A Case Study on The Use of Google NotebookLM in Enhancing Legal Decision-Making for Procurement and Logistics

Abstract

Managing regulatory compliance in logistics and procurement is a recurring problem that frequently requires the tedious, error-prone, and manual examination of complex legal documents. In order to solve these problems, this thesis explores the possibilities of Google NotebookLM, a source-grounded large language model. Through a cost benefit analysis, the study seeks to assess NotebookLM's effectiveness and precision in assisting with compliance tasks such tax verification, customs documentation, and contractual regulation. This thesis evaluated how the tool increases productivity by cutting down on time spent on sourcing regulations (RQ1) and how much it reduces errors connected to compliance (RQ2). Additionally, the study had pinpointed the main expenses and difficulties related to its integration, such as data security and training (RQ3).

This thesis finds that the tool provides significant efficiency gains, cutting contract review time by up to 50% and enabling teams to vet 50% more suppliers. Furthermore, Google NotebookLM substantially reduces high-cost compliance errors, such as those related to wrong HS code declarations. While integration presents challenges, particularly regarding staff training and data security, the overall QCBA confirms a positive net value. The highest return on investment (ROI) is found in Validating Incoterm Clauses and Contract Review, demonstrating that AI offers a beneficial solution for improving day-to-day regulatory operations.