Washington: US President Donald Trump has ordered federal agencies to ditch Anthropic’s AI technology following a public spat with the Pentagon over safety concerns. The US President’s order followed after Anthropic’s CEO Dario Amodei refused to allow unrestricted military use of its Artificial Intelligence (AI), citing concerns over autonomous weapons and mass surveillance. 

Trump gave the Pentagon 6 months to phase out Anthropic’s tech, saying, “We don’t need it, we don’t want it, and will not do business with them again!” The Trump administration’s decision benefits Elon Musk’s Grok chatbot and sends a warning to Google and OpenAI, which also have military contracts.

Following the decision, the US government is set to cease using Anthropic’s AI technology following a public disagreement with the Pentagon over artificial intelligence safety. President Trump announced on Friday that federal agencies will immediately stop using Anthropic’s products, with a 6-month phase-out period for the Department of Defense.

The dispute centres on Anthropic’s refusal to allow the Pentagon unrestricted use of its AI technology, citing concerns over autonomous weapons and mass surveillance. Dario Amodei stated that his company “cannot in good conscience accede” to the Pentagon’s demands. Meanwhile, Trump’s decision is likely to benefit competing AI companies, including Elon Musk’s Grok, which the Pentagon plans to use in classified military networks, while the move also puts pressure on Google and OpenAI, which have contracts to supply AI tools to the military.

The Pentagon had given Anthropic until 5.01 pm local time on Friday to agree to its demands, threatening to designate the company a supply chain risk if it refused. The designation could harm Anthropic’s business partnerships and reputation. Meanwhile, Anthropic’s refusal to compromise has garnered support from tech workers and industry leaders, including OpenAI CEO Sam Altman, who questioned the Pentagon’s threatening approach. 

According to the analysts, the standoff between Anthropic and the Pentagon represents a critical test of voluntary AI governance frameworks. The dispute centres on whether companies can enforce usage restrictions on their AI models once they are deployed in national security contexts. 

Anthropic asserted that its AI models, like Claude, are not reliable enough for autonomous lethal decision-making and should not be used for mass surveillance. The company has already softened its safety policy, removing a commitment to pause training if model capabilities outpaced safety controls.

However, the Pentagon insisted on full operational flexibility, stressing that vendors should not retain the ability to restrict lawful military applications. 

The industry leaders are divided on the issue, with some supporting Anthropic’s stance and others backing the Pentagon’s position. Elon Musk has sided with the Trump administration, while OpenAI and Google employees have expressed solidarity with Anthropic.

Retired Air Force General Jack Shanahan, who led the Pentagon’s AI initiatives, has questioned the wisdom of targeting Anthropic, arguing that the company’s red lines are reasonable. The standoff has further polarised the tech industry, with some companies and individuals supporting Anthropic’s stance on AI safety and ethics. Also, the outcome is expected to affect the development and deployment of AI in national security contexts.