Identifying legal norms on legal entities using Artificial Intelligence: Legal system fragmentation and the potential of the LEI Identifier

Abstract

This article presents an innovative approach to legal analysis through the application of artificial intelligence (AI) tools – particularly large language models (LLMs) – to the identification of legal norms concerning legal entities. The study aimed to map and assess the regulations that define and identify legal entities within the Polish legal system, with a specific focus on the global Legal Entity Identifier (LEI). The introduction discusses the problem of system fragmentation in legal entity identification: various branches of law and jurisdictions rely on different definitions and registries, which hampers effective risk management and economic transparency. The literature review highlights the predominance of traditional doctrinal methods in legal scholarship, while noting the growing role of empirical approaches using computational tools, including AI. The methodology combines three perspectives: doctrinal legal analysis (textual examination of legal acts), functional analysis (assessment of whether existing norms fulfill the identification function), and empirical analysis using AI (automated extraction and processing of legal provisions related to legal entities). The findings reveal a high degree of regulatory fragmentation – multiple, inconsistent definitions and identification requirements for legal entities were found across legal acts. At the same time, the LEI is shown to have significant potential as a universal standard capable of integrating dispersed identification regimes and enhancing legal coherence. The conclusion includes de lege ferenda recommendations, such as extending the mandatory use of the LEI beyond the financial sector in both national and EU law, harmonizing legal terminology concerning legal entities, and incorporating AI tools into the legislative and legal application processes. The article argues that integrating innovative methods (AI/LLMs) with classical legal scholarship can significantly improve the analysis and refinement of legal systems.