The Wave Is Here. Europe Is Not Ready.

Bill Gates said out loud last week what many of us have been saying for two years. In his essay "A Turbulent AI Era and Critical Choices to Make", published on GatesNotes.com, he called this moment "the most turbulent time in human history" and said he is "deafened by the silence" from leaders who should be preparing. He is right. And in Europe, the silence is especially loud.

I have been building AI infrastructure since 2016. I have watched this technology move from curiosity to capability to disruption at a speed that has surprised even the optimists. Kai-Fu Lee saw it coming. In AI Superpowers in 2018, he noted that AI would not arrive gradually that the concentration of its power would reshape economic competition and national sovereignty before most institutions were ready to respond. Mustafa Suleyman went further in The Coming Wave in 2023: what is coming is not just AI, but a convergence of powerful technologies spreading so fast that governance cannot keep pace with capability. Containment, he argued, is the central challenge of our era.

And then there is Geoffrey Hinton. The man who built the foundations of modern AI and quit Google in 2023 specifically so he could warn us about it. He thought superintelligence was 30 to 50 years away. He now says he was wrong. His concern about competitive pressure is one I keep coming back to: companies are releasing AI at speeds driven by competition, not by readiness. Nobody wants to be second. The result is a race with no agreed rules and no agreed finish line.

Gates is now saying the same, and his framing matters: AI is not just a productivity tool. It is a cognitive replacement technology. It does not just help humans think faster. At some point, it will simply think instead of them and companies will have every economic incentive to let it.

"Who controls the intelligence, controls the future. Europe learned that lesson the hard way in energy. We are about to learn it again in AI." - Mikael Munck, CEO & Founder of 2021.AI

So what does this mean for Europe?

Most European organisations today run their AI on American infrastructure, American models, American terms of service. They have no real visibility into what happens to their data. They have no control over model updates that change their AI's behaviour without warning. They have no fallback if a provider pivots, gets acquired, or decides their use case is no longer commercially interesting.

I see this every week in conversations with defence organisations, government agencies and large enterprises across the continent. The dependency is real, and most decision-makers either do not know how deep it runs, or have quietly accepted it as the price of access.

Hinton put it directly when he left Google: he called for people to figure out how to manage technology that could "greatly empower a handful of governments or companies." That is the concentration risk. Not a robot apocalypse. Just power, enormous, compounding, structural power, in very few hands. And right now, most European organisations are helping that concentration along by default.

That is not a technology choice. It is a sovereignty choice and most organisations are making it by default rather than by design.

The Gates essay is a warning about jobs and equity. But the deeper question underneath it is about power: who controls the intelligence, controls the future. Europe learned that lesson the hard way in energy. We are about to learn it again in AI — unless we decide, now, that this time we build the infrastructure ourselves.

‍