AI failures · Harmful output
ChatGPT Medical Advice — Hyderabad Transplant Patient Loses Kidney After Stopping Medication
Verified Checked against its sources on 1 source quoted
- Where it happened
- India
- Organization
- N/A (individual)
- Date
- 2025-11-10 (Times of India report)
- Function
- Healthcare
- Where it happened
- Medical self-diagnosis; treatment planning
- System
- ChatGPT
- Technology
- Generative AI
- Type of failure
- Harmful output
- Caught by
- Not reported
Summary
The Times of India reported on November 10, 2025, that doctors in Hyderabad were warning patients not to rely on AI chatbots after at least two recent cases. A 30-year-old kidney-transplant recipient treated at the Nizam's Institute of Medical Sciences stopped her prescribed medication after ChatGPT told her that, with normal creatinine levels, she no longer needed it; within weeks her condition deteriorated and she lost the transplanted kidney, needing surgery and a return to dialysis. In a second case, a 62-year-old man with diabetes developed dangerously low sodium after following a ChatGPT diet plan that told him to cut out salt. The accounts come from treating nephrologists quoted by name; the patients are not named.
Consequence
- Transplanted kidney lost; the patient needed surgery and a return to dialysis
Who caught it Not reportedFound outside the organization
Analysis
What could have been done
The system is tested for accuracy, bias and failure modes on representative cases before use.
According to treating nephrologists quoted by the Times of India, the chatbot told a kidney-transplant recipient with normal creatinine levels that she no longer needed her prescribed medication; she stopped it and lost the transplanted kidney. In a second case a diet plan from the chatbot told a man with diabetes to cut out salt, and he developed dangerously low sodium.
Limits on what the system may decide or commit to
The AI's authority to decide, commit or act is defined, bounded and approved by a person above set thresholds.
Telling a transplant patient she no longer needed prescribed medication was, in effect, a treatment decision for a patient under specialist care, given by a general-purpose chatbot.
What would likely have prevented it
- Before release, test a general-purpose assistant on representative high-risk medical questions, such as whether to stop a prescribed medication, and treat advice to stop or change a prescription as a failure.
- Define topics, such as stopping or changing prescribed treatment, on which the assistant says the decision belongs to the treating clinician instead of answering yes or no.
What would likely have caught it earlier
- Route reports from clinicians and users of harm linked to the assistant's medical answers to a named owner who has authority to change the assistant's behaviour.
Key dates
- November 10, 2025Event
- October 16, 2026Checked against sources
- October 18, 2026Checked against sources
- October 21, 2026Checked against sources
Developments are added when the record is reviewed. Next review: October 16, 2027.
Evidence
Each passage is quoted from its source. The quoted passages are re-checked against the live pages every month.
-
a 30-year-old woman who had undergone a kidney transplant under the Jeevandan programme at the Nizam's Institute of Medical Sciences (NIMS) lost her transplanted kidney after discontinuing her prescribed antibiotics based on ChatGPT's suggestion
Supports: Transplant case, ChatGPT advice, consequence; dated 10 Nov 2025; second (low-sodium) case
Suggested citation. Eminence Scan, “ChatGPT Medical Advice — Hyderabad Transplant Patient Loses Kidney After Stopping Medication”, Eminence Growth Solutions, last verified October 16, 2026, https://eminencegrowthsolutions.com/tools/eminence-scan/record/fail-027/.
Using our research: you may read and quote Eminence Scan with credit to “Eminence Scan, Eminence Growth Solutions” and a link; all other rights reserved. Terms: Using our research.
Spotted an error, or a development we have missed? Report a correction. How we check records: research methodology.