A secret phone call between Mark Zuckerberg and Donald Trump in mid-August focused on the potential risks of slowing down the deployment of advanced software models. The Meta CEO argued that adding a layer of government approval could delay new AI releases by one month and undermine U.S. competitiveness against China, according to a senior White House official. This opposition targeted a proposal for an independent, FINRA-style regulatory body championed by Google DeepMind CEO Demis Hassabis to test models for risk before broader deployment.
Trump administration officials are now considering a voluntary, industry-led standards group modeled on the Motion Picture Association as an alternative, an approach suggested by David Sacks. This shifted focus follows a draft executive order for a self-regulatory agency that had stalled in the White House. Other tech firms, including OpenAI and Anthropic, were separately previewed on the regulator idea in mid-August, but Zuckerberg specifically told Trump the a national regulator was a flawed idea during the week of Aug. 17.
Key sources
- SOURCE@sophiacai99“people the WH might appoint to the regulatory body should reflect Trump’s own approach to AI”x.com
- SOURCE@politico“involvement of one of the world’s top tech executives in a fraught White House policy debate”x.com
- SUPPORT@charlesrollet1“create an independent, FINRA-style body to regulate new AI models”x.com
- SUPPORT@leomschwartz“executive order draft on a self-regulatory agency for AI companies had stalled in the White House”x.com
- SUPPORT@wallstengine“slowing model releases, even by a month, could hurt U.S. leadership over China”x.com
- SOURCE@ajenglish“A kid growing up today will never be smarter than AI.”x.com
- SUPPORT@clashreport“For every 38,000 ChatGPT queries, that is the same amount of water that is used in the production of a single almond in California”x.com
- SOURCE@techmeme“joined Trump officials in lobbying policymakers against heavy AI regulations”x.com