Regulatory FAQs
How Unlearn's digital twin methodology meets regulatory standards — our approach, common questions, and the evidence behind it.
Regulatory
EMA & FDA
Our approach
The FDA's seven-step framework for AI credibility
At Unlearn, we follow the FDA's Seven Step Framework for AI use.
- 1. Define Question of InterestIdentify the scientific or regulatory problem the AI model addresses.
- 2. Define Context of UseDescribe how and where the AI model will be used.
- 3. Assess AI Model RiskEvaluate the influence of the model and consequences of incorrect predictions.
- 4. Develop Credibility Assessment PlanPlan how the model will be validated and evaluated.
- 5. Execute the PlanPerform validation activities and test performance.
- 6. Document ResultsSummarize validation results in a Credibility Assessment Report.
- 7. Determine AdequacyConfirm whether the model is credible for its intended use.
Will regulators allow me to use digital twins in my trial?
[Answer to be supplied by the regulatory team.]
Has the FDA reviewed or accepted Unlearn's digital twin methodology?
[Answer to be supplied by the regulatory team.]
Is PROCOVA accepted by regulators outside the US and EU?
[Answer to be supplied by the regulatory team.]
What documentation do I need to submit to regulators alongside digital twin results?
[Answer to be supplied by the regulatory team.]
Does using digital twins change my trial's regulatory risk profile?
[Answer to be supplied by the regulatory team.]
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
No regulatory resources match your search.
Regulatory documents
Date
Title
Source
View
Date
2022
EMA Qualification Opinion — PROCOVA
Prognostic covariate adjustment qualified for confirmatory analysis.
Source
EMA
Whitepapers & guides
Date
Title
Format
Download
Date
Jul 2026
AI in the Trial Analysis: Interfacing with Regulatory Guidance
Three uses of AI in clinical trial analysis: adding power to RCTs, enabling reduced sample sizes, and supporting single-arm trials, with regulatory considerations.
Format
Whitepaper
Date
Jun 2025
A Risk-Based Approach for Leveraging AI in Clinical Trials
How to apply the FDA's seven-step credibility framework to digital twins, and how PROCOVA increases statistical power while preserving trial validity.
Format
Whitepaper
Date
Feb 2024
How to Position Clinical Trials with Digital Twins for Regulatory Success
A framework for incorporating participants' digital twins in clinical trials using Unlearn's regulatory-qualified PROCOVA methodology.
Format
Whitepaper
Date
—
PROCOVA™ Handbook for the Target Trial Statistician
How PROCOVA uses historical data and prognostic modeling to reduce uncertainty in treatment effect estimates in Phase 2 and 3 trials.
Format
Handbook
From the blog
Date
Title
Publisher
Read
Date
Oct 8, 2026
Built on Regulatory Rigor: Advancing Digital Twins in Clinical Trials
Publisher
Unlearn
Date
Jul 28, 2026
Not Every Use of AI in Trial Analysis Carries the Same Risk
Publisher
Unlearn
Date
Jul 1, 2026
Our Response to the FDA's Pilot on AI in Early-Phase Trials
Publisher
Unlearn
Date
Dec 11, 2025
Unlearning the Old Ways: Reflections from the FDA-CTTI Workshop
Publisher
Unlearn
Date
Dec 4, 2025
Derisking Clinical Development with Digital Twins: Key Takeaways from the MRCT Center Webinar Series
Publisher
Unlearn
Date
Jan 9, 2025
Why the FDA's New AI Guidelines Matter for Clinical Research
Publisher
Unlearn
Publications
Date
Title
Publisher
Read
Date
May 2026
Regulatory Considerations for Using Artificial Intelligence Models to Reduce Sample Sizes in Registrational Studies
Publisher
arXiv
