Description

Validation of Regulated Computers that Use AI

How to Apply Validation Concepts to Applications that Use AI

This seminar is focused on Artificial Intelligence (AI), Machine Learning (ML), and Large Language Models (LLMs) used in FDA-regulated computer systems. We will discuss these newer technologies and how they are being applied in both the FDA and industry.

There will be a discussion about the current FDA framework for compliance of AI-enabled systems and how it is evolving. We will cover specific areas of FDA compliance and enforcement, in general, and how these apply to AI-enabled systems regulated by FDA.

This seminar will provide insight as to how to validate software applications that use AI. We’ll discuss the broader Computer System Validation (CSV) and Computer Software Assurance (CSA) landscapes, indicating how the approaches to validation can be leveraged for software using AI. We will discuss GAMP 5, 2nd Edition and how it aligns with the concepts of CSA.

We’ll cover the key FDA requirements for validation that must be met regardless of the underlying technology. These include planning, requirements, testing, traceability, and documentation.

We’ll discuss test acceptance criteria, risk management and controls, as the practitioner will need to take a risk-based approach to validation of AI-enabled systems. The FDA 21 CFR Part 11 guidance requirements for electronic records/electronic signatures, and data integrity will also be covered as these are key elements to be tested/verified during validation.

The shortcomings of AI-enabled software will be discussed, along with the role of the Expert-in-the-Loop who uses critical thinking skills and operational expertise to ensure AI outputs are reviewed and judged, leading to good decisions and actions. We will discuss critical thinking and how this has become essential to validation of AI-enabled systems. We will also cover how this became a “lost art” through recent decades as rote behavior became the norm in FDA-regulated operations.

Finally, we will discuss current industry best practices for AI-enabled validation. Overall, the seminar will position the validation practitioner to be able to use AI-enabled systems effectively and in compliance with regulatory requirements.

Agenda:-

September 23, 2026 ( 9:00 AM EST - 5:00 PM EST ) - DAY 1

• 9:00 am - AI Use by FDA and Industry
• 9:30 am -  Current State of FDA Compliance Guidance for AI-Enabled Systems
• 10:00 am - Current Trends in FDA Compliance and Enforcement
• 10:30 am - 10-Minute Break
• 10:40 am - Risk-Based Approach to Validation Compliance: CSV, CSA, and GAMP 5, 2nd Edition
• 12:00 pm - 1-Hour Lunch Break
• 1:00 pm - 21 CFR Part 11 and Data Integrity Compliance
• 2:00 pm - Vendor Audit
• 2:20 pm - Inspection Readiness
• 2:50 pm - 10-Minute Break
• 3:00 pm - AI Shortcomings: The Role of the Expert-in-the-Loop, and Critical Thinking
• 4:30 pm - Industry Best Practices: Validation of AI-Enabled Systems
• 4:45 pm - Q&A

Learning Objectives:-

• Understand how AI is currently used by FDA and industry
• Learn about current trends in FDA compliance and enforcement related to use of AI in GxP systems
• Learn how to validate software that uses AI
• Understand the traditional approach to validation, Computer System Validation (CSV)
• Learn about the newer approach to validation, Computer Software Assurance (CSA) and the benefits and risks of following it
• Learn about GAMP5, 2nd Edition and how it aligns with FDA’s CSA approach
• Understand how to plan, test, and document computer system validation activities
• Learn how to be compliant with 21 CFR Part 11, FDA Guidance for Electronic Records and Electronic Signatures
• Learn how to be compliant with FDA’s 2018 Data Integrity Guidance
• Understand the shortcomings of AI and need to compensate for them
• Understand the difference between validating deterministic systems versus AI-enabled systems, which are probabilistic
• Recognize warning signs of hallucinations, bias, drift, and unreliable AI behavior
• Learn about the Expert-in-the-Loop role as the human guardrail for AI
• Learn about how to safely and effectively use AI in regulatory operations
• Identify where human review and the Expert-in-the-Loop are essential
• When testing AI-enabled software, understand defensible AI acceptance criteria aligned to patient safety, product quality, and data integrity
• Learn the importance of monitoring in maintaining AI-enabled software systems in a validated state over their lifetime
• Apply risk-based critical thinking to AI oversight and validation activities
• Understand how to think critically and why this has become a “lost art” in regulated industry
• Reduce regulatory exposure associated with unchecked AI output
• Build greater trust in AI-assisted GxP processes without creating excessive documentation burden
• Prepare for increasing regulatory scrutiny surrounding AI usage in life sciences
• Learn about industry best practices applicable to AI-enabled system validation

Areas Covered:-

• Artificial Intelligence (AI), Machine Learning (ML), and Large Language Models (LLMs)
• Use of AI-Enabled Systems by FDA and Industry
• Current FDA Guidance, Compliance, and Enforcement Trends
• Computer System Validation (CSV)
• Computer Software Assurance (CSA)
• Validation Planning, Testing, and Maintenance
• Validation Documentation
• Risk Identification, Assessment, Mitigation, and Acceptance
• GAMP5, 2nd Edition and Alignment with CSA
• FDA 21 CFR Part 11, Guidance for Electronic Records and Electronic Systems
• FDA Guidance for Data Integrity (ALCOA+ principles)
• Vendor Audit of Cloud-based Systems
• AI Shortcomings – Hallucinations, Bias, Performance Drift
• Expert-in-the-Loop Role in Validation
• Critical Thinking
• Inspection Readiness
• Industry Best Practices: AI-Enabled Software Validation

Background:-

Artificial Intelligence is rapidly being embedded into regulated GxP operations — from quality systems and manufacturing support to clinical, laboratory, and regulatory processes. But one question continues to challenge industry and regulators alike. AI-enabled software and tools must be reliable, but what does that actually mean when used in a regulated environment?

Traditional validation approaches were designed for deterministic systems that consistently produce the same output. AI introduces probabilistic behavior, evolving models, hallucinations, bias, performance drift, and opaque decision-making that require a fundamentally different way of defining acceptance criteria.

This session explores how life sciences organizations can establish a meaningful, risk-based approach to validation of AI-enabled systems in compliance with regulatory requirements and expectations. Attendees will learn how to move beyond checkbox validation and instead define measurable standards for trust, reliability, oversight, explainability, and human review within AI-driven GxP processes. This requires a new role, the Expert-in-the-Loop, who uses critical thinking. This will be described in relation to validation and use of AI-enabled systems.

The discussion emphasizes the role of risk management and the Expert-in-the-Loop as the critical human guardrail for AI-enabled decision making.

Why Should You Attend:-

Attendees will gain practical insight into one of the most urgent and misunderstood challenges in regulated AI adoption:

How do you prove an AI system is fit for intended use when outputs may vary?

Participants will learn how to understand how AI is being used by FDA and industry. In particular, FDA is studying how industry uses AI as the applicable regulatory framework is evolving. AI is already being used, and we will discuss some of the trends in how FDA is evaluating the use of these systems and what that means for industry.

AI-enabled systems must be validated, and we will cover the important aspects of both traditional validation, CSV, and the newer approach to validation, CSA. We will discuss how GAMP 5, 2nd Edition aligns with AI.

The attendee will be walked through the validation process, including planning, requirements, testing, and documentation. We will cover testing and acceptance criteria for validating AI-enabled systems with testing that will meet regulatory scrutiny. The seminar will teach the attendee how to build greater trust in AI-assisted GxP processes without creating excessive documentation burden.

The importance of continuous monitoring of AI-enabled systems will be discussed to ensure inspection readiness by maintaining a validated system in a state of validation through its entire lifecycle. We will also discuss the approach to vendor audit for this newer technology.

The attendee will learn about 21 CFR Part 11 and data integrity compliance for systems using AI. We will also cover the approach to be taken to audit vendors of AI-enabled systems in a cloud environment.

AI shortcomings will be discussed, including hallucinations, bias, and performance drift. These characteristics of the AI “problem child” require an Expert-in-the-Loop role to be played by a human with the appropriate operational expertise and critical thinking skills.

Critical thinking will be discussed as a key requirement for the Expert-in-the-Loop, and how this differs from the current state of doing validation by rote, following a checklist mentality.

The FDA regulatory framework is still evolving, and there will be changes coming. We will discuss several new draft guidance documents from FDA.

In the interim, we will discuss validating and maintaining AI-enabled systems inspection ready by focusing on the key components of CSV and CSA that apply. This will prepare the attendee to validate AI-enabled systems that are inspection ready.

This session is especially valuable for organizations actively evaluating or implementing AI tools in regulated environments.

Who Should Attend:-

This full-day seminar is designed for leaders and practitioners involved in regulated computerized systems, digital transformation, AI governance, and quality oversight, including:

• Information Technology (IT) System Analysts
• Business Analysts
• IT Developers and Testers
• IT Maintenance Professionals
• QC/QA Managers and Analysts
• Analytical Chemists
• Compliance and Audit Managers
• Laboratory Managers
• Automation Analysts and Managers
• Manufacturing Specialists and Managers
• Supply Chain Specialists and Managers
• Regulatory Affairs Specialists
• Regulatory Submissions Specialists
• Clinical Data Analysts
• Clinical Data Managers
• Clinical Trial Sponsors
• Computer System Validation Specialists
• GMP Training Specialists
• Business Stakeholders/Subject Matter Experts
• Business System/Application Testers
• Business Stakeholders responsible for computer system validation planning, execution, reporting, compliance, maintenance and audit
• Consultants working in the life sciences industry who are involved in computer system implementation, validation and compliance
• Auditors engaged in internal inspection
• This seminar will also benefit any vendors and consultants working in the life sciences industry who are involved in AI-enabled computer system implementation, validation and compliance. It will also help those in software development companies who support the life science industries with AI-enabled tools and software.