The Trump administration secured a major diplomatic win Wednesday as all 20 G20 member nations — including China and Russia — unanimously endorsed a U.S.-backed framework advocating for light-touch, pro-innovation artificial intelligence (AI) regulation at the conclusion of a two-day summit.
The accord, the Carolina Principles, encourages member states to adopt sector-specific approaches to AI regulation, avoid establishing new regulatory agencies, and collaborate closely with private industry to evaluate emerging technologies. Although the framework is non-binding, it sets the stage for formal adoption at the upcoming G20 leaders summit, scheduled for December at Trump National Doral in Florida.
Commerce Secretary Howard Lutnick announced the agreement alongside White House Office of Science and Technology Policy Director Michael Kratsios. Lutnick noted that securing unanimous backing required “an enormous amount of work,” adding that “when you bring global leaders together, there’s always lots of politics.”
“The U.S. message to our guests has been very simple: we want more growth and innovation around the world,” Kratsios said during a joint news conference.
The consensus included full endorsement from major U.S. rivals, including China and Russia. The consensus comes just weeks ahead of a planned meeting in Washington between President Donald Trump and Chinese President Xi Jinping, where AI governance is expected to be a key agenda item.
Prominent Silicon Valley executives attending the summit echoed the White House’s pro-innovation, light-touch stance.
NVIDIA Corp. CEO Jensen Huang urged governments to regulate “practical and actual harm, and not regulate theoretical and hypothetical harm,” describing AI as “the great equalizer” that will “democratize intelligence.”
OpenAI CEO Sam Altman framed AI adoption as a global necessity, comparing the shift to the adoption of electricity and calling its use “non-negotiable” for modern economies.
SpaceX CEO Elon Musk warned that over-regulation threatens to stifle technological growth, projecting that AI could expand the global economy by 20% to 30%.
Palantir CEO Alex Karp acknowledged the risks associated with AI but pushed back against widespread “doom messaging.”
Energy and infrastructure were key topics during the gathering.
Anthropic co-founder Tom Brown urged officials to streamline permitting for data center construction, praising a recent social media statement by President Trump that warned communities opposing data centers would “end up being backwards and poor.”
Lutnick similarly defended data center infrastructure, arguing the facilities “bring prosperity” and dismissing concerns over their environmental impact and water usage. Lutnick also confirmed that Anthropic has restored its working relationship with the White House and remains in good standing.
Despite the broad agreement, international perspectives diverged on regulatory necessity. Henna Virkkunen, executive vice president for Tech Sovereignty at the European Commission, supported fostering innovation but highlighted emerging security threats, including recent attacks by rogue AI models.
Virkkunen cautioned that ongoing international coordination remains critical, noting, “It’s not easy to agree among so many different countries.”