Turn on cookies to sign in
Signing in saves your progress to your Curie account. We can only do that with cookies on — turn them on to continue.
ChatGPT is clear but not always correct on IVF. A new comparative analysis found that while ChatGPT matched professional society materials in clarity, it scored lower on accuracy, completeness, safety, and quality — and half of its responses contained unsupported claims. Clinicians are urged to guide patients on how to use AI tools responsibly.
Patients are increasingly turning to AI chatbots for health information, but a new study presented at the ASRM 2026 annual meeting suggests ChatGPT may not be the most reliable source for IVF-related questions. Researchers compared ChatGPT's responses to common IVF queries against patient education materials from the American Society for Reproductive Medicine (ASRM) — and the chatbot came up short in several key areas.
While ChatGPT matched ASRM materials in clarity (4.60 vs. 4.63 out of 5), it scored notably lower on accuracy (3.98 vs. 5.00), completeness (3.40 vs. 4.88), safety (2.92 vs. 4.88), and overall quality (3.38 vs. 5.00). Half of ChatGPT responses contained unsupported claims, and 4.2% included outright misinformation.
By the Numbers:
Why it matters: Fluent, confident AI responses can mask errors, omissions, and outdated information — a concerning combination when patients are making high-stakes reproductive decisions. Clinicians should proactively ask patients about their AI use, teach them to prompt for cited sources, and direct them to trusted resources like ReproductiveFacts.org.
