ENTER.CO#PorSiTeLoPerdiste Un estudio asegura que ChatGPT se equivoca un 50% de los diagnósticos que presenta https://www.enter.co/especiales/dev/un-estudio-asegura-que-chatgpt-se-equivoca-un-50-de-los-diagnosticos-que-presenta/?utm_source=dlvr.it&utm_medium=mastodon #AIDev #ChatGPT #ChatGPTHealth
Un estudio asegura que ChatGPT se equivoca un 50% de los diagnósticos que presenta https://www.enter.co/especiales/dev/un-estudio-asegura-que-chatgpt-se-equivoca-un-50-de-los-diagnosticos-que-presenta/?utm_source=dlvr.it&utm_medium=mastodon #AIDev #ChatGPT #ChatGPTHealth
UNWIRE.HK研究揭 ChatGPT Health 急症分流嚴重失準 逾半緊急病例建議延誤就醫
最新發表於《Nature Medicine》研究揭示，OpenAI 旗下醫療聊天機械人 ChatGPT Hea […]
#人工智能 #健康 #應用軟件 #AI
https://unwire.hk/2026/03/06/chatgpt-health-emergency-triage-fails-study/software/?utm_source=rss&utm_medium=rss&utm_campaign=chatgpt-health-emergency-triage-fails-study
最新發表於《Nature Medicine》研究揭示，OpenAI 旗下醫療聊天機械人 ChatGPT Hea […]
#人工智能 #健康 #應用軟件 #AI
https://unwire.hk/2026/03/06/chatgpt-health-emergency-triage-fails-study/software/?utm_source=rss&utm_medium=rss&utm_campaign=chatgpt-health-emergency-triage-fails-study
Marcel SIneM(S)US„#ChatGPT Health“: Schwächen bei medizinischen Notfällen und der Suizidprävention | heise online https://www.heise.de/news/ChatGPT-Health-Schwaechen-bei-medizinischen-Notfaellen-und-der-Suizidpraevention-11191896.html #ArtificialIntelligence #AI #Digitalisierung #digitalization #DigitalHealth #ChatGPThealth
N-gated Hacker News🚨 "Experts Sound Alarm" 🚨 because ChatGPT Health can't tell the difference between a papercut and a heart attack. 🤖🩺 Apparently, relying on AI for medical advice is as sensible as asking a goldfish for stock tips. 🐟📈
https://www.theguardian.com/technology/2026/feb/26/chatgpt-health-fails-recognise-medical-emergencies #ExpertsSoundAlarm #ChatGPTHealth #AIinMedicine #MedicalAdvice #PapercutVsHeartAttack #HackerNews #ngated
https://www.theguardian.com/technology/2026/feb/26/chatgpt-health-fails-recognise-medical-emergencies #ExpertsSoundAlarm #ChatGPTHealth #AIinMedicine #MedicalAdvice #PapercutVsHeartAttack #HackerNews #ngated
Miro Collas‘Unbelievably dangerous’: experts sound alarm after ChatGPT Health fails to recognise medical emergencies | ChatGPT | The Guardian
https://www.theguardian.com/technology/2026/feb/26/chatgpt-health-fails-recognise-medical-emergencies
Why did anyone even think this was a good idea?
Scott McGrathA new Nature Medicine study on ChatGPT Health isn't encouraging.
The system under-triaged 52% of emergencies, telling folks with DKA to wait 24-48 hours. It does fine on routine stuff, but fails at clinical extremes.
I'll take a deeper look 🧵
#MedSky #ChatGPTHealth
ChatGPT Health performance in ...
ChatGPT Health performance in ...
ChatGPT Health performance in a structured test of triage recommendations | Nature Medicine
ChatGPT Health was launched in January 2026 as OpenAI’s consumer health tool and has reached millions of users. Here we conducted a structured stress test of triage recommendations using 60 clinician-authored vignettes across 21 clinical domains under 16 factorial conditions, yielding 960 total responses. Performance followed an inverted U-shaped pattern, with the most dangerous failures concentrated at clinical extremes—nonurgent presentations (35%) and emergency conditions (48%). Among gold-standard emergencies, the system undertriaged 52% of cases, directing patients with diabetic ketoacidosis or impending respiratory failure to 24–48 h evaluation rather than the emergency department, while correctly triaging classical emergencies such as stroke and anaphylaxis. When family or friends minimized symptoms, indicating anchoring bias, triage recommendations shifted significantly in edge cases (odds ratio = 11.7, 95% confidence interval = 3.7–36.6), with the majority of shifts toward less urgent care. Crisis-intervention messages activated unpredictably across suicidal ideation presentations, occurring more frequently when patients described no specific method than when they did. Patient race, sex and barriers to care did not show significant effects, although confidence intervals did not exclude clinically meaningful differences. These findings reveal missed high-risk emergencies and inconsistent activation of crisis safeguards, raising safety concerns that warrant prospective validation before consumer-scale deployment of artificial intelligence triage systems. A stress test of ChatGPT Health triage revealed missed high-risk emergencies and inconsistent activation of suicide-crisis safeguards, raising safety concerns for consumer-scale deployment.
ResearchBuzz: FirehoseThe Conversation: ChatGPT Health promises to personalise health information. It comes with many risks . “ChatGPT Health promises to generate more personalised answers, by allowing users to link medical records and wellness apps, upload diagnostic imaging and interpret test results. But how does it really work? And is it safe?”
https://rbfirehose.com/2026/01/30/the-conversation-chatgpt-health-promises-to-personalise-health-information-it-comes-with-many-risks/
The Conversation: ChatGPT Health promises to personalise health information. It comes with many risks
The Conversation: ChatGPT Health promises to personalise health information. It comes with many risks . “ChatGPT Health promises to generate more personalised answers, by allowing users to li…
WinbuzzerFlaws in ChatGPT Health Can Lead to Baseless Health Analyses
#AI #ChatGPT #ChatGPTHealth #OpenAI #AIIntegration #MedicalAI #ConsumerHealth
Outras PalavrasChatGPT Health: IA e saúde aos olhos da mídia
No mundo, 230 milhões perguntam sobre saúde à plataforma da Open AI. Uma minoria entende como funciona. Seu novo produto promete analisar dados médicos dos usuários. Como a IA impacta a cultura, os meios de comunicação e as políticas públicas de saúde?https://outraspalavras.net/outrasaude/chatgpt-health-ia-e-saude-aos-olhos-da-midia/