Published in Eucrim.eu, 12.10.2023
In April 2021, the European Commission put forward a proposal for a Regulation to harmonise rules on artificial intelligence (AI) across the EU, including AI in the context of law enforcement. Its horizontal character raised concerns in the police community, prompting a response by some Member States arguing for a separate legal act on the use of AI by law enforcement agencies. Two controversial components that have drawn the attention of the Council of the EU and the European Parliament are remote biometric identification and emotion recognition technologies. While the Council’s general approach aligns with the Commission’s proposal to balance law enforcement and human rights protection, the European Parliament pursues a narrower approach, advocating for the prohibition of real-time remote biometric recognition and emotion inference applications. It goes without saying that the outcome of the ongoing inter-institutional negotiations (trilogue) between the EU co-legislators and the Commission is being anticipated by law enforcement bodies with considerable interest. After all, this will define how the opportunities provided by AI are leveraged in law enforcement settings as well as how to deal with the misuse of this evolving technology by terrorists and criminals. This article reports on the institutions’ positions on remote biometric identification and emotion recognition and highlights the – in the authors’ view – flawed approach by the European Parliament toward law enforcement.