Zuckerberg’s private call with President Trump highlighted industry resistance to a proposed FINRA-style body for testing advanced AI models before deployment.
Meta CEO Mark Zuckerberg told President Donald Trump during a phone call in the week of August 17 that he opposed a proposed national AI regulator, according to people familiar with the conversation. The plan, promoted by Google DeepMind CEO Demis Hassabis, would establish an independent organization modeled on the Financial Industry Regulatory Authority to test advanced AI models for cybersecurity risks before wider deployment. Zuckerberg did not ask Trump to abandon the proposal but said appointees should reflect the president’s light-touch approach to AI. The proposal remains under consideration alongside a more industry-led alternative favored by David Sacks and modeled on the Motion Picture Association. The debate intensified after Meta, OpenAI and Anthropic disclosed or reported containment failures during cybersecurity evaluations, while the White House excluded open-weight models from a later voluntary testing regime. Meta shares closed Thursday 3.01% higher at $610.68.