The panel argues that Europe has not yet lost the AI race, but it is clearly behind the U.S. on capital, compute, and product adoption. The debate centers on whether Europe should mainly regulate AI or urgently build its own data centers, models, and data infrastructure to avoid dependence on American and Chinese platforms.
Watch on YouTube ›Get the market thesis, key claims, assets, contradictions, and follow-up questions from any financial video — then unlock a version personalized to your portfolio, watchlist, and favorite speakers.
This discussion frames AI as both an industrial competition and a sovereignty issue. Thierry Breton argues that Europe has not “lost” the race because AI rests on three pillars — data, software/models, and computing power — and Europe still has regulatory capacity, a large market, and the ability to build infrastructure if it chooses to act now. He repeatedly insists that the key issue is not abstract ideology but ownership of data and compute, and that Europe must move immediately to build its own data centers and “modéliser nos données” before the ecosystem becomes locked into American or Chinese models. Several speakers emphasize how far behind Europe already is on capital and scale. The host cites 2026 digital investment figures of 647 billion euros in the U.S. versus 132 billion in Europe, along with 70% of global AI compute being located in the U.S. versus 4% in the EU. …
Tactically, the key risk is Europe continuing to lag while U.S. AI platforms deepen penetration in Europe. Any credible EU push on data centers, compute, or harmonized AI rules would be the near-term catalyst to watch.
Over the next few months, the base case is that Europe remains behind unless it commits real capital and simplifies regulation; without that, foreign AI adoption likely keeps rising. Confirmation would be visible in domestic compute buildout, funding, and faster policy execution.
Structurally, AI looks like a sovereignty regime where compute, data ownership, and platform architecture determine who controls the value chain. Europe’s long-run relevance depends on whether it becomes an AI builder with its own standards or stays a regulated consumer of U.S. and Chinese systems.
Europe has not lost the AI race because AI depends on data, software, and computing power, and Europe still has leverage on data and regulation.
Breton argues that AI is made of three elements and says Europe has rules on data and has not lost.
Europe is behind on AI adoption and capital, with much lower digital investment than the U.S.
The host cites a large investment gap between the U.S. and Europe, and Bacqué says Europe is very late.
French users are already heavily dependent on non-European AI tools, mostly American ones.
Bacqué cites usage figures showing foreign AI penetration in France.
Il faut une capacité financière?
Raphaëlle Bacqué répond que oui, pour pénétrer le marché domestique. Elle rappelle que ChatGPT existe depuis 2022, qu'en France 45 à 50% des gens utilisent une IA non européenne (essentiellement américaine), et que nos données sont très largement parties ailleurs. Elle reconnaît que l'Europe est en retard mais dit que la bataille n'est pas terminée et qu'on espère que l'Europe va pouvoir y rentrer à pleine vitesse.
Unlock the full claims, asset map, scores, related transcripts, follow-up questions, and AI chat — shaped around your portfolio, watchlist, favorite speakers, and risks.
