Pentagon inks deals with AI giants, but not Anthropic

The US Department of Defense on Friday signed agreements with seven artificial intelligence (AI) giants to deploy their powerful AI systems on its secure classified networks.

The deals notably left out Anthropic, the firm behind the popular AI assistant Claude, due to a dispute over safety restrictions on military use.

What do we know about the Pentagon deals?

The agreements include well-known firms such as OpenAI — the company behind ChatGPT—  Google, Microsoft, Amazon Web Services, NVIDIA, SpaceX, and Reflection — a startup making powerful open-source AI.

The goal is to expand access to advanced AI technology within the United States military for defense and security work.

A statement from the US Department of Defense said it would deploy the firms' AI capabilities for "lawful operational use."

The US government classifies sensitive information into three main levels: confidential, secret and top secret, with secret being the middle level.

The new AI systems will be integrated into the Pentagon’s secure classified networks, known as Impact Levels 6 and 7. These networks are approved to handle information up to the secret level.

This will give more defense and military personnel access to powerful AI tools while maintaining strict security controls. Unauthorized disclosure of secret information could seriously damage national security.

The Pentagon's main AI platform, GenAI.mil, has been used by more than 1.3 million Defense Department personnel in the five months since it launched, the department said.

The press statement added that GenAI.mil had cut "many tasks from months to days," noting that adding the other AI platforms "will give warfighters the tools they need to act with confidence and safeguard the nation against any threat."

Anthropic excluded from deals

The agreements notably exclude Anthropic, which was barred by the Pentagon and its contractors earlier this year after being labeled a supply-chain risk — a term the US government uses when it believes a company or product could pose a potential security threat to sensitive data.

Defense Department Chief Technology Officer Emil Michael told CNBC on Friday that Anthropic remains a supply chain risk.

However, he described the company’s Mythos AI model — known for its advanced cyber capabilities — as “a separate national security moment.”

How AI search is changing what we find online

To view this video please enable JavaScript, and consider upgrading to a web browser that supports HTML5 video

Anthropic CEO Dario Amodei met with senior Trump administration officials at the White House earlier this month to discuss the model, a meeting both sides described as productive.

US President Donald Trump said last week that Anthropic was "shaping up" in the eyes of his administration, raising the possibility that the company could eventually have its blacklisting reversed.

Edited by: Dmytro Hubenko