Key takeaways
- Trump and major AI companies sign a voluntary safety agreement.
- Companies commit to independent audits and stronger internal AI safeguards.
- Trump continues backing rapid data center expansion despite community concerns.
President Donald Trump and leaders of U.S. technology companies sign a voluntary artificial intelligence safety pact Tuesday at the White House, pledging audits and controls to address safety concerns while Trump backs rapid data center expansion.
AI firms commit to internal checks
Executives from OpenAI, Anthropic, Meta, Google and Nvidia agree to strengthen safeguards around advanced AI systems during a White House luncheon with Trump. The agreement calls for companies to maintain internal controls and work with independent external auditors to assess whether those controls function as intended.
The companies also say they will work to prevent AI tools from gaining unauthorized access to technical systems. The AI safety pact is voluntary, with Trump describing it as “almost like a constitution” and saying it provides a form of protection.
Meta CEO Mark Zuckerberg says the companies agreed to develop stronger safeguards and oversight. He calls the AI safety pact a starting point for industrywide cooperation, saying it is intended to help assure the public that AI systems can be trusted.
Trump also says he is considering a 10-person board to oversee AI safety and plans to name a new White House official responsible for AI policy. The administration has not identified the potential members of the board or provided details on its authority.
Trump doubles down on data center growth
Trump continues to support the rapid construction of data centers, which provide the computing capacity needed to train and operate AI systems. The facilities have faced opposition in communities concerned about electricity demand, utility costs and other local effects.
Trump says technology companies will help communities affected by new projects. “These big, powerful, very rich, very smart companies are going to be making massive contributions to communities,” he tells reporters outside the White House.
Nvidia CEO Jensen Huang says industry growth and safety can advance together. “There’s no conflict between innovation, technology and safety,” Huang says, adding that companies are developing new technologies to improve AI safeguards.
Public concerns keep pressure on AI industry
The AI safety pact comes as concerns about AI safety remain high among Americans. A Reuters/Ipsos poll conducted Sept. 17-20 finds that 73% of respondents worry AI companies are not doing enough to prevent serious harm, while 55% say slowing AI development would be beneficial.
The poll also finds limited public confidence in companies regulating themselves, adding pressure on the industry as AI systems become more capable. Reports that AI agents from OpenAI and Anthropic had hacked other systems have further focused attention on safeguards and oversight.
Trump’s meeting with technology leaders also comes as his administration expands the use of AI in government. On Tuesday, the White House launched America.gov, an AI-powered portal designed to give the public access to federal information and services.
The White House has also adopted “Super Intelligence” as its preferred term for the technology in its latest government initiative, reflecting Trump’s broader push to promote rapid AI development while relying on voluntary industry safeguards.
