OpenEvidence CAIHL draft report

OpenEvidence: CAIHL reassessment

September 8, 2026. Published AI-assisted draft.

OpenEvidence’s historical profile describes clinician-controlled evidence and Visits workflows. Current official pages could not be retrieved in the September reassessment. The historical score is preserved provisionally, without treating the failed retrieval as a fresh confirmation or an agency penalty.

Scope and agency posture

Clinician evidence and Visits workflows described in the June baseline. Current public verification was blocked.

Posture: Institution-directed, patient-affecting.

Axis: 18/100, retained provisionally. No numerical recalibration was performed.

Sources and documented findings

• The frozen June report describes clinician-gated evidence synthesis, editable clinical outputs and patient handouts. This is historical evidence, not newly verified behavior. See the linked baseline report.

• The focused pass attempted the official Visits guide and privacy policy. Both returned HTTP 403. https://www.openevidence.com/user-guide/visits-overview and https://www.openevidence.com/policies/privacy

• One targeted fallback to the official homepage also returned HTTP 403. No current product or policy contents were recovered. https://www.openevidence.com/

Patient authority

Inference from the documented workflow: The historical deployment places the clinician in control of questions, context and outputs. Nothing newly recovered establishes direct patient control, but blocked access cannot establish its absence.

Critical capacity

Editorial assessment: Historical source-grounded answers may strengthen clinician reasoning and patient materials. Whether patients receive inspectable sources or can challenge the generated context remains unverified.

Informed control

Editorial assessment: Advertising separation, patient notice, refusal and correction need deployment evidence. Previously stated privacy protections must retain their June provenance.

Assessment and limits

The institutional interpretation has a specific historical basis. A fresh confirmation is unavailable, so this is an explicit uncertainty disposition rather than a verified keep or a new penalty.

Confidence: Low for present deployment, moderate for the frozen historical scope.

Remaining uncertainty: All current patient-facing controls and policy changes remain unverified after the bounded access failures

Published documentation is evidence of stated conditions, not proof of actual implementation. Unknowns did not receive automatic negative points. Funding, sponsorship, and public code do not determine agency by themselves.

Review provenance

• Reviewer/model: OpenAI Codex / GPT-6.
• Method: Focused public-source reassessment using CAIHL: patient authority, critical capacity, and informed control. Existing evidence plus one focused primary-source pass and at most one targeted follow-up. No live product testing. Numerical scores remain provisional editorial placements, not a new calculation.
• Human review: Hugo Campos authorized publication of these AI-assisted draft reassessments on September 8, 2026. This does not claim comprehensive human verification of every finding.
• Earlier review: 2026-06-10. Historical assessment. Earlier claims are not automatically reverified by this publication.