US defense contractors, including Lockheed Martin, are expected to remove Anthropic’s AI tools from their supply chains days after President Donald Trump ordered all federal agencies to stop using them immediately.
In the aftermath, Defense Secretary Pete Hegseth vowed to designate Anthropic a national security supply chain risk.
Read also | Trump banned Anthropic – hours later, the US military used its AI in airstrikes against Iran
The Trump Administration vs. Anthropic
“Effective immediately, no supplier, vendor, or partner that does business with the United States military may conduct any commercial activity with Anthropic,” Hegseth said in a post on X.
However, he added that Anthropic will continue to provide the War Department with its services for a period of six months for a smooth transition.
According to Hegseth, Anthropic’s position is fundamentally incompatible with American principles.
The Trump administration and Anthropic, a maker of artificial intelligence tools including Cloud, have been at odds over the company’s refusal to remove “safety rails” that prohibited their use in fully autonomous weapons systems and mass home surveillance.
What the CEO of Anthropic said
Anthropic CEO Dario Amodei denied the US demand, saying the company could not “in good conscience” agree to terms that would allow Cloud to be used for:
Anthropic argued that using artificial intelligence to track American citizens is incompatible with democratic values.
Anthropic also said its current AI models are not reliable enough to make lethal targeting decisions without human oversight, and that doing so would put civilians and warfighters at risk.
US used Anthropic for Iran war
Although the Trump administration banned the use of Anthropic, the Wall Street Journal reported that its artificial intelligence tools, including Cloud, were used by the US Central Command for intelligence assessment, target identification and simulation of battle scenarios before attacking Iran.
Claude was reportedly used through Palantir’s Maven Smart System to analyze intelligence, identify targets and simulate battle scenarios.
How defense contractors are responding
After the Trump administration banned the use of Anthropic, defense contractors reportedly began complying with the order, which legal experts say could be overturned by the courts.
“We will follow the instructions of the president and the Department of War,” Lockheed Martin said in a statement to Reuters.
We expect minimal impacts,” the company said, adding that it does not depend on any single AI vendor “for any part of our work.”
Read also | Anthropic promises legal action after flagging supply chain risk
With major government contracts at stake, defense contractors would quickly comply with the Pentagon’s ban, lawyers said.
“Most companies that do significant business with the government are very aware of what the US government wants and are probably already taking steps to clean up their supply chains from Anthropic,” Franklin Turner, a lawyer who specializes in government contracts, told Reuters.
“Regardless of the legal reasoning, I think this is a threat … it has already caused damage, significant damage to the company,” he added, referring to Anthropic.
Anthropic out OpenAI in
While Anthropic is standing its ground and has vowed to challenge the ban in court, one of its main rivals, OpenAI, has signed a deal to deploy its models on classified Pentagon networks.
Read also | In the midst of the “Abolish ChatGPT” trend, Anthropic brings a new feature to move to Claude
According to OpenAI CEO Sam Altman, their agreement includes similar bans on autonomous weapons and mass surveillance that Antropic has sought, though the administration has been more accommodating to OpenAI.
Key things
- The Trump administration’s ban reflects growing concerns about the role of AI in military and surveillance applications.
- Defense contractors prioritize compliance with government directives to secure lucrative contracts.
- Anthropic’s stance on the ethical use of artificial intelligence presents significant challenges in the evolving military technology landscape.
