Predictive Justice and Artificial Intelligence supporting judge’s decision-making in Genoa. Sant’Anna School and the Court of Genoa announce collaboration on computational analysis of the rulings
Sant’Anna School researchers will survey the Genoa judges implementing the open data policy for judicial decisions. The analysis processing case-law by algorithms is the big data processing system currently used at the Dirpolis Institute LIDER-lab (observatory on personal damages, tort and liabilities) to highlight models, correlations and trends. This processing method (machine learning, data science) for the analysis of trends of decisions, also predicting court decisions, will identify existing discrimination, through grouping or classifying data relating to individuals or public and private stakeholders.
Particular care must be taken when the processing is directly or indirectly based on sensitive data including racial or ethnic origin, socio-economic background, political opinions, religious or philosophical beliefs, trade union membership, genetic data, biometric data, health-related data or data concerning sexual life or sexual orientation. Consideration must be given to corrective measures to limit or neutralize the risks and as well as to awareness raising among stakeholders.
The use of machine learning and multidisciplinary scientific analyses to combat discriminations in the general complexity of legal systems has highlighted that judicial reasoning is above all a matter of assessment and interpretation. Researchers have stressed that coherent arguments and legal syllogism can lead to different judgments. Algorithms predicting a risk associated with a certain defendant’s behavior built upon historic data with higher levels of intolerance could reflect the law enforcement disproportional targeting.
Therefore, researchers will focus on training to uncover the bias and dirty data in a regular decision-making so that the legal professionals become more conscious about the use of discriminatory language or fraudulent data. A higher bias awareness, combined with a basic understanding of the machine learning processes, will improve the use of AI models in decision-making process.
The objective of the method is to create a decision-making tool in order to reduce excessive variability in court decisions, in the name of the principle of equality of citizens (and workers) before the law. It will concern civil litigation, abuse at workplace, compensatory allowance and severance pay, divorce proceedings, compensation for personal injury, and small disputes, inter alia.
Public debate involving legal professionals, legal tech companies or scientists, is focusing on possible impact of artificial intelligence to predict judges' decisions and devise the ethical framework in which it must operate. The debate comes from the private sector involving citizens wanting to reduce legal uncertainty and the unpredictability of judicial decisions.
Genoa press conference comments section:
Enrico Ravera, President of the Court of Genoa: “We are happy to accept innovative technologies and the use of machine learning in decision-making processes, including judicial practice. Artificial Intelligence is one technology that shows a great impact on the individual’s personal and professional life”.
Domenico Pellegrini, President of the Civil Division - Genoa: “Predictive justice offers a tool as regards anticipating and harmonizing court decisions. It will help judges to focus on cases in which their expertise is of greater added value”.
Giovanni Comandè, LIDER Lab: “46 years ago we started to collect case law data. Thanks to the legal technology tools, we are now able to offer statistical analyses, which can be used to predict the chances of success in a particular case. Lawyers can adapt their strategy and even their arguments to be in line with formulations and language rules that have proven effective in the past”.
Denise Amram, LIDER Lab: “The justice of the future, predictive justice and artificial intelligence are at the forefront of the debate on the functioning of judicial systems. Through a multidisciplinary approach judges can perform consistently and render fair decisions making the AI system ethical”.
Cover photo: from left to right, Denise Amram, Giovanni Comandè, Domenico Pellegrini, and Enrico Ravera.