Politicians are currently working on two levels to ensure that Europe and Germany can be at the forefront of the AI transformation of the economy, science and society, while also reducing dependence on the USA and China. So far, Europe seems to be playing second fiddle – less in research than in the implementation of the business models based on it. What's more, the serious differences in investment levels speak for themselves.

Just a few days ago, the EU digital Commissioner Henna Virkkunen gave the go-ahead for five so-called „gigafactories“. They are intended to provide a kind of ecosystem of computing capacity, data storage and secure access to electricity as well as a favourable environment for researchers and companies. The Jülich Research Centre, where powerful supercomputers are already located, sees great opportunities as a location in Germany. And in the coalition agreement, the CDU/CSU and SPD have pledged to make Germany an „AI nation“ and a „top location for future technology“. Cloud and AI infrastructure are to be massively promoted in conjunction with robotics, and regulation is to be designed to be „innovation-friendly“.

The timing seems favourable, as the USA, a leading AI nation, is being held back by Trump's tariffs announcements, and some scientists are even said to be considering moving away due to his aggressive policies. In any case, the costs of building, equipping and operating data centres in the USA will rise dramatically, warns Sam Altmann, head of OpenAI. Even if some technology components are now temporarily exempt from tariffs again, „there are many other parts of a data centre that will become more expensive“, he said in a speech at the Vanderbilt Summit. Google alone had announced that it would be spending 75 billion dollars on data centres this year, while Microsoft is even aiming for 80 billion dollars.

„I am very concerned,“ Andrew Ng, the former head of the Google AI lab, is quoted as saying. Some US AI companies could now decide to build data centres outside the USA. This will depend on how attractive the AI location is. Germany looks to hold good cards here, as the latest AI Index Report published by Stanford University shows. It summarises all the important information relating to AI and attracts a great deal of attention in politics and business due to its gigantic amount of data. In terms of the number of AI researchers, India recorded the largest growth in 2024, but Germany ranked fourth in terms of talent density after the USA, India and the UK. Germany is also in sixth place in terms of inflow of AI workers.

In terms of public and private investment, the USA is in first place by a wide margin. The vast majority of public investment in the USA also comes from the Pentagon, while the lion's share of private investment comes from the big tech companies. However, the market is becoming increasingly differentiated into smaller segments and appears to have entered a consolidation phase.

Integration into existing structures is now more important than new language models. Specialised data is of crucial importance. This applies, for example, to medical applications, materials research and product development, the use of autonomous vehicles or collaborative robots. The report shows that many European companies are successful in this area.

The fact that approval of AI is growing in Europe, according to the surveys cited in the report, could also prove helpful. When asked whether AI brings more advantages than disadvantages, only 39% of respondents in the USA agreed, while 47% in Germany agreed with this assessment. Also favourable for companies setting up here: System costs have fallen dramatically. At the level of AI models, they have fallen more than 280-fold since 2022, hardware costs are 30% lower and energy efficiency has improved by 40% annually.