Abstract

This essay is intended to elaborate upon the impact of artificial intelligence on judicial decisions. The statement that I want to prove is that, while any unofficial use of artificial intelligence does not imply any significant change in the legal system, any institutionalized recourse to artificial intelligence (even as a simple «support» to decision-making) requires a major systemic change at constitutional level. This conclusion is argued from two different viewpoints. First, I will endeavor to show that procedural law requires that the ascertainment of facts has to be outwardly justifiable. In other words, it is irrelevant that the judge ascertains the truth or how he or she actually reaches his or her conclusions. All that matters is that such ascertainment can be objectively justified, which means that it is grounded on elements having an outward relevance such as the proofs or the rules governing the decision-making process. A decision reached through the aid of an algorithm does not have any of these features. Secondly, I will argue that the judge’s independence requires that his or her decision must not rely on third parties’ opinions. Therefore, any institutionalization of artificial intelligence support in the decision making process would be a limitation of the judge’s independence.