OpenAI, Nvidia, Alphabet Sign Pentagon AI Deals—Anthropic Left Out
Pentagon signed deals with 7 tech firms for classified military AI use, excluding Anthropic over surveillance disputes. The U.S. military aims to become an
- The Pentagon signed classified AI agreements with seven major tech firms to deploy AI on its highest-security IL6 and IL7 military networks.
- Anthropic was excluded after refusing to loosen “red lines” around autonomous weapons and mass domestic surveillance, and later sued the Trump administration over what it called unlawful retaliation.
- Google’s deal faced internal opposition from roughly 600 employees who sent a letter to CEO Sundar Pichai asking him to reject Pentagon AI partnerships.
The Pentagon announced agreements with OpenAI, Alphabet, Nvidia, SpaceX, Microsoft, Amazon, and Reflection to deploy their AI tools on classified military networks for “lawful operational use.” The deals aim to “accelerate the transformation toward establishing the United States military as an AI-first fighting force,” the department said.
Anthropic is conspicuously absent. The AI startup had a $200 million contract to handle classified materials but refused to loosen its restrictions on mass domestic surveillance and fully autonomous weapons. The Trump administration responded by labeling the company a national security threat.
Anthropic sued the administration in March, accusing the government of an “unlawful campaign of retaliation.” The company later won a court order blocking the AI ban—but still isn’t part of the new classified agreements.
The deals integrate AI capabilities into the Pentagon’s Impact Level 6 (IL6) and Impact Level 7 (IL7) network environments—the highest classification levels for military systems. According to the department, the AI will streamline data synthesis, improve situational awareness, and augment decision-making in combat operations.
OpenAI signed its deal with the Pentagon shortly after Anthropic was pushed out. The agreement was later amended to prohibit the use of OpenAI tools for domestic surveillance of U.S. citizens, following public criticism.
Google’s deal, reached earlier this week, allows Gemini AI systems onto classified networks. It hasn’t been smooth internally: roughly 600 Google employees sent a letter to CEO Sundar Pichai asking him to reject future Pentagon AI partnerships. The company signed anyway.
“The deal ‘to use Google’s AI models for classified purposes illustrates the growing importance of AI for U.S. national security,'” said Michael Horowitz, a former senior defense official and professor at the University of Pennsylvania.
Defense Secretary Pete Hegseth has made AI integration a stated priority, describing his goal as transforming the U.S. military into an “AI-first warfighting force.” The Pentagon’s GenAI.mil platform is already operational, according to the department’s announcement.
Anthropic co-founder Jack Clark said earlier this month that the company was in talks with the federal government about Claude Mythos, its new AI model the company claims can exploit security vulnerabilities in every major operating system and web browser. CEO Dario Amodei visited the White House in April following the Pentagon model ban, though the ethical standoff has not been resolved.
The Pentagon previously signed agreements worth up to $200 million each with major AI labs in 2025, including Anthropic, OpenAI, and Google. Anthropic’s current exclusion represents a reversal of that relationship—one the company is still fighting in court.
AI Ethics and Military Applications
The Pentagon’s AI deals raise significant ethical questions about the role of artificial intelligence in warfare. The agreements allow AI systems to be used for “lawful operational use,” which includes combat operations, intelligence analysis, and decision support. Critics argue that this could lead to autonomous weapons systems that make life-or-death decisions without human intervention.
Anthropic’s refusal to participate highlights the growing divide within the AI industry over military applications. Some companies see defense contracts as a major revenue opportunity and a way to contribute to national security. Others worry that military AI could lead to dangerous escalation and undermine international arms control efforts.
The internal opposition at Google reflects broader concerns among tech workers about their companies’ involvement in military projects. Similar protests have occurred at other tech companies, including Microsoft and Amazon, over contracts with defense agencies. The protests have had mixed success, with some companies canceling controversial projects while others have proceeded despite employee opposition.
The Pentagon’s push for AI integration comes as part of a broader modernization effort. The military is investing heavily in autonomous systems, cyber capabilities, and advanced analytics. AI is seen as a critical technology for maintaining military superiority against near-peer competitors like China and Russia.
