Unlearn.AI builds AI-generated digital twins of clinical trial participants — forecasts of how each individual patient's outcomes would evolve — and uses them to sharpen design, interim, and readout decisions in drug development.
Unlearn.AI builds AI-generated digital twins of clinical trial participants — forecasts of how each individual patient's outcomes would evolve — and uses them to sharpen design, interim, and readout decisions in drug development.
AI-generated patient digital twins
Model-based forecasts of each participant's clinical outcomes, generated from baseline data already collected, that act as a virtual comparator or as additional information per enrolled patient.
PROCOVA design and analysis method
A twin-based covariate adjustment approach that Unlearn describes as EMA-qualified and FDA-supported, used to reduce sample size in randomized controlled trials while maintaining power.
Trial design scenario modeling
Build and compare design scenarios — endpoints, eligibility criteria, sample size, assumptions — so governance decisions are made with quantified tradeoffs instead of positions.
Interim and readout analysis support
Pre-specified twin-based analyses added to the statistical analysis plan before unblinding, plus twin-adjusted analysis delivered at interim and at readout.
강점 및 장점
+Digital twins are generated from data already collected, so no extra patients are required to gain information
+Publicly reported, quantified outcomes — sample-size reductions, cost savings, and enrollment time saved — rather than only qualitative claims
+
Covers the whole trial lifecycle: design, interim, and readout, not just one decision point
+Named collaborations with biotech and pharma sponsors, plus regulatory-aligned methodology
+Awards and press coverage indicate an active, growing organization
한계
–Pricing is not published, so budgeting requires a direct sales conversation
–The value proposition depends on regulatory acceptance of twin-based methods, which varies by agency and indication
–Best suited to sponsors with existing clinical data and statistical teams; not a self-serve tool
–Public metrics come from specific studies and may not transfer directly to other indications