{
  "id": 7320703,
  "title": "Fine-tuning medical AI can improve diagnosis but also creates privacy risks",
  "url": "https://urgent.news/2026/09/14/fine-tuning-medical-ai-can-improve-diagnosis-but-also-creates-privacy",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-14T12:40:10.000Z",
  "source": {
    "name": "Medical Xpress",
    "slug": "medical-xpress",
    "url": "https://medicalxpress.com/news/2026-09-fine-tuning-medical-ai-diagnosis.html"
  },
  "original_language": "en",
  "account": "Qingyu Chen, a biomedical informatics and data science professor at Yale School of Medicine, led a study on how artificial intelligence language models are adapted for medicine. The research revealed that fine-tuning these models, while improving diagnostic performance, also increases the risk of reproducing sensitive patient information. The study, published in Nature Communications, highlights the need to balance the capabilities and reliability of medical AI.",
  "summary": "Qingyu Chen, PhD, and his team set out to study how artificial intelligence language models are adapted for medicine and found that what these models memorize can be both useful and risky. A model may retain valuable medical knowledge, but in a controlled study using real hospital records, the same fine-tuning—the added training that adapts a model to a specific task—that improved diagnostic…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}