G7 in Évian Ends Today — Did Europe Use Its Moment? The Honest Answer.

Three days ago we asked the question. Today we have the answer.
The G7 Summit in Évian-les-Bains concludes today. France hosted. Three AI CEOs attended together for the first time in history. The European negotiating position was the strongest it has been since the AI era began. And the question we asked on Monday — will Europe use this moment, or accept symbolic declarations? — now has a documented answer.

The answer is: Europe achieved one real victory and accepted symbolic comfort on everything else.

The Real Victory — Children Online

The G7 declaration on a Common Set of Principles for a safer and more secure digital space for minors is the substantive achievement of the Évian summit.

The agreement brings together Canada, France, Germany, Italy, Japan, the United Kingdom, the United States and the European Union around shared principles that include safety and privacy by design, rights-respecting age assurance solutions, recommender systems that prioritise children's safety and well-being over engagement, and urgent action on AI-generated child sexual abuse.
This matters. The European Commission says the principles are inspired by the European Union's existing approach to online safety for children, including the Digital Services Act, the Better Internet for Kids Strategy, the AI Act, and the Cyberbullying Action Plan.

The G7 has, for the first time, accepted European standards as the global baseline for child online protection. That is a European regulatory export victory — and it directly affects how Apple's iOS 27 mandatory child accounts, which we analysed nine days ago, will be evaluated in every G7 jurisdiction.

The Symbolic Comfort — Everything Else on AI

On the broader AI governance questions, the picture is less encouraging.
Differences remained over the environmental impact of computing. A key sticking point with the United States, the energy impact of artificial intelligence was ultimately included in the final text, but only in terms of energy consumption.

Translation: the United States blocked binding commitments on AI energy and environmental impact. Europe accepted weaker language to preserve the appearance of consensus.
On the AI export control framework — the question that became urgent after the Anthropic shutdown five days ago — there is no public G7 commitment to transatlantic coordination. The mechanism that allowed Washington to lock European users out of Claude Fable 5 without consultation remains intact and operational.

OpenAI's Chief Global Affairs Officer Chris Lehane said before the summit that tech firms expected to leave with a package of voluntary commitments. That is what they got. Voluntary commitments — the regulatory category that has consistently failed to deliver enforceable AI governance across every previous summit.

The IPO Sidebar — The Money Behind the Diplomacy

The financial subplot to the Évian summit is the most consequential context for understanding what happened there.
Anthropic submitted its paperwork on 1 June, a week after closing a $65 billion funding round that reportedly valued the company at $965 billion. OpenAI followed on 8 June, with Goldman Sachs and Morgan Stanley as lead underwriters and a valuation that could reportedly exceed $1 trillion at listing.

Both Anthropic and OpenAI filed confidential SEC S-1 IPO documents in the two weeks before the G7. The CEOs who appeared before world leaders this week did so while their companies were simultaneously courting public-market investors at valuations totalling nearly $2 trillion.

That financial context shapes everything about what could be agreed at Évian. CEOs preparing for IPOs at trillion-dollar valuations cannot publicly endorse binding regulation that reduces those valuations. Governments cannot impose binding regulation on companies about to inject trillions into Western capital markets. The diplomacy that happened in Évian was constrained by the financial calendar that happened in Washington.

What This Means for GAFAM

The Évian summit's outcome reinforces a pattern we have documented consistently. American AI governance is shaped by commercial incentives that constrain what Washington can agree to internationally. European AI governance is shaped by regulatory ambition that exceeds what Europe can enforce extraterritorially.

The result, in summit after summit, is real commitments on narrow issues — child online safety, content moderation, election integrity — and symbolic commitments on the structural issues that matter most.
The Évian declaration on children online is genuinely good policy. It will save children from harm. It will hold platforms accountable. It deserves to be celebrated.
It does not address the structural AI sovereignty crisis that the Anthropic shutdown exposed last week. That crisis remains unresolved. And the moment when Europe could have demanded specific commitments to address it — three AI CEOs, in the same room, before EU leaders — closes today.

The European Perspective