The United States has developed strict new guidelines on artificial intelligence due to disagreements with Anthropic.

The Trump administration has established strict rules for civil artificial intelligence contracts, requiring AI companies to allow for "any lawful" use of their models, against the backdrop of a stalemate between the Pentagon and Anthropic. The draft guidelines seen by the Financial Times in the UK stipulate that AI companies wishing to do business with the government must grant the United States an irrevocable license to use their systems for all lawful purposes. This guidance issued by the General Services Administration will apply to civil contracts and is part of the government's efforts to strengthen the procurement of AI services. An informed source indicated that the principles in the guidelines are similar to measures being considered by the Pentagon for military contracts. These new terms have been drafted over the past few months, and just a week before, the US Department of Defense announced it would cancel a $200 million contract with Anthropic. The reason being that the AI company refused to give the Pentagon complete discretion on the use of the technology, citing concerns over domestic surveillance and deadly autonomous weapons. The White House has also identified Anthropic as a supply chain risk company. The startup, valued at $3.8 billion, argues that if allowed for "all lawful purposes," its powerful technology could be used for large-scale domestic surveillance and is pushing for specific clauses to provide safeguards.