All 20 members of the Group of 20, including China and Russia, endorsed a U.S.-backed framework for governing artificial intelligence on Wednesday at a technology ministerial in Chapel Hill, North Carolina, handing the Trump administration a diplomatic win built around a deliberately light regulatory touch that stands in sharp contrast to the European Union's approach.
The non-binding accord, dubbed the Carolina Principles, calls on G20 nations to apply sector-specific rules to AI rather than sweeping cross-sector legislation, to avoid standing up new regulatory bodies dedicated to the technology, and to lean on collaboration with industry rather than top-down mandates. White House science adviser Michael Kratsios, who promoted the framework, argued that "policymakers do not need to approach each innovation in isolation," according to Al Jazeera's account of the summit. Commerce Secretary Howard Lutnick called reaching consensus among all 20 members "no small feat."
The two-day ministerial produced a broader consensus statement organized around six pillars, including pro-innovation policy frameworks, workforce development and supply-chain investment, along with a companion set of "AI Prosperity Objectives" and an "AI Prosperity Compact" aimed at expanding industry partnerships, according to the White House's summary of the statement. The document carries no treaty obligations, fines or compliance deadlines — it is a statement of shared intent, not enforceable law.
A pointed contrast with Brussels
The principles are, by design, close to the opposite of the path the European Union has taken. The EU's AI Act sorts systems by risk tier and imposes binding compliance obligations on the highest-risk categories, and the European Commission confirmed this week that it has sent compliance inquiries to more than 30 AI companies worldwide even as the G20 statement was being finalized. Henna Virkkunen, the Commission's tech sovereignty vice president, said Brussels intends to "ensure that AI in Europe is developed, released and used safely and transparently" — language that underscores how far apart the U.S. and EU positions on AI governance remain, even with China and Russia now nominally aligned with Washington's lighter-touch framework.
That divide was visible at the summit itself: American technology executives present, including Elon Musk, pressed for minimal constraints on AI development, with Musk arguing that innovation requires letting new technologies be "default legal as opposed to default illegal," in contrast to what he described as Europe's default-illegal posture.
What happens next
The Carolina Principles are non-binding and will be presented for formal adoption at the full G20 leaders summit in December, hosted at a Trump golf property in Doral, Florida — a step that would elevate the framework from a ministerial statement to something closer to an official G20 position, though still without enforcement mechanisms. Whether China and Russia's endorsement translates into any change in their own domestic AI rules remains an open question, and the EU's parallel enforcement push suggests the practical divide between light-touch and prescriptive AI regulation will persist regardless of what leaders sign in Florida.
