ChatGPT Often Inaccurate for IVF Guidance, Study Finds

Patients considering fertility treatment should verify AI-generated data against vetted clinical resources.

Updated on Oct. 30, 2026 in Pregnancy

Bold vector editorial illustration featuring a stylized anatomical model in coral and teal, illustrating fertility health guidance concepts. — A new study reveals that ChatGPT often provides inaccurate information regarding IVF, suggesting patients should prioritize vetted clinical resources over AI-generated guidance. AI Illustration. Upload story photo >

Live Poll

Would you trust medical information provided by an AI chatbot over professional medical society documents?

Physicians evaluating AI responses found that ChatGPT provides less accurate and complete information regarding IVF compared to educational materials from the American Society for Reproductive Medicine. While the chatbot offers similar clarity, patients should be aware of potential misinformation when seeking fertility guidance online.

Why it matters

As more patients turn to AI for complex medical information, researchers warn that these tools may present unsupported claims or inaccuracies that could impact fertility treatment expectations. This study highlights the importance of relying on vetted clinical sources for sensitive reproductive health decisions.

In a review of AI performance against professional standards, ChatGPT achieved a 94.1% checklist adherence rate compared to 100% for American Society for Reproductive Medicine documents. Researchers identified that 4.2% of chatbot responses contained misinformation, including inaccurate estimates for IVF live birth rates.

The players

American Society for Reproductive Medicine

A professional organization that develops educational standards and clinical guidelines for fertility care.

University of Pittsburgh

An academic institution where researchers are currently investigating the clinical safety of AI in reproductive health.

The details

Researchers at the University of Pittsburgh tasked four physicians with rating 12 AI-generated responses against two official ASRM educational documents using a five-point scale. The team compared ChatGPT’s success estimates against national data from the CDC and the Society for Assisted Reproductive Technology. While the chatbot successfully generated information at an accessible reading level, it frequently struggled with precision, notably overestimating birth rates for 35-year-olds and underestimating them for ages 37 and 40.

October 2026: Study results were reported at the ASRM annual meeting.

Health Landscape

This study underscores a growing trend of evaluating generative AI against long-standing clinical practice guidelines. It highlights the gap between conversational accessibility and medical precision in an era where AI tools are increasingly integrated into patient information pathways.

If you are researching IVF or fertility treatments, use chatbots only for general concepts rather than specific clinical statistics or success rates. Always verify any medical information you find online by discussing it directly with your reproductive endocrinologist.

The takeaway

Generative AI can offer helpful, readable summaries, but it often lacks the clinical rigor required for complex fertility planning. Prioritize official medical association materials to ensure your health decisions are based on accurate and complete data.