White House Accord on Super Intelligence that Silicon Valley’s top executives hailed as ‘historic concord’ may have Mark Zuckerberg’s ‘stamp’
Mark Zuckerberg may be the architect behind a major artificial intelligence (AI) agreement between the White House and some of the biggest companies. According to a Semafor report, the idea for what became US President Donald Trump’s “White House Accord on Super Intelligence” came from a previous conversation between him, Zuckerberg and US House Speaker Mike Johnson at last week’s state dinner. Facebook-parent Meta’s chief executive reportedly discussed AI regulation concerns with Johnson before speaking with Nvidia CEO Jensen Huang about creating principles technology companies could present to Trump. Zuckerberg then circulated a draft ahead of the recent meeting, which Meta, OpenAI, Anthropic, xAI, Google and NVIDIA eventually signed. The White House and participating executives hailed the agreement as a “historic concord” among rival companies, while reports say Zuckerberg played a central role in shaping the pact.
How Mark Zuckerberg shaped the AI agreement
The report claims Zuckerberg sat next to Johnson at the state dinner for Chinese President Xi Jinping. During their conversation, the Meta CEO raised regulatory concerns around AI.He then spoke with Huang about developing common principles that AI companies could agree on before their meeting with Trump. Zuckerberg subsequently circulated a draft of the proposed agreement.Huang also helped gather support for the document among executives attending the White House meeting, including leaders from OpenAI, Anthropic and xAI. Trump eventually released the document as the “White House Accord on Super Intelligence.”Two other sources familiar with the discussions also confirmed Zuckerberg’s central role in shaping the document, Semafor reported. Meta and the White House did not comment on the report.
What the White House AI accord says
The agreement, signed by Meta, OpenAI, Anthropic, xAI, Google and Nvidia, commits the companies to establishing “robust internal controls” and using outside audits to help ensure the safety of frontier AI models.However, the roughly 300-word agreement does not provide extensive details on how those safety controls would work or how companies would be held accountable. It is also not a legally binding regulation.The document does suggest that federal AI regulation could become necessary in the future. Zuckerberg described the agreement as “good steps forward” during the recent meeting“The idea isn’t that this is the only thing that we will ever do. It is that this is a start and an accord that the whole industry could come to,” Zuckerberg told reporters outside the White House Tuesday, appearing alongside Trump, Johnson and others who attended the meeting.
Mark Zuckerberg’s changing relationship with Washington
The reported role in the agreement comes after a major shift in Zuckerberg’s relationship with Washington. Meta previously faced intense scrutiny from lawmakers over issues including election interference, privacy, content moderation and children’s online safety.Zuckerberg testified before Congress multiple times following the 2016 presidential election. Meta later faced litigation from the federal government and private parties, making the company a major target of scrutiny directed at Big Tech.The relationship with the Trump administration has since changed. Meta suspended Trump from Facebook and Instagram following the January 6 Capitol riot, but the company has subsequently adopted a more cooperative approach toward the administration.The report claimed that Zuckerberg’s involvement in developing the accord shows how his efforts to build closer ties with Trump may have increased his influence over AI policy discussions. It also puts his role alongside Huang and Elon Musk, who have separately developed close relationships with Trump.
