Artificial intelligence is a disruptive technology that is causing profound change to the economy and society. In Europe, we need to take greater advantage of the opportunities arising from this, said Bundesbank Executive Board member Fritzi Köhler-Geib in an interview with the “Süddeutsche Zeitung”. In the interview, she discusses the opportunities and challenges that AI presents for Europe and the stability of the financial markets.

Europe lagging behind in AI

According to Köhler-Geib, Europe is lagging behind the United States and China with regard to large AI models. As a recent example, she cited the US company Anthropic, whose latest AI models were blocked for the rest of the world by order of the US Administration. This very clearly shows how strategically significant these technologies have become, she emphasised. If Europe does not have access to models with such capabilities, then it needs to develop its own alternatives: We need to advance our digital sovereignty.

AI in financial markets

The use of AI in the financial markets brings opportunities, but also harbours risks. The decisive factors will be the models used and the data with which they are trained. If all models are based on the same datasets, this quickly leads to them all moving in the same direction, so to speak, explained Köhler-Geib. This could amplify herd behaviour: In human terms, it would be as if every equity investor had the exact same education, the exact same models, and would make the exact same decisions. The phenomenon of herd behaviour has been known in the markets for a long time now. With AI, however, it is taking on a new technological dimension. 

Central banks and supervisory authorities are already taking action to address these risks. They are trying to obtain a picture at an early stage so that they can understand the potential risks and – where necessary – put guardrails in place, said Köhler-Geib. One example of this is the BIS Innovation Hub’s Project Logos, in which central banks analyse the behaviour of AI agents within a simulated test environment.

New generation of trading algorithms

Algorithmic trading has been around for decades, but AI is shifting the boundary at which humans have direct influence over trading decisions. With what is known as ‘agentic’ AI, which is also becoming increasingly powerful, systems are making more and more decisions autonomously without human intervention in each individual case, explained Köhler-Geib in the interview.

The European Banking Authority is currently working to obtain a better overview of where and how AI is being used for trading. However, in this context, the risks arising from these models do not necessarily need to be greater than those arising from traditional algorithmic trading, said the Bundesbank Executive Board member. 

Limitations of AI models

According to Köhler-Geib, biases can also creep into AI models, similar to human investors: These models are often very good at identifying such biases. However, they have not yet been able to reliably address these biases. This is due, in part, to the fact that these models lack consciousness and self-awareness. This may see further development in the coming years – but, as things stand today, it remains a key constraint for the way in which AI is used in the financial markets, she said.