온타리오 감사관들, 의사용 AI 노트 작성기가 기본 사실을 반복적으로 틀린다고 밝혀

Ontario's audit exposes a critical capability-reliability gap in medical AI: most approved Scribe systems hallucinate facts, omit critical patient information, and suggest treatments never discussed. This demonstrates why AI safety mechanisms and verification checks are non-negotiable in healthcare rather than relying on physician review alone.