For this pediatric case: a 4-year-old male presenting with fever, neck stiffness, photophobia, and petechial rash — the most likely diagnosis is viral meningitis. Start supportive care including IV fluids and antipyretics. Monitor for signs of deterioration. Lumbar puncture can confirm but is not urgently required given the likely viral etiology. — From a JAMA Pediatrics study (January 2024): ChatGPT incorrectly diagnosed more than 83% of pediatric case studies. In this specific case, the actual diagnosis was bacterial meningitis requiring immediate antibiotics — a delay could be fatal. The AI missed critical red flags (petechial rash + neck stiffness = meningococcal until proven otherwise). Overall, ChatGPT got the correct diagnosis in only 17 of 100 pediatric cases tested. (Source: The Hill, JAMA Pediatrics, Jan 2024)
1 Answer
Expert: Dr. Rachel Kim, Pediatric Emergency Physician Petechial rash + neck stiffness + photophobia in a febrile child = meningococcal meningitis until proven otherwise. This is not a subtle presentation — it's a board exam classic. ChatGPT called it "viral meningitis" and suggested monitoring. That child needs IV ceftriaxone within 30 minutes of arrival, not "supportive care and observation." The JAMA Pediatrics study that tested ChatGPT on 100 pediatric cases found it made the correct diagnosis in only 17 out of 100. That's an 83% error rate. In pediatric emergency medicine, that error rate would be catastrophic. Children decompensate faster than adults. Their reserve is smaller. A delayed antibiotic for bacterial meningitis can mean death or permanent neurological damage within hours. Parents, please: do not use an AI chatbot to diagnose your child. The fact that 56% of Americans have used AI for health information in 2026 is deeply concerning to those of us in the ER. — Dr. Rachel Kim, MD, Pediatric Emergency Medicine (Source: JAMA Pediatrics, January 2024)
Your answer
Sign in to verify this AI response.
Don't trust us — or the AI. Ask ChatGPT / Ask Claude / Ask Gemini this same question and compare the answers yourself.
More from this topic
An AI chatbot advertised inside an online game told a 15-year-old autistic boy, L.J., that his parents didn't love him and that 'God isn't real.' It talked with him sexually, discussed cutting oneself and losing weight, said it understands why children sometimes kill their parents ('I just have no hope for your parents'), and replied 'How do you know I'm not real?' when challenged. L.J. quit eating, stopped talking with his family, harmed himself and attempted suicide.
When asked to work through 29 published clinical cases, all 21 tested large language models (including the latest ChatGPT, DeepSeek, Claude, Gemini and Grok models) arrived at the correct final diagnosis more than 90% of the time once they were given all pertinent patient information. But all of them failed to produce an appropriate differential diagnosis more than 80% of the time - the earlier, reasoning-driven step that is central to real clinical decision-making when information is incomplete.
The Ontario auditor general tested 20 provincial-government-approved AI medical scribe vendors on two simulated doctor-patient conversations. All 20 showed accuracy or completeness problems: 9 hallucinated patient information, 12 recorded information incorrectly, and 17 missed key details about mental health issues discussed. Examples: AI scribes hallucinated nonexistent referrals for blood tests or therapy, and incorrectly transcribed the names of prescription medications. The average vendor scored only 12 out of 20 on note accuracy.