Trustworthy artificial intelligence and the European Union AI act: On the conflation of trustworthiness and acceptability of risk

Abstract

In its AI Act, the European Union chose to understand trustworthiness of AI in terms of the acceptability of its risks. Based on a narrative systematic literature review on institutional trust and AI in the public sector, this article argues that the EU adopted a simplistic conceptualization of trust and is overselling its regulatory ambition. The paper begins by reconstructing the conflation of “trustworthiness” with “acceptability” in the AI Act. It continues by developing a prescriptive set of variables for reviewing trust research in the context of AI. The paper then uses those variables for a narrative review of prior research on trust and trustworthiness in AI in the public sector. Finally, it relates the findings of the review to the EU's AI policy. Its prospects to successfully engineer citizen's trust are uncertain. There remains a threat of misalignment between levels of actual trust and the trustworthiness of applied AI.