Abstract: South Africa has undergone legislative reform that is significant inthe areas of performers’ rights and copyright in recent years. Theline between legal protection and creativity is often blurred. TheCopyright Amendment Bill and the Performers’ Protection Amendment Billsparked waves of public commentary and have generated a wide range ofpublic submissions from stakeholders in the creative industriesincluding the music industry, yet the depth and meaning of thesesubmissions are often lost in legal language. These submissions arerepresentative of an underutilized and valuable data source that canbe transformed into digital tools that support legal practitioners,music artists and policymakers. These submissions can be transformedinto structured and searchable insights. This project aims to leverageNatural Language Processing (NLP) techniques for the analysis of thesepolicy and legal texts by making use of models suitable forclassification such as LegalBERT, Logistic regression and randomforest. The goal is to build a tool that identifies the issues and keyactors raised during the legislative process, evaluate how theseissues are addressed by current statutes and make this informationaccessible through the tool. The project aims to bridge the gapbetween public understanding and legal data by using data science toanalyze, classify and visualize stakeholder input on South Africanmusic law. For the final output, the user-interface tool must acceptlegal documents or submissions and identify stakeholder actors and keylegal issues. Furthermore, it must match them to legal statutes ortaxonomy categories and provide visual summaries and filtercapability. This contributes to the improvement of access to justice,empowerment of music industry participants and strengthens theinternational positioning of South African music law.