Europe 2031 is a five-year scenario, written by a group of researchers, think-tankers and investors, all of whom have spent their careers at the intersection of AI and European policy. It begins in January 2025, with the public release of the DeepSeek R1 model, and runs to April 2027, at which point the scenario branches.
One branch is labelled The Drift. It is not a prediction; it is an exercise in disciplined imagination — internally consistent, traceable to current political conditions, and intended to clarify the cost of inaction. The piece closes with a Post Mortem set in 2034, looking back on the choices that brought Europe there.
A central part of the scenario concerns compute — the data centres and chips required to train and run frontier AI systems. Our default estimate is that Europe is on track to host only around 3–5% of global AI compute by the end of the decade, compared to roughly 70% in the United States and 15% in China.
The forecast is anchored in projections by the AI Futures Project, with a logistic fit through 2031 and supply-chain bounds derived from EUV-lithography capacity. Read the full methodology →
We believe AI is a big deal, and will become an even bigger deal very soon. Collectively, however, Europe is asleep at the wheel: if current trends continue, the AI transition will render Europe irrelevant on the global stage. Europe's values will atrophy, and its more careful instincts will carry no weight over the most consequential transition of our time.
AI's impact will be comparable to the Industrial Revolution. Unlike past technologies, AI will at some point be able to automate virtually all work — already reshaping cognitive labour, soon physical labour once it powers robotics. Jobs reshuffled at scale, entire industries reordered, the global balance of power redrawn.
AI progress will be fast. More compute, richer datasets, and better algorithms are driving exponential growth in AI capabilities. There is no good reason to expect it to stop, or to halt at human level. Systems already write most of the code at frontier labs and will increasingly help build their smarter successors. Major implications appear years away, not decades.
Europe is on track to become irrelevant. Europe's AI strategy is 10–100× less ambitious than it needs to be, and aimed at the wrong target. Existing sovereignty efforts mistake what sovereignty actually requires: building leverage and accepting uncomfortable trade-offs. In every past industrial revolution, the biggest losers were countries that failed to adapt.
Many Europeans believe AI won't be a big deal, that progress will stall, or that Europe can easily catch up. If AI is as big a deal as we claim, why doesn't everyone believe this already?
An uncomfortable truth. Accepting AI's significance is psychologically painful. For centuries we told ourselves that humans are special; AI is now mirroring traits we took to be uniquely human. The geopolitical picture is similarly dire — on the current trajectory, Europe becomes closer to a developing country than to a global power. Many Europeans are looking for reasons why AI is fake.
An unfortunate messenger. The clearest signals about what's next in AI come from Silicon Valley elites — circles that Europeans resent and choose to ignore, often for understandable reasons. But that does not make them wrong about AI: they have consistently been mostly right about the overall trajectory.
An unequipped state. Technical AI experts are scarce and rarely work for the public sector. Governments lack in-house expertise; attention is consumed by political crises; reforms are politically costly. Europe is also fragmented — coordination slowed by consultations, roundtables, and expert panels. The US has these problems too, but compensates with a dense private-sector inner circle. Europe has no equivalent.
No. It is an exercise in disciplined imagination — internally consistent, traceable to current political conditions, and intended to clarify the cost of inaction.
A group of AI researchers, think-tankers and investors who have spent their careers at the intersection of frontier AI and European policy. The authors are Michiel Bakker, Daan Juijn, and Stan van Baarsen.
Lorem ipsum dolor sit amet — replace with a real contact route (email, X handle, mailing list) before publication.