How does Trump’s White House AI accord work?

What the report says

U.S. President Donald Trump met with leaders from Meta, Nvidia, xAI, OpenAI, Google and Anthropic at the White House on Tuesday to promote a voluntary artificial intelligence pact, which the administration called the White House Accord on Super Intelligence. Al Jazeera reported that the agreement is intended to encourage companies to strengthen internal safeguards as AI systems become more powerful, while avoiding immediate government-imposed rules.

Under the accord, participating companies would keep internal controls, assign a team to oversee them, use outside auditors and create an independent board committee to review findings. The pact is not legally binding, sets no penalties for non-compliance and does not require public disclosure of audit results. It also leaves companies in control of how they apply the measures, although the text suggests the steps could later be turned into formal laws or regulations.

Trump described the approach as a form of self-policing and said it was morally binding. Google chief executive Sundar Pichai also supported the initiative publicly, saying it could help build confidence while preserving the benefits of the technology. The deal comes as some of the same companies have faced pressure to accept stricter oversight, and as lawmakers in Congress consider more binding proposals on frontier AI safety.

The broader policy debate centers on how to manage advanced systems often described as frontier AI or, in industry terms, superintelligence. Al Jazeera noted that Trump has also pushed federal agencies to use “Super Intelligence” terminology in official communications, reflecting a White House effort to shape the language and direction of AI policy.