This research investigates the use of advanced artificial intelligence (AI) technologies including Llama-Parse, Retrieval-Augmented Generation (RAG), and large language models (LLMs) to automate andenhance compliance with European Union (EU) product regulations. Traditional compliance processesare manual, time-consuming, and error-prone, posing particular challenges for small and medium-sizedenterprises (SMEs) lacking in-house legal expertise. To address this, develop and evaluate an AI-drivencompliance advisor system capable of interpreting complex regulatory texts and providing actionablecompliance insights.The system integrates LlamaParse for high-fidelity document parsing, FAISS for efficient vector-basedsemantic search, and GPT-4 Turbo for reasoning and report generation. Through a structured pipeline,it processes regulatory documents, retrieves relevant clauses, and generates compliance assessments tailoredto product descriptions. The study follows the Design Science Research Methodology (DSRM),with iterative development and validation in collaboration with Chemity AB, a Swedish compliance consultancy.Key findings show that the AI system achieves 89% accuracy and 95% completeness in regulatory analysis,significantly outperforming non-experts and approaching expert-level performance. It reduced errorsby 51% and accelerated document processing by 42–63% compared to human reviewers, yielding substantialefficiency gains. While 82% of users rated the system valuable for business applications, a 7%hallucination rate underscores the need for expert oversight to ensure legal precision. Overall, the findingshighlight the potential of AI-powered compliance systems to streamline regulatory workflows, improveefficiency, and enhance accuracy, with particular promise for SMEs navigating the complexities of EUproduct regulations.