Healthcare AI Safety Gap · Don't break patients
    Peer-reviewed / officialFeb 2026·US

    Leading chatbots give unsafe answers to real patient questions

    Physician red-teaming of four major models found up to 13.5% of answers unsafe and up to 43% problematic.

    13.5%
    of chatbot answers judged unsafe

    What the research found

    In a physician-led red-teaming study, researchers evaluated 888 responses from Claude, Gemini, GPT-4o and Llama to 222 real patient-posed primary-care questions. Problematic responses ranged from 21.6% (Claude) to 43.2% (Llama), and outright unsafe answers from 5% (Claude) to 13.5% (GPT-4o), several with the potential to cause serious harm.

    Why it matters for providers

    When patients put real medical questions to general chatbots — exactly what is happening now — a meaningful share of answers are unsafe, and the safest model was more than twice as safe as the worst.

    Original source
    npj Digital Medicine (Nature)
    Read the full report ↗
    Peer-reviewed or official source — the most reliable tier.