This Article examines the potential use of large language models (LLMs) like ChatGPT in constitutional interpretation. LLMs are extremely powerful tools, with s

Exploring, analyzing, and interpreting law can be tedious and challenging, even for legal scholars, since legal texts contain domain-specific language, require knowledge of tacit legal concepts, and are sometimes intentionally ambiguous. In related, text-based domains, Visual Analytics (VA) and large language models (LLMs) have become essential for working with documents as they support data navigation, knowledge representation, and analytical reasoning. However, legal scholars must simultaneously manage hierarchical information sources, leverage implicit domain knowledge, and document complex reasoning processes, which are neither adequately accessible through existing VA designs nor sufficiently supported by current LLMs. To address the needs of legal scholars, we identify previously unexamined challenges and opportunities when applying VA to jurisprudence. We conducted semi-structured interviews with nine experts from the legal domain and found that they lacked the ability to articulate their tacit domain knowledge as explicit, machine-interpretable knowledge. Hence, we propose leveraging interactive visualization for this articulation, teaching the machine relevant semantic relationships between legal documents. These relationships inform the predictions of VA and LLMs, facilitating the navigation between the hierarchies of legal document collections. The enhanced navigation can uncover additional relevant legal documents, reinforcing the legal reasoning process by generating legal insights that reflect internalized, tacit domain knowledge. In summary, we provide a human-is-the-loop VA workflow for jurisprudence that recognizes tacit domain knowledge as essential for deriving legal insights. More broadly, we compare this workflow with related text-based research practices, revealing research gaps and guiding visualization researchers in knowledge-assisted VA for law and beyond.

Use of legal fictions in public law has shown no sign of abating, as demonstrated in the enactment of the Safety of Rwanda (Asylum and Immigration) Act 2024. Ac

At the start of the new legal year, literally tens of thousands of professional lawyers and their law firms from across the English-speaking world are

Échange avec Stéphane Cottin, chef du service de la documentation et de l’aide à l’instruction du Conseil constitutionnel, autour de son ouvrage "Recherche documentaire juridique – Méthodologie" (6e édition), co-écrit avec Vincent de Briant.

Continuons notre petite série de vidéos dédiée aux ouvrages juridiques. Avec une découverte de la 6e édition de l’ouvrage « Recherche documentaire juridique – Méthodologie » par MM. Stéphane …

A foundational principle of the law is that decision-makers must explain their reasons: judges write opinions, government agencies write reports detailing why t

This article examines the impact of generative AI (GenAI) on the legal profession, focusing on the rise of AI-generated fake and inaccurate citations in l

This document reports research conducted from December 2022 to August 2024, and in particular, Part I experiments conducted from May 20 to July 12, 2024, and Pa