In the highly regulated pharmaceutical industry, ensuring compliance with advertising guidelines is a critical yet complex task. This paper introduces a novel, AI-driven approach to automate compliance verification in pharmaceutical advertising, addressing the challenges of efficiency, accuracy, and consistency in the current manual processes. We present a multi-agent system that leverages advanced computer vision, natural language processing, and machine learning techniques to decompose advertising assets and verify their compliance with industry-specific regulations. Our system comprises two primary AI agents: (1) an Asset Decomposition Agent that employs few-shot learning, computer vision, and optical character recognition to break down complex advertising materials into analyzable components, and (2) a Compliance Verification Agent that interprets regulatory guidelines, transforms them into verifiable conditions, and assesses the decomposed asset data against these conditions. This approach not only significantly reduces the time and resources required for compliance checking but also improves accuracy and consistency. We evaluate our system on a diverse dataset of pharmaceutical advertisements, demonstrating its effectiveness across various media types and regulatory scenarios. Our results show a 95% accuracy in compliance verification, with a reduction in processing time from days (and sometimes weeks) to minutes compared to manual methods. This research contributes to the growing body of work on AI applications in regulatory compliance and demonstrates the potential for intelligent systems to address complex, domain-specific challenges in highly regulated industries. The proposed approach has significant implications for improving patient safety, reducing regulatory risks, and enhancing the efficiency of pharmaceutical marketing processes.