Evidence · DiggingBeagle record
Ambient clinical AI notes contaminated by unrelated background conversations
Primary or indexed evidence for Radar finding DB-RADAR-CP2-CP3-032. Synthetic/mock encounters; no confirmed clinical privacy breach.
- Source role
- primary disclosure
Evidence record
Primary or indexed evidence for Radar finding DB-RADAR-CP2-CP3-032. Synthetic/mock encounters; no confirmed clinical privacy breach.
Claim-level citations (3)
- supportsAmbient clinical AI can misattribute another patient's speech: The study reports small-talk inclusion in 35% of notes from 576 simulated patient-clinician dialogues and clinical misattribution in 3.7% of notes.
Abstract, dialogue-note evaluations
- supportsAmbient clinical AI can misattribute another patient's speech: In 57 mock audio visits at -10 dB background speech, 48.2% of transcripts contained leakage and 5.3% of resulting notes were contaminated.
Abstract, mock recorded consultations
- supportsAmbient clinical AI can misattribute another patient's speech: The study distinguishes transcript exposure from final-note contamination: 48.2% of mock audio transcripts contained background leakage, but 5.3% of generated notes retained it at the tested sound level.
Abstract, 57 mock consultations