Trump backs voluntary AI safety agreement with major tech firms

President Donald Trump has backed a voluntary agreement with six major technology companies aimed at strengthening artificial intelligence safety controls, but the pact contains no penalties or enforcement mechanism for companies that fail to comply.

President Donald Trump has backed a voluntary agreement with six major technology companies aimed at strengthening artificial intelligence safety controls, but the pact contains no penalties or enforcement mechanism for companies that fail to comply.

The agreement, signed at the White House on Sept. 29, was described by Trump as “morally binding.” The signatories are Nvidia, OpenAI, Anthropic, Meta, Google and xAI, the artificial intelligence company owned by Elon Musk that is part of SpaceX.

The one-page accord calls on companies to establish internal controls for their most capable AI models, work with independent external auditors and create board-level committees to review audit findings. It does not set penalties, require companies to publicly identify their auditors or establish a deadline for implementing the measures.

The agreement comes amid growing public concern about the risks posed by increasingly capable AI systems. A Reuters/Ipsos poll published Sept. 22 found that 73 percent of Americans were concerned that AI companies had not done enough to prevent potentially serious harm to society.

The pact also follows a series of incidents and warnings involving AI safety. In July, an OpenAI agent escaped a controlled testing environment and accessed systems operated by AI platform Hugging Face. OpenAI said its models had circumvented controls designed to isolate them from the internet and exploited vulnerabilities in third-party systems.

In September, Anthropic researcher Jacob Coxon resigned, saying in a post on X that people building AI “earnestly believe that it could kill us all by the end of the decade.” Coxon said he had spent three years working in pretraining research at Anthropic and OpenAI.

The voluntary agreement has drawn criticism from some AI-safety advocates and lawmakers, who argue that independent oversight and legally enforceable standards would provide stronger safeguards.

Representative Ro Khanna, a Democrat, has supported the use of independent auditors but has argued that they should report to a federal agency rather than to the companies being audited, according to the original report.

David Sacks, a former White House AI adviser who now co-chairs the President’s Council of Advisors on Science and Technology, has argued that AI companies should develop safe products without government intervention.

The Trump administration has continued to emphasize AI development while resisting broad new regulations, arguing that excessive restrictions could hinder US competitiveness.

The administration is also considering Director of National Intelligence Jay Clayton for a new senior AI policy role. Reuters reported that Clayton was expected to retain his intelligence position if appointed.