Photo: Tima Miroshnichenko / Pexels
Zuckerberg's push against centralized AI oversight aligns neatly with the Trump administration's own deregulatory instincts, creating an unusual alliance between Silicon Valley and the White House.
Mark Zuckerberg has made his position on AI regulation about as subtle as a billboard on the Las Vegas Strip. The Meta CEO told President Trump directly that creating a national AI regulator would be a mistake, adding his voice to a growing chorus of tech executives who view centralized oversight as a threat to American innovation rather than a safeguard for it.
The Trump administration has separately ruled out establishing any kind of central licensing agency for artificial intelligence, drawing a clear line against anything resembling an FDA for AI models.
Open source as the anti-regulation play #
Zuckerberg’s argument isn’t just “regulation bad.” He’s been warning consistently throughout 2026 about the risks of what he calls “regulatory capture,” the phenomenon where the agencies designed to police an industry end up being controlled by the very companies they’re supposed to oversee.
In a Financial Times interview on July 29, Zuckerberg acknowledged that peer reviews of AI systems could have value. But he drew a sharp distinction between collaborative oversight and a permanent government body with licensing authority.
Meta has bet heavily on open-weight AI models, releasing its Llama series for broad public use. A centralized regulator with the power to approve or reject AI models before deployment would fundamentally threaten that strategy.
That approach puts Meta squarely at odds with rivals like OpenAI and Anthropic, which have generally favored more structured regulatory frameworks. Those companies operate with closed or semi-closed models, meaning a licensing regime would create less friction for their business models and potentially more friction for Meta’s open-source playbook.
A White House already leaning his way #
The Trump administration’s AI policy has been moving away from prescriptive regulation, with officials making public statements as early as July 2026 explicitly ruling out the creation of a national AI regulator.
Sriram Krishnan, who served as a former AI adviser to the administration, has echoed similar concerns about overregulation stifling the pace of AI development.
Since the 2024 campaign, Zuckerberg has actively engaged with Trump, gradually positioning Meta’s corporate interests alongside the administration’s broader economic philosophy.
At the G20 ministerial meeting in September 2026, US tech executives including Zuckerberg pushed the same message on an international stage, calling for minimal AI regulations and promoting open-source approaches.
The stakes behind the talking points #
The question isn’t whether AI should be governed at all. It’s whether governance should be centralized in a single federal body or distributed across existing agencies, industry self-regulation, and market forces.
Zuckerberg and his allies counter that AI moves too fast for a bureaucratic agency to keep pace, and that concentrating approval power in one place would inevitably favor incumbents with the resources to navigate complex compliance requirements.
The European Union has already moved forward with its AI Act, creating exactly the kind of comprehensive regulatory framework that Zuckerberg opposes domestically. If the US continues down the deregulatory path, American and European AI ecosystems could diverge significantly, creating compliance headaches for companies operating in both markets.
Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our