FDA Maps AI Drug Development to Existing Review Pathways
The U.S. Food and Drug Administration updated its guidance on May 1, 2026, outlining where sponsors should direct AI-related questions during drug development. Rather than creating a single approval route for AI tools, the FDA tied different AI applications to established regulatory programs based on their intended use.
Sponsors can request formal meetings to discuss AI use tied to a specific development program. When requesting a meeting, they should include the investigational new drug (IND) or pre-IND number if applicable and notify the relevant review team.
Five Engagement Pathways
- CDER Center for Clinical Trial Innovation (C3TI): For AI in late-stage trial design and conduct
- Complex Innovative Trial Design (CID) meeting program: For novel clinical trial designs
- Drug Development Tools (DDT) qualification: For AI-based tools used as biomarkers or outcome measures
- Digital Health Technologies (DHT) program: For AI-enabled digital health tools not tied to a specific drug program
- ISTAND pilot program: For additional support on specific development questions
The FDA also provides triage contacts to route inquiries to relevant review groups or subject-matter experts.
What This Means for Development Teams
The mapping reduces ambiguity about which FDA group to contact for technical feedback on validation, data provenance, and performance evaluation. Data scientists and clinical teams working on AI-based tools can now identify the correct pathway before submitting questions, potentially shortening feedback cycles in complex programs.
Different AI applications will follow distinct regulatory interactions. A tool intended for trial simulation may follow one pathway, while an AI-derived biomarker may qualify under a different program. This separation reflects the FDA's approach of fitting AI uses into existing review structures rather than creating blanket rules.
What to Watch
Practitioners should monitor for follow-up guidance on model evaluation metrics, transparency expectations, and example data packages. Public workshops or draft guidances on validation standards for AI-derived biomarkers and algorithm updates during product lifecycle could provide more detail on regulatory expectations.
For those working on AI tools in clinical development, understanding these pathways now can clarify which validation and documentation standards apply to your specific use case.
AI for IT & Development professionals building regulatory compliance systems or clinical trial platforms should review how these pathways affect tool architecture and data requirements. Data scientists working on drug development tools need to understand the model evaluation and transparency standards tied to each pathway.
Your membership also unlocks:
