AI Acceleration vs. Precaution

When he was president of France in the 1960s, Charles De Gaulle intuitively understood that his nation could not be a sovereign player on the world stage during the Cold War between the United States and the Soviet Union unless it possessed its own nuclear weapons.

What was true for France then is true today for the European Union, as China and America dominate AI. The continent cannot achieve strategic autonomy as a sovereign entity unless it joins the club with its own significant capacity.

American Big Tech already dominates Europe, which has struggled to start up its own industry, with the exceptions of the French company, Mistral AI, and the critical Dutch manufacturer of high-end chips, ASML. In the U.S., OpenAI, Microsoft, Google, DeepMind, Amazon, Meta and Nvidia are spending hundreds of billions on AI research and infrastructure. Driven by state investment, China is spending comparative billions and has shown its ability to compete globally through open-source AI models such as DeepSeek.

AI differs from nuclear weapons because it is a foundational technology that will transform all aspects of life. As such, it is not merely a technological achievement, but a cultural project. It is here that Europe’s precautionary temperament clashes with the accelerationist fever of Silicon Valley.

Does this place Europe at a competitive disadvantage that will fatally impede its advance in AI? Or will Europe’s deliberative vigilance save humanity from handing over the keys of the kingdom to intelligent machines?

The core conflict between America and its European geopolitical allies is their differing approaches to AI; the former seeks to “build first, regulate later,” while the latter seeks to “regulate first, build later.”

To explore this divergence within the West, Noema invited two top thinkers on technology to debate the topic. Benjamin Bratton directs the Antikythera project on planetary-scale computation. Francesca Bria is Barcelona’s former chief technology and innovation officer. Their exchange is more polemical than Noema’s tone usually accommodates, an expression of the passions aroused when the stakes are so high.

When The Precautionary Principle Backfires

Bratton maintains that Europe’s “regulate first, build later (maybe)” approach is backfiring, only resulting in “greater dependency and frustration, rather than any hoped-for technological sovereignty.”

As he sees it, “Europe choked its own creative engineering pipeline with regulation and paralysis by consensus. The precautionary delay was successfully narrated by a Critique Industry that monopolized both academia and public discourse. Oxygen and resources were monopolized by endless stakeholder working groups, debates about omnibus legislation and symposia about resistance — all incentivizing European talent to flee and American and Chinese platforms to fill the gaps.”

Bratton analogizes Europe’s present wariness of AI with how it killed the nuclear power industry in recent decades (outside of France), only to end up being dependent on Russian oil and gas for energy.

The lesson he draws for Europe is that it must not repeat its mistakes. “Do not ban, throttle or demonize a new general-purpose technology with tremendous potential just because it also implies risk. The precautionary principle can be literally fatal. And yet that is precisely what is happening around the newest emerging technological battleground: artificial intelligence. The same terms used to vilify nuclear power — ‘techno-fascist,’ ‘extractive,’ ‘existential risk,’ ‘Promethean madness’ and ‘fantasy’— are now regularly voiced by today’s Critique Industry to describe AI. …

“Europe surely can and should regulate the emergence of AI according to its ‘values,’ but it must also be aware that you can’t always get what you want. Europe is free to attempt to legislate its preferred technologies into existence, but that doesn’t mean that the planetary evolution of these technologies will cooperate. …It is up to Europe to decide. That is, Europe may have strong AI regulation, but this may actually prevent the AI it wants from being realized at all (again making it more reliant on American and Chinese platforms). Europe has the right to put its AI under ‘democratic control’ and supervised ‘consent’ if it wants to, but it does not have a right to be insulated from the consequences of doing so.”

In short, the accelerationists have all the momentum wherever the flow of capital is abundant and regulation is scarce. 

Digital Colonization & Infrastructure Dependency

For Bria, “Europe occupies a paradoxical position: a regulatory leader but an infrastructural dependent. We Europeans have set global standards through GDPR and the AI Act. Our research institutions remain world-class. Yet just 4% of global cloud infrastructure is European-owned. European governments, businesses and citizens depend entirely on systems controlled by Amazon, Microsoft, Google and other companies subject to the U.S. CLOUD Act’s extraterritorial surveillance requirements. When we use ‘our’ digital services, we’re actually using American infrastructure governed by American law for American interests.

“This dependency isn’t abstract — it’s existential. In the 21st century, those who control digital infrastructure control the conditions of possibility for democracy itself. Europe faces a choice: build sovereign technological capacity or accept digital colonization.”

Bria doesn’t hold back: “Many blame Europe’s digital paralysis on its critical intellectuals — those who push back against Silicon Valley accelerationism, crypto hyper-libertarianism and the rise of techno-authoritarianism. But this is misdirected. …The real choice facing Europe isn’t between criticism and construction but between authoritarian technological models or democratic alternatives.”

The European model, she argues, “differs fundamentally from Silicon Valley’s extractive optimization and Beijing’s state control. Europe should start from local demands and strengths, building AI as critical public infrastructure that serves democratic accountability, social purpose and citizen empowerment — rather than shareholder primacy or state surveillance. When critics dismiss this approach as inefficient, they only expose their ideological commitment to oligopolistic concentration.

“The difference isn’t capability but values and political imagination. Silicon Valley optimizes for extraction — how to capture maximum value from users. Europe optimizes for empowerment — how to distribute agency across society. These aren’t compatible goals, which is why importing Silicon Valley’s model would mean abandoning European democracy. …

“Critics present false choices: either embrace Silicon Valley’s model or abandon technological ambition. Either accept surveillance capitalism and digital colonialism or abandon digital transformation. Either submit to what some refer to as ‘planetary evolution’ or retreat to analog irrelevance. These binaries serve those benefiting from current arrangements by making alternatives seem impossible.”

Rather than becoming marginalized, Bria envisions that “Europe’s constraints will become competitive advantages” if AI infrastructure “operates within planetary boundaries while serving democratic rather than extractive purposes. When data centers must run on clean electricity, when water consumption faces strict limits and when carbon pricing reflects true costs.”

Europe’s Vocation

It is safe to say that Bria encapsulates the general European temper. Jacques Attali, the founder and first president of the European Bank for Reconstruction and Development, echoes her claims in more philosophical terms. AI, he has recently written, “is the most radical upheaval humanity has ever known: a shift from the logic of tools to the logic of minds. A moment where decisions are taken before we think, where desires are anticipated before they are born, and where the boundaries between freedom and prediction, between democracy and algorithm, blur into opacity.

“In this grand transformation of AI,” he continued, “the role of Europe is not to dominate, but to orient. Not to build the biggest servers, but to write the rules that will preserve our humanity. Not to chase others’ empires, but to become the guardian of meaning in a world flooded with data.

“Europe has always had this singular vocation: to think about the world before transforming it. In the face of artificial intelligence, it is once again Europe’s task not to slow down progress, but to ensure that progress remains human. …Its regulations classify risks, set boundaries, prohibit surveillance dystopias, and affirm that some technologies, however efficient, have no place in a democratic society. Thus, Europe sets a precedent: a civilization where machines are not above the law, and where the digital world adheres to the same moral imperatives as the physical one.

“AI systems learn, decide, recommend, and exclude — sometimes without anyone understanding why. Europe refuses this opacity. It demands transparency, explicability, and accountability — principles that seem philosophical, but are deeply political. What is at stake is the very notion of justice. In tomorrow’s world, a decision to grant a loan, assign a school, detect a crime, or prescribe a treatment may be made by a machine. The EU reminds us that a decision is only legitimate if it can be explained, challenged, and appealed. It affirms that freedom begins where comprehension begins. Elsewhere, AI is seen as a lever of supremacy. In Europe, it is — or should be — viewed as a means to serve the common good.”

America As A Check & Balance On Europe, & Vice-Versa

The issues raised by Attali, Bria and Bratton are being debated the world over as AI penetrates every society. Each is right in their own way. To stall when the U.S. and China are accelerating as a strategic objective, bolstered by immense resources and political will, is to fall so inexorably behind that catching up will be impossible. Yet, the more AI in the mold of the tech superpowers is firmly established, the harder it will be to ever make course corrections and challenge systemic hegemony.

The wise position, as in so much else, would be to not settle on zero-sum convictions but value the creative tension in their opposition. Technological advances, and the risks they always pose, are part and parcel of human becoming. But, so, too, “withholding from becoming” is what makes and keeps us human when the implications and consequences are unclear, but clear enough to raise reasonable concerns.

To date, a kind of division of labor appears to have evolved between relatively young America’s innovative impulse, always conquering new frontiers as the first mover, and Europe’s precautionary instinct cultivated throughout its long past of triumphs and catastrophes.

Diminishing this duality by more evenly sharing each other’s attributes would benefit both. Europe needs America to spur the unleashing of its strangled innovative potential and cannot serve as a counterweight unless it does so. America needs Europe to question and constrain the no-holds-barred hubris bent on moving fast and breaking things without considering how and where the pieces will scatter.

Together, they constitute a necessary check and balance on each other as the global center of gravity shifts East, where an altogether new set of challenges awaits.