The United States on Tuesday called on Group of Twenty members to adopt a more cautious and restrained approach to artificial intelligence governance, urging them to refrain from establishing an excessive number of new regulations for the technology, according to a Reuters report. The position was articulated during a meeting held in North Carolina, which brought together representatives from major technology companies and trade ministers from various countries.

America's stance largely aligns with the interests of the world's largest AI firms, nearly all of which are headquartered in the U.S. These companies want governments worldwide to impose less regulation on the rapidly expanding AI sector, or at minimum, to have meaningful influence over the rule-making process. For AI enterprises, mandatory government oversight measures could weigh on profitability. If new regulations slow down the release of new models, or force companies to alter how their products operate to address safety concerns, these changes could raise operational costs and dampen business growth.

Michael Kratsios, the U.S. technology adviser co-chairing the two-day gathering, said he hopes participating nations will endorse the so-called "Carolina Principles." In remarks delivered to G20 officials on Tuesday morning, Kratsios indicated that countries signing onto the principles agree to avoid creating wholly new regulatory frameworks specifically for AI, instead focusing on crafting rules tailored to the "novel" use cases that arise around the technology. "Policymakers do not need to view every innovation in isolation, nor should they treat every emerging technology as an unprecedented policy problem," Kratsios said during his morning address.

Tech industry heavyweights convene

Anthropic co-founder Tom Brown is scheduled to address attendees on Wednesday, a company spokesperson confirmed. Google DeepMind CEO Demis Hassabis, Meta CEO Mark Zuckerberg, and SpaceX CEO Elon Musk delivered remarks to participants via video link on Tuesday. Hassabis urged G20 officials to establish safety testing mechanisms for AI systems. Zuckerberg argued that countries should not restrict open-weight AI models, which are systems where certain core components are made publicly accessible. In recent weeks, Zuckerberg has aimed to position Meta as a frontrunner in the open-weight AI space, and additional restrictions on such models could derail the company's plans in this area.

Musk, meanwhile, criticized the European Union's technology regulatory policies, stating that the bloc's rules were "holding back progress." He also called on countries outside China to accelerate the development of new energy sources to meet the surging electricity demands of data centers. As AI data center construction expands rapidly, disputes over these facilities have become a significant topic in U.S. politics. With the 2026 U.S. midterm elections approaching, the trajectory of data center projects is shaping voter sentiments toward candidates.

A United Nations expert panel recently cautioned that the pace of AI development is outpacing both the scientific community's capacity to understand it and governments' ability to keep up with policy. A string of recent incidents has also heightened concerns about AI safety, including an attack by a rogue OpenAI AI agent that led to a breach of infrastructure at AI company Hugging Face. AI agents are programs capable of operating autonomously with minimal human supervision. These events have raised questions about whether leading AI firms are adequately monitoring their product testing processes.

U.S. faces rising competition from China's AI capabilities

The G20 officials' meeting comes at a pivotal moment for the global AI industry and the debate over AI governance, as the U.S. contends with intensifying competitive pressure from China in its quest for global AI dominance. Compared to the closed-source models of American AI giants like Anthropic and OpenAI, China's open-weight AI models are becoming increasingly powerful, and the capability gap is narrowing. Chinese-developed open-weight models are also gaining broader adoption among U.S. businesses. "This makes the current U.S. administration more eager to ensure that the rest of the world remains within its technological ecosystem rather than turning to alternatives," said Vivek Chilukuri, a technology and national security fellow at the Center for a New American Security.

This gathering is one of several G20-related meetings slated across various U.S. cities in the coming months, culminating in the G20 leaders' summit scheduled for December in Miami. Reuters previously reported that OpenAI CEO Sam Altman and Nvidia CEO Jensen Huang are set to address delegates on Wednesday alongside U.S. Commerce Secretary Howard Lutnick. Meanwhile, a small group of protesters has gathered outside the G20 venue. John Montavon, a protester from North Carolina, voiced opposition to Musk's push for large-scale data center construction, arguing that rising electricity demand from these facilities would further accelerate global warming. "Mr. Musk has clearly completely detached himself from the reality that most of us live in," he said.