Torbellino Tech
Arguments why Europe is the best candidate place to build AI
August 30 2025
1. AI has not been built
First, before even considering this matter, we need the following claim: Artificial Intelligence has not been built. Here AI understood as a truly autonomous system. Basically, the prelude is that we have seen ``AI models'' deployed but none are truly autonomous. I will indirectly argue for this claim in the following.
2. Comparative arguments
Also, before presenting the actual arguments, we need to see this is a comparative matter. ``Europe is the best candidate place to build AI'' compared to something. Basically, compared to the US and China since this is the current situation.
3. The raw computing capacity argument
A metric very often mentioned when talking about AI is the raw computing capacity available. US and China are leading when it comes to raw computing capacity, a valid argument often used against Europe. However, there is an argument that raw computing capacity is not that relevant, and will progressively be less relevant in this context, since the costs will go down and most importantly, since ultimately the hardest task is at the software level (system design).
4. The scientific argument
This argument basically goes like this: ultimately AI is a scientific endeavour and Europe is best positioned at the scientific level. The engineering, which is strictly necessary nevertheless, should lead to some scientific result. In the long run, Europe will come out ahead because it will catch up at the engineering level and it will outperform at the scientific level.
This is a topic I have covered before (youtube channel). Also, see the following argument.
5. The historical (scientific) argument
This argument complements the previous one. Historically, Europe has been the most important region at science. From the scientific revolutions at the origins of modernity even to the troublesome 20th century, Europe has been the center of scientific production.
6. The talent argument
It is often mentioned that most prominent engineers and scientists in the field are from the US and China (and work for US and China based companies). This may be true at some level (I have not formally checked the statistics). However, it can very well just be that they have been able to mobilize better the existing talent towards this goal for now (at some level). It does not mean that the talent is not there.
Secondary arguments here talk about better economic conditions (e.g. better salaries or better economies of scale), which go to strengthen the main argument.
So the argument basically goes like this: there is talent in Europe, it is just that conditions need to change (and these conditions can change).
7. The still early argument
AI has not been built. We are still in the process of building it. In fact, we are early in the process. For now, we have seen some ``AI models'' deployed, that are iterated every year and measured against some benchmarks (a formidable task in and of itself), but there is still a lot of work to be done to get to a truly autonomous system. Even if Europe is behind for now we are still early in the race.
8. The anarchist argument
This argument is probably the most subtle and difficult to see clearly.
Some have argued very strongly in favor of making the AI matter a political matter. Arguments vary: AI for the common cause of the state, AI needs to be controlled by preexisting, accepted society standards and procedures, AI favors from state fostering it and state sponsored programs, \ldots At some level this seems reasonable. For example, we can clearly see a connection with argument 6 and secondary arguments. However, ultimately a political organization is defined by its (political) goals and, regarding technology, a political organization has to and strives to subordinate technology to achieve its own goals. Therefore, in the long run it is not so obvious how these balances of power will come to a resolution.
This argument basically goes along the lines: ultimately there is an implicit anarchist thesis in the concept of AI and Europe is better positioned in this regard.
Anyways, it begs the question: who is really committed to this endeavour?