The subject “Artificial Intelligence (AI) and justice” can be addressed from two different perspectives: AI as the object of judicial proceedings, from one side, and AI as a tool supporting judges in the exercise of jurisdiction, from the other side.

It is not difficult to foresee that, in the future, an increasing number of disputes will regard the use of AI systems. This is the case, for instance, of claims for damages caused by driverless cars, drones or automated disease diagnosis and treatment systems. In such cases, the main issue is whether, and to what extent, consolidated legal principles on the law of evidence, on damages quantification and on liability, which traditionally refer to human behaviours, can be extended to robotic behaviours.

The use of AI systems may also trigger a different set of issues, when used to assist judicial authorities in exercising jurisdiction. Nowadays, new automated tools for due diligence exercises, for drafting documents and for technical assessments, including calculation of maintenance allowances for spouse or children or damages in the event of personal injury, are available on the market. Law firms and insurance companies increasingly use predictive AI systems to determine the possible outcome of a current or potential legal dispute. Why not using the same tools, then, to increase the efficiency of the judicial system? As a matter of fact, the use of AI systems may help to increase the quality and efficiency of justice. At the same time, however, the use of AI in the judicial field raises a set of new and open questions.

Below, I will address this second perspective. I will try to bring some thoughts on the opportunities and risks deriving from the use of AI in the justice domain.