The AI Act as a Safeguard for Equality of Arms in Data-driven Criminal Investigations

Increasingly, law enforcement agencies are resorting to a new approach to policing, often referred to in the literature as data-driven or intelligence-led policing. This development is spurred by the increasing availability to law enforcement agencies of large amounts of data, as well as the development of sophisticated AI techniques for analysing this data. More recently, the use of these tools in a large number of cases resulting from the EncroChat operation has given rise to concerns about the right to a fair trial: the results of investigations in which these AI tools are used are difficult to assess in terms of reliability and lawfulness, and case law from the ECtHR provides only some clarity on how to address these issues. At the same time, the AI Act has, as one of its aims, the safeguarding of fundamental rights, including the right to a fair trial. In this chapter I will assess, using two examples of AI tools that could be or are employed in these investigations, whether the AI Act indeed lives up to this promise. I will argue that for two main reasons, limited scope and shortcomings in the individual rights provided, the AI Act is unlikely to solve most of the fair trial issues raised by the use of AI tools in this type of investigation.