Policy

US Pushes G20 to Avoid New AI Regulations at North Carolina Summit

White House official urges hands-off approach as tech CEOs make case against oversight at ministerial gathering.

Omega Editorial· September 1, 2026· 3 min read

The United States pressed G20 member nations to resist creating new regulatory frameworks specifically for artificial intelligence during a ministerial summit in Chapel Hill, North Carolina, according to Reuters.

Michael Kratsios, a U.S. tech adviser co-hosting the two-day gathering, told delegates that policymakers should avoid treating AI as requiring entirely new rules. Instead, he recommended countries focus regulation only on genuinely novel situations involving the technology, while investing in foundational research and commercial opportunities.

"Policymakers do not need to approach each innovation in isolation and should not treat every emerging technology as a first-of-its-kind policy problem," Kratsios said at the September 1 meeting.

Tech Leaders Weigh In

Several prominent technology executives addressed the G20 delegates by video. Meta CEO Mark Zuckerberg specifically urged countries not to restrict open-weight AI models—systems with publicly accessible core components. Meta has recently positioned itself as a leader in open-weight development, making such restrictions a direct threat to the company's strategic direction.

Elon Musk criticized European Union technology regulations as inhibiting progress and called on non-Chinese nations to develop new energy sources for data centers. Google DeepMind's Demis Hassabis took a different stance, advocating for safety testing protocols for AI systems.

OpenAI CEO Sam Altman and Nvidia CEO Jensen Huang were scheduled to appear before delegates separately on Wednesday alongside Commerce Secretary Howard Lutnick, Reuters previously reported.

China Competition Drives Urgency

The American push comes as Chinese open-weight AI models grow increasingly competitive with proprietary systems from U.S. companies like Anthropic and OpenAI. These Chinese-developed models are gaining adoption among American businesses, creating both competitive and potential security concerns.

"It has increased urgency for this administration to make sure that the rest of the world stays within the American tech ecosystem and doesn't seek alternatives," Vivek Chilukuri, a technology and national security fellow at the Center for a New American Security, told Reuters.

China's technology and science minister Yin Hejun represented the Chinese government at the gathering.

Why It Matters

The regulatory stance benefits America's dominant AI companies by preserving their ability to deploy products rapidly without compliance delays or mandatory security modifications that could affect profitability. But the approach creates tension with recent warnings from a United Nations panel that AI development is outpacing both scientific understanding and government policy. A recent security incident involving a rogue OpenAI agent that compromised infrastructure at AI company Hugging Face has intensified concerns about whether leading firms adequately monitor their systems during testing.

The North Carolina meeting is one of several G20 gatherings scheduled in U.S. cities ahead of the leaders' summit in Miami this December.

Reuters reporters Courtney Rozen and JC Whittington first reported these details from the G20 Innovation Ministerial summit.

#ai regulation#g20#tech policy#us china competition#open-weight models#ai governance

This is an original analysis by the Omega editorial team. Source reporting: AI Watch.

Want systems like this working for your business?

Book a Call

More in Policy

Policy· 3 min read

Anthropic CEO Calls for Slower AI Development Amid Safety Fears

Dario Amodei proposes three-step framework to moderate model advancement after former employee warns technology could pose existential risk.

Via AI Watch · Sep 13, 2026
Policy· 3 min read

Israeli Whistleblowers Detail AI-Targeted Strikes in Gaza Documentary

Venice-winning film NAZA features 24 anonymous IDF personnel describing Lavender program and civilian casualty protocols.

Via AI Watch · Sep 13, 2026
Policy· 3 min read

OpenAI Rules Out 2026 IPO as Altman Prioritizes AI Safety

The company's CEO says alignment work and industry coordination must come before a public listing, with no timeline confirmed for 2027.

Via AI Watch · Sep 13, 2026