The US Department of Defense has stopped using Anthropic's AI tools entirely, a department official told the BBC, months after Defense Secretary Pete Hegseth designated the company a supply chain risk and ordered its technology phased out of the military.
"The Pentagon has ceased the use of Anthropic products," the department official said in a statement to the BBC on Monday. The confirmation ends a drawn-out transition in which Claude remained embedded in some of the military's most sensitive workflows long after the official deadline for removal passed. For more context on this story, see our ongoing more AI stories.
Claude was still in use last week
Multiple people familiar with the matter told the BBC that as recently as last week, Anthropic's Claude model was still being used by the department for research, analysis and intelligence gathering, as well as in military operations against Iran. The model was viewed as a critical tool for sorting through vast quantities of information, according to former US defense officials and contractors who spoke to the BBC, several of them on condition of anonymity.
Claude was embedded in the Pentagon's Maven Smart System, the primary platform for organizing intelligence and other data, which is operated by Palantir. Analysts used Maven to access and organize visual data including satellite imagery and drone footage, which was then fed into Claude and other large language models for tasks such as identifying potential military targets and compiling one-page briefs, the BBC reported.
Lauren Kahn, a senior research analyst at Georgetown University's Center for Security and Emerging Technology and a former US defense official, said Claude's deep integration into Maven made it difficult to remove quickly. "The fact that it took the Pentagon until now to remove all use of Claude from across the system is indicative that these things are not just plug and play," she told the BBC. "Once they become integrated it can be painful to remove them."
How Anthropic ended up blacklisted
Anthropic had supplied AI tools to the US government and military since 2024 and was the first advanced AI company to have its systems deployed in agencies doing classified work, according to the BBC. That relationship collapsed earlier this year when the Pentagon pressed the company to remove safety guardrails from Claude and give the military unfettered access to its products.
When Anthropic refused, citing concerns about mass surveillance and autonomous weapons, Hegseth designated the company a national security supply chain risk in February — a designation typically reserved for companies based in countries viewed as hostile to the United States. Hegseth set a six-month deadline to phase out Anthropic on February 27, targeting full removal by late August. The BBC noted it is not clear why the removal took longer than ordered.
Anthropic called the move "unprecedented and unlawful" and sued the Trump administration to overturn the designation. Anthropic declined to comment on the Pentagon's statement, the BBC reported.
A split court record
The legal fight has produced contradictory rulings. In August, US District Judge Rita Lin in San Francisco struck down one of the two designations, finding it amounted to unconstitutional retaliation for Anthropic's public criticism of the administration. In September, the DC Circuit Court of Appeals upheld the second designation in a 2-1 ruling, finding the Defense Department had acted within its authority, with the majority writing that continued integration of Claude into department systems presented a national-security risk, according to CNBC.
The Pentagon's confirmation that it has now stopped using Anthropic products entirely means the practical question — whether the military has access to Claude — has been settled by the department itself, regardless of how the litigation ultimately resolves.
Google, xAI and OpenAI fill the gap
As the dispute played out in court, the Pentagon signed contracts with other major technology companies, including Google, xAI and OpenAI, the BBC reported. OpenAI's tools have been more widely adopted in recent months in some military departments, according to two people familiar with AI use inside the defense department.
The episode highlights a growing fault line in the relationship between frontier AI companies and national security establishments. Anthropic has positioned itself as the safety-focused lab willing to walk away from lucrative government contracts over usage conditions, while competitors have accepted fewer restrictions in exchange for defense work.
The court cases are not over. Anthropic's lawsuits continue to work through the appeals process, and the company retains other paths to challenge the designation. But with the Pentagon publicly confirming a completed exit, the practical stakes of the litigation have narrowed: even a favorable ruling would not automatically restore Claude to Pentagon systems, and rebuilding the deep Maven integration that took years to establish could take years more.
For the Pentagon, the transition demonstrates both the strategic value of frontier AI in intelligence work and the operational cost of untangling it once embedded. For Anthropic, losing the defense market removes a revenue stream but preserves the usage red lines the company says are central to its mission.
---
Stay Ahead of AIGet the latest AI news, analysis, and breakthroughs — all in one place.
Read more AI news →
