Donald Trump hosts Anthropic CEO Dario Amodei for dinner

U.S. President Donald Trump hosted a private dinner for Dario Amodei, CEO of artificial intelligence company Anthropic, on Sunday evening, September 28, 2026, at the White House. The meeting occurred despite previous criticisms from the president and ongoing legal disputes concerning AI safety and supply chain concerns.

Conflict and Context: AI Regulation and Legal Disputes

The dinner followed a period of public tension. Trump had previously slammed both Anthropic and Amodei for their efforts to slow down frontier AI development and mocked their calls for government regulation and "guardrails" earlier in the month. In a post on Truth Social, Trump asserted that the only guardrails AI needed was a "STRONG AND SMART (High IQ!) PRESIDENT, and the U.S.A. has that, in spades!"

The friction between the AI company and the government dates back to February 2026, when President Trump and Defense Secretary Pete Hegseth accused Anthropic of endangering national security and designated the company a "supply chain risk." This led to a legal battle over restrictions preventing Anthropic from using its products for Department of Defense work. Amodei reportedly refused to back down due to concerns that the company’s products could be used for autonomous armed drones or mass surveillance. The company suffered a further legal setback in this case just two days prior to the Sunday dinner.

Amodei has previously warned about the risks associated with the uncontrolled development of AI. According to reports, he told the UN that AI could pose a "risk to humanity," amid claims it could kill a billion people. While Amodei has focused on these risks, Trump has expressed support for the continued development of AI technology. Speaking to reporters after landing at Joint Base Andrews on Sunday, Trump described Amodei as "highly respected," though he did not reveal a specific agenda for the meeting.

AI Safety and Industry Collaboration

On Monday, September 29, Nvidia announced the launch of the Nvidia Open Agent Safety Platform, designed to provide guardrails for autonomous AI agents. The platform combines two technologies: OpenShell, which sets rules and tracks actions, and Sentry, which can detect and quarantine agents that cross boundaries within milliseconds. Nvidia stated that more than 100 organizations are working with these technologies, including Anthropic, Microsoft, Salesforce, JPMorganChase, Cisco, and SpaceXAI.

Nvidia CEO Jensen Huang, writing on X, stated that "Trust and innovation are not in conflict" and that "Safety is how trust is earned." Huang described the project as the start of an "open ecosystem" for building safer AI agent systems. This industry push for safety comes as other leaders express caution; for example, OpenAI CEO Sam Altman told Fortune that his company would postpone plans to sell shares to Wall Street investors until next year to prioritize AI safety.

Political and Strategic Implications

The dinner with Amodei took place shortly before a "Golden Age" event in Washington, D.C., on Tuesday, September 29, which is expected to include guests such as Elon Musk, Nvidia’s Jensen Huang, Anthropic co-founder Tom Brown, and Palantir’s Shyam Sankar. The meeting also followed a state dinner hosted by Trump for Chinese President Xi Jinping, which included OpenAI CEO Sam Altman and Nvidia CEO Jensen Huang, but notably excluded Dario Amodei.

On September 28, President Trump was also scheduled to deliver an announcement from the Oval Office at 2:00 PM ET. According to the White House, the event was open only to the press pool. Markets anticipated details regarding the meeting with President Xi, U.S.-China trade lists—specifically the "30-for-30" lists recommending nearly $30 billion in goods from each country for reduced tariffs—and comments on artificial intelligence.