Annex 22 and AI Governance for GMP: Human Review, Model Monitoring, and Change Control

Providing safe and effective FDA-regulated products is in the best interests of all those involved in their development, manufacturing, testing, and distribution. With newer technologies such as AI in the mix, it means opportunity for greater efficiency and efficacy, but also poses more challenges for companies that develop, test, and support software applications in the life science industries. AI can hallucinate, and show bias and performance drift. The human expert must be able to critically assess the output and make the judgment calls for better decisions, quality, and compliance, and to optimize ROI. In this webinar, you will learn just how AI, ML and LLMs, such as ChatGPT can increase efficiency and effectiveness of software development life cycle (SDLC) activities, enabling the delivery and support of computer solutions and new innovative products that will drive industry over the coming years. In particular, you'll learn about the need for human review by the expert who applies critical thinking to ensure better outcomes, decisions, quality, patient safety, compliance, and ROI. Artificial Intelligence (AI), Machine Learning (ML) and Large Language Models (LLMs), such as ChatGPT are beginning to find a presence at life science companies. While life science companies tend to lag behind other markets in using these technologies, they are catching up and we are seeing much more activity related to AI use in software applications used to develop, produce, test, and manage life science products with quality and compliance. As the pace of technological innovation and evolution becomes more intense, there is a critical need for computer system validation, 21 CFR Part 11 (Electronic Records and Electronic Signatures) compliance, and data integrity assurance to continue in environments where artificial intelligence (AI) and machine learning (ML) are becoming more prevalent. In addition, the European Union (EU) has issued Annex 22 for AI governance. AI can hallucinate, and show bias and performance drift. AI solutions need human guardrails — the experts who can evaluate output and, based on broad and deep expertise, make the right judgment calls for better outcomes and quality decisions. Without this safeguard, we may never achieve the great potential in ensuring quality, compliance, and real ROI from these technological innovations. We'll cover all key aspects of EU Annex 22 as it applies to Human Review, Model Monitoring, and Change Control. We'll provide an overview of validation, including CSV and Computer Software Assurance (CSA), and the latest GAMP®5, 2nd Edition that aligns with CSA. We'll walk you through the Software Validation and Maintenance approach that will bring clarity to what FDA is looking for in terms of compliance. Code: FDB2849