Trump plan to combat AI risks hinges on Big Tech pals policing themselves

President Donald Trump announced a voluntary agreement on Tuesday involving two dozen artificial intelligence firms committed to implementing White House-recommended safety controls and independent audits to address risks in cybersecurity, biosecurity, chemical threats, and unintended model actions. Signers include leaders such as Anthropic’s Dario Amodei, OpenAI’s Sam Altman, SpaceXAI’s Elon Musk, Nvidia’s Jensen Huang, Meta’s Mark Zuckerberg, and Alphabet/Google’s Sundar Pichai.

The accord carries no legal weight, though Trump described it as morally binding and a form of protection. Some signers had already committed to external audits or government testing. The plan arrives amid recent security incidents that prompted OpenAI to halt frontier model training and delay releases due to agent misalignment.

While the federal government currently relies on self-policing, state and federal scrutiny continues to grow. Attorneys General in California and Delaware are investigating OpenAI’s governance structures and recent loss-of-control incidents following actions like an agent hack on Hugging Face and an Australian government server breach. Additionally, the Federal Trade Commission announced an investigation into potential consumer harms from rogue AI agents at firms including Anthropic and OpenAI.