TL;DR
Get health and wellness essentials delivered free — and shop member deals
- Fast, free delivery on millions of items
- Access to Prime Big Deal Days deals on October 6–7
- Prime Video, Amazon Music and more included
A guest essay on Astral Codex Ten recounts how a couple used ChatGPT for questions during six years of fertility treatment and says the chatbot suggested an MRI that revealed a previously undetected uterine fibroid. The account does not establish that the AI caused a successful pregnancy or that the same approach is appropriate for other patients.
A couple seeking answers after six years of infertility treatment say ChatGPT suggested an MRI that found a 6-centimeter uterine fibroid not detected in their earlier workup. The account, published as a guest post on Astral Codex Ten by Drew Housman, describes the chatbot as a research and question-answering aid—not a substitute for the couple’s fertility doctors—and does not report the eventual outcome of treatment.
Housman writes that he and his wife could produce embryos, but their embryo transfers had not resulted in a pregnancy. After doctors had not identified a cause, the couple were told they had “unexplained infertility.” They began using ChatGPT after GPT-4 became available, asking about scans, hormone levels, follicle counts, procedures, supplements and other aspects of care. Housman says they checked the chatbot’s answers with their doctors and found the responses aligned, though he also recounts that it occasionally misread ultrasound images.
The turning point came when the chatbot suggested an MRI of his wife’s uterus, according to Housman. He writes that their fertility doctor was initially reluctant to order the scan, and that the couple had to insist. The MRI found a submucosal fibroid about 6 centimeters in diameter, a noncancerous growth in the uterine wall. Housman says the doctor reviewed the images and determined it should be removed.
The report provides a personal account rather than a clinical study or independent review of the medical records. It does not establish that the fibroid caused the couple’s failed embryo transfers, that ChatGPT’s suggestion would be useful for other patients, or what happened after the recommendation for surgery. The account also does not identify the model version used for the MRI suggestion or provide dates for the scan and treatment.
When AI Suggests Another Test
The episode highlights a possible role for chatbots in helping patients organize questions and discuss diagnostic options during time-limited appointments. Housman describes using ChatGPT to explore details between visits, then taking a suggested test back to a clinician. In this case, the account says that discussion led to imaging that identified a finding missed during years of prior fertility care.
That is meaningful to readers because fertility treatment can involve repeated procedures, substantial costs and long periods without a clear explanation. But the account does not show that an AI system can reliably identify missed conditions. The MRI finding and the doctor’s response are reported by the patient; the article offers no clinical documentation or independent assessment of whether the fibroid explains the infertility.
The practical distinction is between helping a patient ask about an option and making a diagnosis or treatment decision. Medical imaging, interpretation and decisions about surgery remain matters for qualified clinicians, who can consider the patient’s history, risks and alternatives.
As an affiliate, we earn on qualifying purchases.
Six Years Without an Explanation
Housman describes a fertility journey in which the couple could create embryos but had repeated unsuccessful transfers. After testing did not produce a clear explanation, clinicians used the term unexplained infertility. He writes that the label left the couple uncertain about what else might be investigated and that appointments offered limited time for questions.
The couple turned to ChatGPT for answers they could not readily get during visits. Housman says they asked about medical terms, test results and possible next steps, and compared its answers with those from their care team. He also describes limitations: the model sometimes made mistakes interpreting images, and its responses could vary or be blocked. The account therefore documents how one couple used a chatbot alongside clinical care; it does not compare AI advice with standard fertility practice or measure outcomes.
“That’s when our LLM doctor stepped in.”
— Drew Housman, in the Astral Codex Ten guest post
As an affiliate, we earn on qualifying purchases.
What the Account Cannot Establish
The guest post does not include medical records, imaging reports or comments from the fertility clinic, so the account cannot be independently verified from the material provided. It is also unclear why the doctor initially hesitated to order an MRI, whether the fibroid was responsible for the unsuccessful transfers, or what treatment and reproductive outcome followed its discovery.
Housman’s report describes one couple’s experience. It does not show how often a chatbot would suggest an appropriate test, how often such a suggestion would be wrong, or whether using AI changes fertility outcomes. Chatbot answers can be inaccurate, including in medical settings, and the author himself recounts an incorrect ultrasound interpretation. The story should not be read as a recommendation for patients to seek a particular scan or procedure.
As an affiliate, we earn on qualifying purchases.
Treatment Outcome Not Reported
Housman writes that the fertility doctor determined the fibroid should be removed after reviewing the MRI, but the supplied account ends before describing whether surgery took place or what followed. No later pregnancy or treatment result is established in the report. Further details from the couple or their care team would be needed to clarify the clinical outcome and how the finding affected their care.
For now, the reported development is the discovery of a fibroid after ChatGPT suggested discussing an MRI—not evidence that the chatbot diagnosed infertility or that AI-guided care improves results. Any decisions about imaging or treatment need to be made with a qualified clinician familiar with the patient’s circumstances.
As an affiliate, we earn on qualifying purchases.
Key Questions
What did ChatGPT suggest?
According to Drew Housman’s guest post, ChatGPT suggested that his wife get an MRI during their fertility workup. The scan later showed a 6-centimeter submucosal fibroid.
What is a uterine fibroid?
A fibroid is a noncancerous growth that develops in or around the uterus. Housman’s account describes the finding as submucosal and says the fertility doctor recommended removal; decisions about an individual fibroid require clinical evaluation.
Did the chatbot diagnose the cause of infertility?
No diagnosis by the chatbot is independently established in the report. The account says it suggested an MRI, which identified a fibroid, but does not establish that the fibroid caused the couple’s unsuccessful embryo transfers.
Did the couple later have a successful pregnancy?
The supplied guest post does not report whether the fibroid was removed or whether the couple later became pregnant. The eventual outcome remains unclear.
Should patients use AI to decide which fertility tests to get?
The report is a personal account, not medical guidance or evidence that AI can reliably recommend tests. Patients should discuss symptoms, test options and treatment decisions with a qualified clinician who knows their medical history.
Source: hn
Cold & flu season Picks
humidifiers
As an affiliate, we earn on qualifying purchases.
