😺 OpenAI found 18 rare diseases

2026-06-19 · Source: The Neuron · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Emerging Technologies & Innovation, Software Development & Engineering · Depth: Fundamental Awareness, long

Summary

OpenAI has advanced its health intelligence initiatives, with GPT-5.5 Instant now providing enhanced health answers to over 230 million free ChatGPT users weekly, improving urgent-care recognition and context gathering. Concurrently, a collaborative NEJM AI study involving Boston Children's Hospital, Harvard, and OpenAI utilized an OpenAI reasoning model, o3 Deep Research, to reanalyze 376 previously unsolved pediatric rare-disease cases. This effort led to the confirmation of 18 new diagnoses by human clinicians after expert review, testing, and clinical validation. These developments highlight AI's evolving role from general advice to workflow support, positioning it as a "tireless second reader" for complex medical scenarios rather than an autonomous diagnostician.

Key takeaway

For individuals seeking health information or preparing for medical appointments, you should utilize AI tools like ChatGPT as a structured pre-appointment assistant, not a diagnostic replacement. Focus on organizing your symptoms, questions, and relevant context to present a clearer agenda to your human clinician. This approach helps you ask better questions and prepares you for care, mitigating the risk of misinterpreting AI's polished answers as definitive medical advice.

Key insights

AI models can significantly aid human clinicians in diagnosing complex rare diseases by generating evidence-linked leads.

Principles

• AI acts as a "tireless second reader" in medical diagnosis.
• AI health tools support workflow, not autonomous diagnosis.
• Human clinicians retain final decision-making authority.

Method

Researchers reanalyzed 376 de-identified unsolved cases using an OpenAI reasoning model, o3 Deep Research, to generate evidence-linked leads for human clinicians.

In practice

• Use ChatGPT as a pre-appointment prep assistant.
• Organize symptoms, questions for clinicians.
• Ask for uncertainty, not definitive answers.

Topics

• OpenAI
• Rare Disease Diagnosis
• AI in Healthcare
• ChatGPT Health
• Diagnostic AI
• Clinical Decision Support

Best for: Executive, AI Scientist, Research Scientist, AI Student, Tech Journalist, General Interest

Related on AIssential

• Universal Medical Intelligence: OpenAI's Plan to Elevate Human Health, with Karan Singhal — OpenAI's ChatGPT Health achieves physician-level performance, integrating AI into global medical practice with robust safety and privacy.
• How good are ‘AI doctors’ — and will they take over medicine? — Advanced AI models are matching or exceeding human diagnostic accuracy in specific medical contexts, but real-world application is complex.
• Spotlight on innovation: Google-sponsored Data Science for Health Ideathon across Africa — Google's Ideathon fostered African innovation in healthcare AI using open models like MedGemma and MedSigLIP.
• Exploring the feasibility of conversational diagnostic AI in a real-world clinical study — Supervised conversational AI for pre-visit history taking is feasible, safe, and well-received by patients and clinicians.
• 🗞️ Techie uses ChatGPT and AlphaFold to build DIY mRNA cancer vaccine, saves dog — AI tools empower individuals without specialized training to conduct complex scientific research and develop novel solutions.
• ‘Unbelievably dangerous’: experts sound alarm after ChatGPT Health fails to recognise medical emergencies — ChatGPT Health frequently misdiagnoses medical emergencies and suicidal ideation, posing significant patient safety risks.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by The Neuron.