The disconnect between the goals of trustworthy AI for law enforcement and the EU research agenda

The disconnect between the goals of trustworthy AI for law enforcement and the EU research agenda

연구 분야: Strategies

학회: AI and Ethics

In this paper, we investigate whether AI deployment for law enforcement will enable or impede the exercise of citizens' fundamental rights by juxtaposing the promises and policy goals with the crude reality of practices, funded projects, and practicalities of law enforcement. To this end, we map the projects funded by H2020 in AI for law enforcement and juxtapose them to the goals and aims of the EU in terms of Trustworthy AI and fundamental rights. We then bring forward existing research stressing that AI implementation in sensitive domains such as defense and law enforcement does not come without drawbacks, especially regarding discrimination, surveillance, data protection, and human dignity. We thoroughly analyze and assess human-centric and socially-driven lens risks and threats of using AI factors from an ethical, legal, and societal perspective (ELSA), including organizational and gender worries.