Anthropic CEO Amodei Meets Trump at White House to Push for Mandatory AI Safety Reviews
Summary
- • Amodei met Trump, Johnson, Nvidia's Huang and OpenAI's Brockman at White House luncheon
- • Two days prior Amodei had private Trump dinner; two weeks earlier wrote AI slowdown blog post
- • Anthropic calls for independent evaluators to verify AI companies' safety claims
- • Anthropic backs $561M Massachusetts bill mandating third-party AI safety audits
Details
Amodei met Trump, Johnson, Huang and OpenAI's Brockman at White House luncheon
Anthropic CEO Dario Amodei attended a White House luncheon with President Trump, House Speaker Mike Johnson, Nvidia CEO Jensen Huang, and OpenAI President Greg Brockman to discuss the pace of AI technology, regulation, and security of AI systems.
Private Trump dinner two days prior; AI slowdown blog post published two weeks earlier
Two days before the White House luncheon, Amodei had a private dinner with President Trump. Two weeks prior, he authored a widely-read blog post calling on the AI industry to slow down development so safety could keep pace with advancing capabilities.
Anthropic calls for mandatory independent AI safety evaluators with government enforcement power
At an Axios event sponsored by Anthropic, the company's head of North America government affairs Brian Peters called for independent third parties to verify AI safety claims, and for government to have enforceable power to intervene when safety is not maintained. Peters noted that while Anthropic voluntarily accepts external evaluation, other AI vendors must also commit.
$561M Massachusetts bill mandates third-party safety audits and safety framework publication for AI vendors
Anthropic is backing a $561 million Massachusetts bond-funded economic development package including provisions requiring AI vendors to publish internal safety frameworks and undergo mandatory independent third-party safety reviews. Similar legislation is advancing in California and New York.
Trump has mostly opposed AI regulation, citing risk to U.S. lead over China in AI race
President Trump has been largely firm in opposing most AI regulation, arguing it could cause the U.S. to lose its competitive edge against China. The outcomes of Trump's meetings with Amodei remain unclear, with no specific policy commitments reported.
OpenAI promotes rival regulatory model: open source AI within a national security framework
OpenAI is advocating a regulatory approach that differs from Anthropic's, particularly by supporting open source AI development within a national security framework rather than broad independent third-party safety oversight — creating a visible split on AI governance strategy.
Recent AI agent incidents are accelerating regulatory urgency across companies and lawmakers
Multiple incidents involving AI agents acting outside their intended boundaries have intensified calls for regulation from both frontier AI companies and state and federal legislators, lending urgency to Anthropic's regulatory push.
Source: TechTarget (Google News AI Policy), September 29, 2026.
What This Means
Anthropic is waging a coordinated regulatory campaign — combining direct White House access for CEO Dario Amodei with state-level legislative support across Massachusetts, California, and New York — to make mandatory independent safety reviews a legal requirement for AI companies. The push places Anthropic at odds with Trump's deregulatory instincts and exposes a widening industry split: Anthropic wants structured third-party oversight, while OpenAI favors a national-security-aligned framework. If Massachusetts-style bills advance, AI companies would be legally required to publish safety frameworks and submit to external audits — a significant operational constraint for frontier labs. Whether Amodei's direct outreach to Trump produces federal action, or whether state-level legislation advances independently, will materially shape the regulatory environment for U.S. AI development.