Mission
The Justice and AI Jean Monnet Centre of Excellence: Effective Judicial Redress in the Rising European and Global AI litigation (JUST-AI JMCE) strives to become a Belgian and European ‘hub of excellence’ on issues pertaining to procedural fairness in the field of AI. With the participation of a network of Belgian and international experts on AI, and in view of reaching out to expert and non-expert audiences around the world, the activities launched under the Jean Monnet Centre of Excellence label aim at promoting innovative, pluri- and interdisciplinary research, creating original and interactive educational content and raising awareness on the most salient current and future issues on procedural fairness in the era of AI. The Centre’s activities will, in particular, focus on two research axes:
1. AI liability and fair trial requirements
In recent years, we have seen a rise in the number of judicial instances where courts were called to apply existing procedural frameworks on evidence to the proof of harm and causation in cases involving AI systems. Unsurprisingly, these cases revealed never-before-seen procedural difficulties in the administration of evidence and allocation of liability, namely because the apparent author of a harm is often an AI system having acted with little or no human supervision. In AI liability cases, an important practical difficulty that litigants experience when debating harm and causation is the access to evidence. The topical examples of algorithmic biases (recruitment, credit scoring, recidivism, access to education) show that harm-causing automated decisions often result from ‘black box’ decisional processes, inscrutable and undiscernible by the victims, users and even programmers. This inability to access and give evidence affects litigants’ exercise of constitutional procedural rights namely, the right to access courts, the right to defense, the equality of arms and the contradictory debate principle.
One of the key ambitions of JUST-AI JMCE is to develop research and teaching focused on the ways in which the administration of evidence - and by that, of justice - in AI liability disputes can be made more effective, thus upholding a standard of judicial protection that is not lower than that applied in ‘ordinary’ disputes (i.e. disputes not involving AI technologies).
2. Regulation and procedural capabilities of litigants in AI-related disputes
Does (EU) law establish the adequate regulatory frameworks for AI programming, deployment or use? This is the question raised, within the Centre, in connection to the regulation of AI, in particular in the EU. With procedural fairness as fil rouge, the main focus is placed on whether any existing or forthcoming AI regulation supports the litigants’ procedural abilities, allowing them to have an ‘effective shot’ at a fair adjudication of AI liability disputes. On the basis of a well-developed ‘in house’ expertise on the risk/rights-based approach to regulation of AI, the JUST-AI JMCE seeks to go further, by opening a new subfield of research and teaching, meant to explore and uncover if national and European procedural frameworks (AI Liability Directive (COM(2022) 496 final, Revised Product Liability Directive (COM (2022) 495 final) take into account the litigants’ concrete needs to access and give specific evidence, to receive and give specific explanations, to effectively make their views known before a court and respond to their adversary’s arguments. In addition to coordinating various teaching and research activities on the procedural aspects of AI-related adjudication, the JUST-AI JMCE seeks to gather the views from experts and non-experts alike, by launching a global public consultation within the PROXAI project the aim of which is to assemble data and conceptualize the procedural needs that potential litigants flag as necessary for the purpose of effective and fair unfolding of trials dealing with the proof and allocation of AI liability.
