On October 1, multiple AI company CEOs met with President Trump at the White House to sign a declaration of principles regarding AI safety. During the event, industry leaders, including Nvidia CEO Jensen Huang, challenged Anthropic CEO Dario Amodei over his public warnings about the extreme risks associated with AI models.
The discussion took place in the Roosevelt Room among a small group of executives and White House advisors. Industry leaders questioned why Amodei emphasizes the potential dangers of AI, such as cyberattacks and mass unemployment. Amodei maintained that it is essential to be transparent about model capabilities rather than downplaying risks.
Meta CEO Mark Zuckerberg played a significant role in finalizing the AI principles, frequently consulting with President Trump and Commerce Secretary Howard Lutnick. Zuckerberg and Huang advocate for a lighter regulatory approach, opposing suggestions from researchers at Anthropic and OpenAI that model development should be slowed down in coordination with the government.
The tension between Anthropic and other industry players may complicate efforts to implement self-regulation through internal controls and model evaluations. While Trump praised Amodei as brilliant, the disagreement highlights a divide in the industry regarding the balance between rapid development and safety oversight.
