Breaking

Trump and AI executives sign a voluntary White House safety accord

September 29, 2026 at 18:11 EDT

Policy & Regulation
AI Agents

President Donald Trump and leaders of major U.S. artificial intelligence companies signed a voluntary safety accord at the White House on Tuesday, committing firms to internal controls, outside review, and board oversight while stopping short of new binding federal rules. Trump called the one-page document “morally binding” and said he was seeing “tremendous self-policing,” after a lunch that brought rival executives into the East Room as concern over runaway AI systems has grown.

Sept. 29, 2026 · White House

A voluntary safety accord, not a new statute

President Trump and rival AI executives signed a one-page pledge of internal controls, outside review, and board oversight. He called it morally binding.

Four control layers, stacked

layers on one page

1–2. Internal cybersecurity controls, plus a team that checks monitoring.

3. An outside operator assesses the models.

4. An independent board committee gets the reports.

The height of the stack is the pledge: four checks, no penalty clause attached.

company signatures on the posted copy

people on a committee Trump said was discussed

3–4

days he gave for naming an AI czar

How the pledge is supposed to work

1 Internal controls

→

2 Outside audit

→

3 Board review

→

4 Shared standards

At the table

Jensen Huang said there is no conflict between innovation and safety. Attendees described the session as a push for common, auditable standards.

In Congress

Senator Mark Warner said the most powerful models need mandatory testing, evaluation, and incident reporting, not only a rename and self-regulation.

Still voluntary

No new binding federal safety law was announced. Enforcement stays with the companies, while a 10-person committee and an AI czar remain plans, not appointments.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.