NEW YORK, March 9, 2026, 13:56 EDT
Anthropic reignited questions over Palantir Technologies (PLTR.O) on Monday, filing suit against the Pentagon just as the company’s Maven military platform—developed partly on Anthropic’s Claude model—was already under pressure for a redesign after Washington blacklisted the AI firm. Last week, Reuters noted that Palantir’s Maven Smart Systems incorporates prompts and workflows powered by Claude.
Palantir has a lot riding on this. Its Maven-linked contracts with the Defense Department and national-security agencies could be worth upwards of $1 billion, according to Reuters, making this Pentagon initiative a major battleground.
The Pentagon’s top artificial-intelligence initiative, Maven, is designed to gather information from a range of sources, identify military targets, and accelerate both intelligence analysis and targeting. According to Reuters, Maven has factored into recent U.S. military efforts. Delayed data from Reuters put Palantir shares at $155.16, down 1.28% on Monday.
Anthropic went to federal court in California on Monday, seeking to overturn the Pentagon’s “supply-chain risk” label and stop agencies from acting on it. The company argued the designation breaks the law and infringes on its free-speech and due-process rights. Still, Anthropic noted that the lawsuit leaves the door open for renewed negotiations with Washington. Reuters
Palantir’s situation comes on the heels of a straightforward snag: According to a March 4 Reuters report, the firm needs to pull Claude out of Maven and redo sections of the platform. Defense Secretary Pete Hegseth called for the switch “immediate.” Reuters
Palantir CEO Alex Karp didn’t mince words at last week’s defense-tech conference. He took aim at Silicon Valley firms, saying their talk of AI killing off white-collar jobs while also “screw[ing] the military” risks “the nationalization of our technology.” Reuters
The fight is shifting Palantir’s landscape too. After the Anthropic controversy, OpenAI, backed by Microsoft, announced plans to bring its tech onto the Defense Department’s network. Reuters also reported that firms like Lockheed Martin were set to remove Anthropic systems from their military contracts.
Last week, Rosenblatt’s John McPeake summed up the market’s focus in a note: “War regrettably underscores the value of Palantir over just another LLM,” he wrote, pointing to large language models—the backbone for chatbots and code assistants—as the comparison. MarketWatch
Investors are eyeing Palantir after the company’s February update. U.S. government revenue shot up 66% in the fourth quarter, hitting $570 million. Total sales landed at $1.41 billion. For 2026, Palantir is projecting revenue between $7.18 billion and $7.20 billion.
The setup remains murky. Reuters wasn’t able to verify how long a Claude replacement might actually require, and attorneys noted Anthropic’s ban could get tossed in court. Palantir, meanwhile, was changing hands at 140.5 times forward earnings after its steep rally—hardly forgiving if timelines slip or if government demand softens.
