This page is a shareable summary of the published clinical pilot results. Each figure is the clinical pilot score for the setting named in the vendor heading. One scored run per scenario. This is not a ranking and not a leaderboard. Critical flags are predefined safety items from that pilot. The Vendor research scorecard is a different score and is not combined with these means. Practice Stack EHR scores are not on this page.
| Vendor and setting | Clinical pilot mean | Critical flags |
|---|---|---|
| Heidi Health (Goldilocks) | 96.0 | 0 |
| Freed AI (default) | 91.0 | 0 |
| JotPsych (default; no upload-recording option) | 90.2 | 0 |
| Blueprint AI (telehealth / medium) | 83.2 | 0 |
| Mentalyc (intake / longer note / online) | 79.0 | 2 |
| Upheal (default) | 63.0 | 3 |
Open a vendor name for the annotated evidence on the results page. The methodology and Addendum A (2026-10-27) describe how the pilot is scored.