XAI-LAW: A Logic Programming Tool for Modeling, Explaining, and Learning Legal Decisions

Explainability as a Core Feature: The tool automatically generates explanations for legal decisions by tracing the logical reasoning behind outcomes, making the judicial decision-making process more transparent and interpretable to both legal experts and potentially the public.

Systematic Handling of Legal Contradictions: The system is designed to identify and properly manage contradictions that arise when encoding criminal law articles, which is critical since real legal codes can contain conflicting provisions that require careful resolution.

Learning from Past Cases: The tool uses inductive logic programming to automatically generalize legal rules from previous judicial verdicts, potentially reducing manual work for legal experts while grounding new decisions in established precedent.

Support Rather Than Replacement: The system is explicitly framed as a tool to "support legal experts during the criminal trial phase" rather than replace human judgment, maintaining human oversight in consequential legal decisions while providing decision support and reasoning transparency.

Validation Against Real Outcomes: The model was tested and refined against actual previous verdicts from Italian criminal trials, suggesting the approach produces legally sound results grounded in real judicial practice rather than theoretical assumptions.