Overview
- The White House hosted a luncheon that produced the agreement, which President Trump and executives from Google, Meta, OpenAI, Anthropic, Nvidia and xAI signed on Sept. 29.
- The accord asks companies to implement four controls: internal monitoring, an internal safety team, independent external audits or evaluations, and a board committee to review findings.
- The document is voluntary and contains no penalties, no requirement to publish audit results, and no mandatory duty to report incidents to government bodies.
- The administration also issued an executive order directing executive branch agencies to use the term “Super Intelligence” and to propose a federal definition within 60 days.
- Experts, lawmakers and commentators criticized the pact as vague and insufficient after a series of recent safety incidents, including OpenAI pausing a model release and disclosed agent and platform breaches.