AI-Enabled Optimization of Early-Phase Clinical Trials Pilot Program; Request for Information
The Food and Drug Administration (FDA or the Agency) is issuing this request for information to solicit input on a proposed pilot program to assess how artificial intelligence (AI)-enabled technologies can improve efficiency, speed, and quality of decision- making in early phase clinical trials. Early-phase clinical trials represent a critical bottleneck in drug development, often characterized by high uncertainty, limited patient populations, and inefficient decision- making processes. This pilot program aims to explore how advances in AI and data science can improve trial efficiency, enhance safety monitoring, facilitate dose selection decisions, and enable more informed early go/no-go decisions (e.g., a regulatory decision as to whether a Phase 1 study may proceed) while maintaining FDA's rigorous scientific and regulatory standards and promoting trustworthy AI systems. The pilot program will be guided by principles aligned with the National Institute of Standards and Technology (NIST) AI Risk Management Framework (AI RMF).
What this rule actually says
The FDA is running a pilot program to see if AI can help drug companies run early-stage clinical trials faster and safer. Right now, these trials are slow bottlenecks in drug development—they involve small patient groups, lots of uncertainty, and slow decision-making. The FDA wants to know: can AI tools help with dose selection, safety monitoring, and go/no-go decisions while keeping trials rigorous and trustworthy?
Who it applies to
- If building AI for clinical trial operations: You're in scope if the AI tool directly supports decisions in Phase 1 or early Phase 2 trials (dose selection, safety flagging, patient monitoring, trial go/no-go recommendations).
- If building general medical AI that doesn't touch trials: Medical scribes, hiring assistants, support chatbots, and diagnostic tools operating outside clinical trial settings are not affected by this.
- Jurisdiction: This is FDA (U.S. federal), so it only applies to AI systems supporting trials regulated by the FDA or conducted in the U.S.
- Data scope: The program focuses on trial decision-making data (safety signals, dosing outcomes, patient eligibility), not patient data in general. If your AI never touches clinical trial workflows, you're out.
What founders need to do
- Determine if you're building for early-phase trials (1 day). Ask: does my AI directly influence Phase 1/early Phase 2 trial decisions? If no, stop here. If yes, continue.
- Monitor the pilot program (ongoing, 2-3 hours/quarter). The FDA is requesting public input until [deadline TBD]. Sign up for FDA regulations.gov updates or check back quarterly. This is still a "request for information"—nothing is mandatory yet.
- Review NIST AI Risk Management Framework basics (2-3 days). The pilot will use NIST's AI RMF principles. Skim the framework's overview to understand safety, explainability, and documentation expectations early.
- Prepare documentation if you apply to the pilot (1-2 weeks, optional). If you want to participate, expect to document: how your AI works, what safety guardrails exist, how decisions are logged, and how you'll validate accuracy in trial settings.
- Get legal review before submitting anything (3-5 days). If you decide to participate, have a healthcare regulatory lawyer review your submission. FDA submissions carry compliance weight.
Bottom line
Monitor for now, but act only if you're building AI that makes clinical trial decisions—and only if the FDA formally launches the pilot and you want to join it.