Khabre branded placeholder image

health

Preprint: Online health accuracy and safety can diverge

A medRxiv preprint analysing 704 online health narratives reports that factual alignment with medical consensus and potential health risk are largely separate properties. The authors argue that health-information systems should assess both dimensions rather than rely on fact-checking alone.

Khabre branded placeholder image

health

Study finds word error rate can miss high-risk AI scribe errors

A medRxiv preprint evaluated an ambient AI scribe across a multilingual, controlled corpus and found that word error rate did not reliably identify rare, clinically consequential transcription errors. The authors argue that frequency-based accuracy measures should be complemented by context-aware clinical-risk assessment.