Verification register Bio, Health & Neuro
Current reading
The readiness gap, in one scan
AI-assisted assembly · derived results
- Claimed
- 75
- Reported
- 66
- Verified
- 57
- Gap
- +18
Public ambition and stated capability
Observed practitioner reporting
Independently supported evidence
Claimed minus verified
Evidence strength Strong
Decision
What the current evidence supports
Human editorial judgment · 2026-06-28
Adopt with guardrails
- Why
- It is the most rigorous public medical-LLM benchmark and a strong relative-ranking tool, but peer-reviewed critique (Digital Health, 2025) is explicit that benchmark performance does not equal clinical readiness — use it to compare models, not to certify safety.
- Next
- Use HealthBench as one signal in model selection, but pair with prospective silent-mode EHR evaluation against real clinician decisions before any clinical reliance
Constraints
Blockers
No named blocker is present in the current public projection.
Evidence summary
Derived counts
AI-assisted assembly
- Total
- 7
- Tier 1
- 0
- Tier 2
- 2
- Tier 3
- 5
- Supports
- 3
- Contradicts
- 3
- Context
- 1
- Latest observed
- 2026-04-30
Counts and dates only. Raw signals, private excerpts, trust records, and internal corpus material are not published here.
Publication record
Revisions
Initial public reading
This is the initial public reading. No earlier readiness change is recorded.