Moral and Legal Autonomy in the Era of Artificial Intelligence

The ethical principle of autonomy is one of the core principles in the field of applied ethics, from bioethics to ethics of artificial intelligence (AI). In this paper, we pursue an ethical inquiry into how AI systems can affect human autonomy according to a moral and legal perspective, that is, both in its moral dimension (as implicit endorsement) as well as in its legal one (as explicit consent). More specifically, after having defined the concept of individual autonomy from a moral and juridical standpoint as the human normative power of self-determination both in the moral sphere and in the legal one, and thus, better substantiating the AI ethics principle of autonomy as currently adopted in the field, we show how the design of novel AI systems, such as machine-learning and deep-learning algorithms, that widely rule the functioning of digital information and communication technology (ICT), can negatively affect both the preconditions of our moral and legal autonomy, by suspending intrinsic consent that individuals can express in order to endorse external information as a true motive of their choices and actions, as well as bypassing the legal conditions for a valid consent, as the external manifestation of the individual legal autonomy. We conclude by highlighting the ethical implications and risks of such phenomenon and proposing ethical and legal design practices to prevent or mitigate them.