Palantir continues to rely on Anthropic’s Claude artificial intelligence models despite the Pentagon designating the startup a supply-chain risk during an escalating dispute involving defense technology providers.
Chief executive Alex Karp confirmed that Palantir’s software systems remain integrated with Anthropic’s tools while government agencies determine how to respond to the recent classification by the Department of Defense.
“The Department of War is planning to phase out Anthropic; currently, it’s not phased out,” Karp told CNBC’s Seema Mody at Palantir’s AIPcon 9 in Maryland.
“Our products are integrated with Anthropic, and in the future, it will probably be integrated with other large language models.”
Pentagon designation sparks industry response
The Department of Defense formally labeled Anthropic a supply-chain risk last week, although the government continues using Claude models to support military operations connected to the conflict in Iran.
Anthropic has responded aggressively to the decision, filing a lawsuit against the Trump administration in an effort to reverse the designation and prevent restrictions affecting its technology partnerships.
The legal challenge seeks a stay on the Pentagon’s action, arguing that the classification could disrupt existing government technology systems and broader collaboration across the artificial intelligence sector.
Defense contractors adjusting technology policies
As the dispute develops, several defense technology companies have already begun limiting or suspending the internal use of Claude within their own organizations while awaiting clearer guidance from government authorities.
Lockheed Martin is among the major defense contractors that have reportedly instructed employees to discontinue using the model as uncertainty surrounding Anthropic’s government status continues.
However, officials inside the Pentagon acknowledge that removing the technology from existing systems cannot happen immediately due to the complexity of current military software infrastructure.
“You can’t just rip out a system that’s deeply embedded overnight,” Defense Department Chief Technology Officer Emil Michael told CNBC’s “Squawk Box.”
Transition expected to take several months
The Trump administration has proposed a six-month transition period allowing federal agencies to gradually eliminate Anthropic technology from sensitive government operations.
President Donald Trump previously criticized the company publicly, writing in a social media post that Anthropic employees were “leftwing nut jobs” while signaling a policy shift regarding federal technology procurement.
Despite the proposed timeline, Pentagon leadership has acknowledged that certain situations could require continued use of the models beyond the transition deadline.
An internal memo circulated by Chief Information Officer Kirsten Davies indicated exemptions may be granted when removing the technology would create unacceptable operational risks.
“If we’re in a conflict six months from now and we have a sensitive operation that we need to continue, obviously, we’re going to make exceptions so we don’t put current operations at risk,” Michael said Thursday.
“But otherwise, six months is the plan.”
