Healthcare AI Safety Gap · Don't break patients
    Peer-reviewed / official20 Nov 2026·Global

    No study has yet shown that clinical AI safeguards reduce harm to patients

    Frontiers in Artificial Intelligence: a scoping review of 111 reports on agentic and related clinical AI workflows found none that both isolated a safeguard's effect and showed reduced patient harm.

    0 of 111
    appraised reports that isolated a safeguard's effect and showed reduced patient harm

    What the research found

    Researchers reviewed 111 reports on clinical AI workflows, 56 of them on functionally agentic systems that choose their own actions, looking at safeguards such as knowledge grounding, rule gates, verification steps and human review. None of the appraised reports both isolated a specified safeguard's effect and showed reduced measured patient harm. Some patient-randomised trials reported patient or process benefits, but they evaluated bundled interventions rather than the contribution of any one safeguard.

    Why it matters for providers

    As AI tools start taking actions rather than just making suggestions, providers should ask vendors for evidence that their guardrails protect patients, not just that they exist.

    Original source
    Frontiers in Artificial Intelligence (Uppsala University, Mohammed Al-Mana College for Medical Sciences and others)
    Read the full report ↗
    Peer-reviewed or official source, the most reliable tier.