#dangerousai — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #dangerousai, aggregated by home.social.
-
the next time your doctor asks you if you’re OK with them using “AI” (#LLMs) to transcribe your appointment, show them this article 💁♀️
Sick and wrong: Ontario auditors find doctors' AI note takers routinely blow basic facts
– 60% of systems mixed up prescribed drugs in patient notes ❌❌❌
– 45% of systems “fabricated information and made suggestions to patients' treatment plans” ❌❌
– 85% of systems “missed key details about the patients’ mental health issues” ❌❌❌❌
– 30% of systems “missed the patients’ mental health issues fully or partially or were missing key details ❌
which perhaps isn’t surprising considering that during the procurement process “criteria tied to accuracy, bias controls, and key security and privacy safeguards made up only a small portion of the total evaluation score” 🤦♀️
#MedicalTranscription #shitAI #dangerousAI #SpicyAutocarrot
^ or as soon as you notice them using it, if they don’t ask for your consent first