All 20 members of the Group of 20 unanimously endorsed the U.S.-drafted Carolina Principles on artificial intelligence Wednesday, closing a two-day ministerial in Chapel Hill, N.C. with a coordinated pledge to avoid new A.I.-specific regulatory bodies and to “reserve new regulation for novel considerations.” Commerce Secretary Howard Lutnick, who hosted, called the sign-on “an enormous amount of work.”

The framework was presented by Michael Kratsios, director of the White House Office of Science and Technology Policy, who argued that governments “do not need to approach each innovation in isolation and should not treat every emerging technology as a first-of-its-kind policy problem.” That sentence is the whole document in miniature: existing consumer-protection, antitrust, and civil-rights law will govern A.I., and no parallel apparatus is coming.

The optics were engineered. Mark Zuckerberg and Elon Musk worked the room Tuesday; Sam Altman, Jensen Huang, and Anthropic co-founder Tom Brown appeared Wednesday. Altman told the ministerial that A.I. adoption is “non-negotiable” and predicted “the greatest boom in entrepreneurship and the creation and development of small businesses the world has ever seen.” Huang, more surgical, asked governments to “regulate practical and actual harm, and not regulate theoretical and hypothetical harm.” Founders got their policy-stage endorsement. Regulators got a photo of consensus.

The split-screen is unmissable. On the same day the ministerial opened, the European Commission confirmed information requests to more than thirty A.I. companies under the E.U. A.I. Act, whose transparency rules took effect in August. Chapel Hill offered posture; Brussels offered process. Small-business operators selling into both markets now inhabit two regulatory realities at once.

Domestically, the accord is a signal, not a statute. It doesn’t preempt state legislation, and it doesn’t touch the Federal Trade Commission’s recent guidance on A.I.-driven pricing disclosures, which continues to bind any firm using algorithms to set prices. What it does do is remove the ambient threat of a sudden federal A.I. rulebook, which matters for owners deciding whether to substitute algorithmic tooling for discretionary marketing spend as a durable capability rather than a hedged bet.

Formalization is expected at the G20 Leaders Summit in December. Between now and then, the interesting question isn’t whether the heads of state will ratify what their ministers already signed. It’s which national parliaments quietly begin drafting the state-level exceptions the Carolina Principles were designed to make politically expensive.

Sources