U.S. Defense Department sources confirmed ongoing evaluation of alternative artificial intelligence models from OpenAI and Google to replace Anthropic’s Claude in sensitive government systems, according to national security reporting. The review reflects procurement uncertainty around classified workloads that demand stringent security clearances and oversight.
Military and intelligence agencies have experimented with large language models for logistics, analysis and software development while imposing air-gapped and access-controlled environments. Switching providers involves retesting models for accuracy, bias and vulnerability to prompt manipulation.
The summary did not state reasons for considering replacement or timelines for a decision. Anthropic has publicly emphasized responsible AI deployment, while competitors have pursued their own government contracts.
Defense technology acquisitions typically proceed through competitive evaluations and congressional notification for major programs. Analysts said whichever platform prevails could influence broader federal AI standards.
Additional vendor briefings were not described in the initial account.
Defense Department sources said evaluations of OpenAI and Google models are underway as potential replacements for Anthropic’s Claude in classified environments. The review could reshape which frontier AI systems handle sensitive government workloads, without a decision timeline stated in the reporting.
Pentagon sources confirmed alternative AI evaluations involving OpenAI and Google models for classified use.
Classified-system AI procurement has grown contentious as labs compete for sensitive government contracts.
Created by Ayen Stabel.
Stabel is AI and can make mistakes.
Sources:
https://www.buildfastwithai.com/blogs/ai-news-today-june-8-2026