Artificial Intelligence as a Subject of Criminal Law: A Corporate Liability Model Perspective
Edita Gruodytė, Paulius Čerka · 2020
“Criminal law generally is retrospective and individual-oriented and not suit-able to direct the development of technologies” (Beck 2016, 141). The content of criminal liability depends on the norms of social behavior prevailing at that time (including moral and ethical norms; Drakšas 2008, 11). Criminal law is cur-rently based on the protection of human rights and freedoms as well as society and state interests (Wessels 2003, 30) and consists of three main components: the human being, the society, and the state (Leroux et al. 2012, 51). In essence, a criminal offense is a violation of legal good and duty. It could be argued that “the illegal content of a criminal offense is revealed through the harmfulness of the consequences of the activity (violation of a protected object and encroach-ment upon it) and harmfulness of actions (the nature and manner of the com-mitted actions)” (Wessels 2003, 30). One important criteria that distinguishes criminal liability from other branches of law is that the aim in criminal law is not compensation of damage, but punishment of the perpetrator by state authorities (Leroux et al. 2012, 51), since it could be argued that, for certain egregious offenses, there should be also condemnation of a person on behalf of the state. If artificial intelligence (ai) committed a criminal offense, could aibe treated as a criminal? This question is especially relevant given recent soci-etal transformations and the “changing nature of society’s perception of, and relationship to, crime” (Keiler and Roef 2019, 37). The previous conception of the welfare state shifted the risk to society as a consequence of “modern indus-trialized and globalized societies” confronted with new risks such as modern technologies (Keiler and Roef 2019, 38).[...]