AI success in R&D depends less on ambition and more on focus. This session explores how leaders can prioritise fit-for-purpose AI applications by grounding decisions in AI-ready data foundations, clear scientific and operational value, and organisational readiness. Drawing on real R&D experience, it outlines practical criteria for selecting fit-for-purpose AI applications that scale into workflows, improve scientific and operational decisions, and ultimately accelerate development and improve outcomes for patients.

AI adoption in clinical trials is less about technology and more about leadership. It is fundamentally about leveraging digital innovation to amplify human capabilities. This session tells the story of introducing AI into Study Leadership—navigating skepticism, GxP concerns, and operational realities. Through practical examples, it demonstrates how study leaders moved from cautious observers to confident users. We will explore the critical framework required to establish AI as a trusted, auditable, and value-adding asset in trial execution, supported by robust guardrails, coaching, and peer champions.

Industry dialogue has shifted from AI experimentation to operational readiness. Drawing from enterprise deployment experience, this session examines strategic approaches for scaling agentic AI in clinical trials—autonomous systems supporting risk identification and decision-making. Key topics: establishing robust data foundations, implementing governance frameworks, navigating regulatory requirements (EU AI Act, FDA guidance), managing organisational change, and measuring business impact beyond proof-of-concept.

Roundtable bullet points:

• Why do most AI pilots fail to scale beyond proof-of-concept?
• Can you build agentic AI on fragmented legacy data—or is infrastructure investment non-negotiable?
  
• Is regulatory compliance (EU AI Act, FDA guidance) a barrier to innovation or a competitive moat?
• How do you measure AI ROI when most value claims are declarative, not demonstrated?
  
• What's harder: deploying the technology or changing organisational behavior to trust it?
  

AI governance looks different depending on where you sit in a clinical organisation. This panel brings together three senior leaders from Eli Lilly's Clinical Development and Design Hub, Investigator Engagement, and Trial Capabilities functions to examine how each built AI readiness from the ground up — with shared principles but function-specific challenges. Attendees will hear a rare single-sponsor, multi-function perspective on what governance, adoption, and operational sustainability actually require across a global clinical organisation.

Join us for an interesting session jointly delivered by three roles across Clinical Development: Global Regulatory, Clinical Operations, and AI Governance for a story about how we in Novo Nordisk are enabling AI to assist with critical decisions in pivotal studies which only can succeed when people, processes, and systems are aligned through integrated workflows, training, and robust controls. When successfully implemented, AI not only strengthens governance and decision quality, but also unlocks meaningful operational efficiencies— accelerating trial execution throughout the study life cycle. 

This presentation outlines Bayer’s journey toward AI governance and operational readiness, from early automation to enterprise level GenAI adoption. It highlights key lessons in organisational enablement, balancing internal and external capabilities, and an example of a holistic approach to transforming regulatory submissions. Attendees will gain practical patterns and pitfalls to help move from isolated experiments to scalable, well-governed AI ecosystems.

This session presents a comprehensive framework for implementing managed, reusable data products in clinical development. Learn how to combine business purpose, semantic standards, technical delivery, and governance into AI-ready assets accessible through standardised interfaces. Discover practical approaches to data contracts, quality rules, lifecycle management, and regulatory compliance that enable consistent, automated decision-making across R&D.

ClinicalAIDEN is an agentic AI–powered decision support system that transforms how clinical trial feasibility and study design insights are generated, evaluated, and translated into strategic options. By orchestrating AI agents across data aggregation, analysis, and scenario evaluation, the system reduces manual effort, shortens insight cycles, and enables teams to explore multiple, evidence-backed development options earlier. This leads to faster, more robust, and better-aligned R&D decisions.

This study presents an AI-driven engine that automates drafting clinical site queries, ensuring quality and reducing redundancy. The system uses context-aware templates, semantic de-duplication, integrity checks, and recursive validation to produce accurate, consistent, and actionable queries. Results show improved query quality, faster responses, and enhanced data integrity, demonstrating a scalable solution for efficient, high-fidelity communication in clinical monitoring.

Role-based AI “Virtual Teams” can generate exploratory clinical analyses by coordinating analysis planning, statistical coding, and narrative summaries. Because generative models are non-deterministic, repeated executions can be treated as stochastic samples from a distribution of plausible analytical ideas. Specifically, we will describe how this AI workflow repeats its processes, compares results across multiple runs, and provides the end user with convergence and variability assessments to enable efficient human-in-the-loop validation and accountability.