Radical novelties in critical technologies and spillovers: how do China, the US and the EU fare?

Critical technologies including artificial intelligence, semiconductors and quantum computing are attracting attention because of their indispensable nature and their role in national security strategies. We compare China, the United States and the European Union in these technologies and their subfields.

We use large language models (LLMs) to identify which patents in these technologies can be considered most groundbreaking (not patented before) and worth replicating. These are ‘radical novelties.’

We find that the US clearly dominates quantum. Chinese and EU progress is similar. The US does slightly better than China in AI with clear dominance in generative AI, but China stands out in some important subfields, such as aerial vehicle technology. China
dominates in a larger number of semiconductor fields but not in the highest value added, which is related to design.

In a second step we assess how quickly radical novelties in these three technologies are transferred from one economic area to another and within each economic area. We find that the fastest transmission overall is for AI. The EU is by far the slowest in replicating radical novelties from the US and China, while the US and China tend to replicate European novel patents relatively quickly. Radical novelties are also replicated quickly between China and the US which is surprising given US controls on exports of critical technologies to China.

Our findings are concerning for Europe because it does not produce enough critical patents in these technologies and because it is slower in replicating patents from the US and China.

This is an output of China Horizons, Bruegel's contribution in the project Dealing with a resurgent China (DWARC). This project has received funding from the European Union’s HORIZON Research and Innovation Actions under grant agreement No. 101061700.