthropic is now under a federal phase-out, with xAI's Grok explicitly permitted in classified environments. The timeline is unclear, but the framing suggests this is not a future contingency — it is already underway. The development comes amid escalating tensions with Iran, a context in which the Pentagon's appetite for AI capabilities with fewer guardrails has reportedly intensified.
Get our free daily newsletter
Get this article free — plus the lead story every day — delivered to your inbox.
Want every article and the full archive? Upgrade anytime.
No spam. Unsubscribe anytime.