Trump Says AI Leaders Signed a Voluntary Accord for Internal and External Reviews

AI founders and investors welcome Trump’s self-policing approach at Tech Week

AI founders and venture capitalists at San Francisco Tech Week welcomed the Trump administration’s hands-off approach to AI safety this week, according to the Associated Press. Their enthusiasm follows a voluntary White House pact that asks advanced AI labs to monitor themselves, without detailed safety requirements or an explained federal process for checking implementation.

Trump hosted technology executives at the White House on September 29 for a luncheon and ceremonial signing. Representatives of OpenAI, Anthropic, Meta, Google, SpaceXAI and Nvidia signed the accord. The agreement puts those companies’ own safety arrangements at the center of the administration’s approach.

The 308-word document calls on advanced AI labs to follow safety protocols and establish internal teams to monitor their progress. It does not spell out detailed safety measures. Asked how the government would monitor implementation, the White House reiterated Trump’s position on balancing innovation and security without directly explaining a checking process.

Several signatories described their safety arrangements to AP. Their responses ranged from plans to bring in outside reviewers to changes in internal standards and board governance. Those are different ways of organizing scrutiny, rather than a single detailed testing requirement set out in the pact.

OpenAI said it is committed to embedding external assessors and finalizing contracts with third-party safety assessors. The company described these as commitments, not completed arrangements.

Anthropic said it is bringing in independent evaluators and discussing collaborations with industry groups. Its response described outside scrutiny alongside possible industry cooperation.

Meta pointed to updated standards for training, evaluating and releasing its most capable models, along with efforts to refine its board’s governance structure.

At a crowded Tech Week event on Wednesday, founders and investors said the government’s hands-off stance would mean higher profits for companies building AI-powered defense and security systems. Panelists praised the administration’s industry support at an investment-focused meeting organized as part of Andreessen Horowitz’s series of events.

Kyle Stanford, a venture capital research director at PitchBook, framed defense spending as an investment opportunity. He told attendees that the U.S. government was allocating $1 trillion to defense, including $13.5 billion for autonomous systems and AI, and said that spending had attracted unprecedented capital to defense technology.

This is no substitute for hard law, and it does give a lot of freedom to the companies.

Robinson, who recently resigned from OpenAI, questioned letting advanced labs police themselves. He said the time between model generations is measured in weeks and warned of serious near-term safety challenges. His concern contrasts with the profit-focused response among founders and investors at the Tech Week gathering.

Voluntary commitments are not new. Under President Joe Biden, leading AI companies made safety pledges that included specific guidelines for tests and transparency reports. Sarah Myers West, a former senior AI adviser to the Federal Trade Commission and now co-executive director of AI Now, told AP that the current accord lacks those specifics.

The administration has also canceled earlier requirements for AI developers to share safety-test results with the federal government. In December, Trump signed an executive order aimed at blocking state-level AI regulations, arguing that a patchwork of rules could stifle the industry. Both steps form the policy backdrop to the Tech Week enthusiasm.

Graham Steele, a former assistant secretary at the Treasury Department, raised a separate concern about who controls evaluations. He warned that internal teams or hired outside evaluators can face undue pressure. He also questioned whether companies should hold decision-making power over matters with broad social and political consequences.

Update from @sundarpichai

Great to meet today with @POTUS, @JDVance, @SpeakerJohnson and Administration + tech leaders. Important conversation and we signed today the White House Accord on Super Intelligence. As I shared today, Google has invested hundreds of billions in the last two years alone, with more to come, across the entire stack, to deliver benefits for America and the world. We’re working to build products that deliver real value for people and businesses, invest in local communities, and build trust in the technology. Industry also has to innovate responsibly. Google’s focused on building the right way, with appropriate testing, evaluations, red-teaming, and other safeguards against misuse and...