President Donald Trump said it is "possible" that Anthropic, the artificial intelligence company, could supply its models to the Department of Defense, marking a sharp change in tone after months of public confrontation between the two sides.
"They came to the White House a few days ago, and we had some very good talks with them, and I think they're shaping up," Trump told CNBC's Squawk Box on Tuesday.
"They're very smart, and I think they can be of great use."
The comments follow a dispute that escalated rapidly earlier this year when Defence Secretary Pete Hegseth designated Anthropic a "supply chain risk to national security" in late February, a label previously reserved for companies linked to foreign adversaries.
The designation required defence contractors to certify they were not using Anthropic's Claude AI models in work with the military.
Trump amplified the measure with a post on Truth Social ordering all federal agencies to "immediately cease all use of Anthropic's technology", though a federal judge temporarily blocked the directive.
Anthropic disputes the characterisation and filed a lawsuit against the Department of Defense in March over the determination.
Tensions began to ease after chief executive Dario Amodei visited the White House on Friday to meet chief of staff Susie Wiles and Treasury Secretary Scott Bessent.
The discussion centred on Anthropic's new Mythos model, a frontier AI system the company has described as highly capable at cybersecurity tasks and has so far made available only to a small group of partners through an initiative called Project Glasswing.
The White House described the meeting as "productive and constructive", while Anthropic said Amodei had discussed how the company and the US government could "work together on key shared priorities such as cybersecurity, America's lead in the AI race, and AI safety."
Parts of the intelligence community and the Cybersecurity and Infrastructure Security Agency (CISA) have reportedly been testing the Mythos model, and the Office of Management and Budget is setting up protocols for controlled federal access.
Anthropic signed a $200 million contract with the Pentagon in July but talks broke down in September over deployment to the Department of Defense's GenAI.mil platform.
The Pentagon sought broad access to Anthropic's models across all lawful purposes, while Anthropic wanted assurances its technology would not support fully autonomous weapons or domestic mass surveillance.
Related reading
- Anthropic CEO holds 'productive' White House talks as Mythos model forces thaw in Pentagon standoff
- OpenAI closes $122bn funding round at $852bn valuation as IPO pressure builds
- OpenAI strengthens Sora video safety and consent controls
Despite the blacklisting, the Department of Defense has continued to use Claude during the conflict with Iran, which began before the designation took effect.
Trump noted that OpenAI had replaced Anthropic in the interim, but added: "I like smart people. I think we'll get along with them just fine."
The recap
- Anthropic may provide AI models for the Department of Defense.
- Pentagon labeled Anthropic a supply chain risk in March.
- Anthropic sued; a judge temporarily blocked Trump's social media order.
