TL;DR
Prime made for students and young adults
- Fast, free delivery for dorm and study essentials
- Prime Video and Amazon Music included
- Member-only deals
In a guest essay published by Astral Codex Ten, Drew Housman says he and his wife used ChatGPT as a supplement during six years of fertility treatment. He says the chatbot suggested an MRI that revealed a large uterine fibroid missed in earlier care; the account is one couple’s experience, not evidence that AI can diagnose infertility reliably.
Drew Housman says ChatGPT prompted him and his wife to ask their fertility clinic for an MRI, which revealed a large uterine fibroid after years of testing and unsuccessful embryo transfers. The account, published as a guest post on Astral Codex Ten, describes how the couple used an AI chatbot alongside clinical care; it does not establish that the system can diagnose infertility or that its advice is dependable for other patients.
Housman writes that he and his wife could create embryos, but repeated transfers did not result in a pregnancy. After consulting multiple doctors without identifying an explanation, they were told their infertility was unexplained. He describes the couple’s frustration with limited time for questions during appointments and says they began using GPT-4 to discuss scans, hormone levels, test results, and possible next steps.
According to Housman, the chatbot—given the fictional persona “Dr. Reid”—suggested that his wife undergo an MRI. He says their fertility doctor was initially reluctant, and the couple had to insist that the clinic order the scan. The MRI found a 6-centimeter submucosal fibroid in the uterine wall, which Housman says had not been detected during the preceding six years of care. The clinic’s notes, as he recounts them, said the fibroid needed removal.
Housman says the treating fertility doctor reviewed the images and also concluded that the fibroid should be removed. The supplied account does not describe the procedure, its outcome, or whether removal changed the couple’s fertility prospects. Housman also acknowledges that ChatGPT sometimes misread an ultrasound and incorrectly told them his wife was pregnant, illustrating that the system could produce confident errors even while he felt many other answers matched what their doctors told them.
AI as a Prompt for Further Testing
The account highlights one possible role for consumer AI in medical care: helping patients generate questions to bring to clinicians, particularly when appointments are brief and a case has not produced a clear explanation. In Housman’s story, the chatbot’s MRI suggestion led the couple to press for a test that identified a finding their earlier workup had not revealed.
That is a report about an individual experience, not a controlled evaluation of an AI tool. The essay does not show that ChatGPT independently diagnosed the cause of the couple’s infertility, that the fibroid was definitively responsible for failed transfers, or that MRI would be suitable for other patients with similar histories. Those decisions depend on clinical evaluation and the details of each case. The distinction matters because a striking anecdote can encourage patients to seek useful follow-up, but it cannot establish diagnostic accuracy or treatment benefit.
The story also points to a practical tension in health care: patients may turn to chatbots when they need more explanation than they feel they can get during appointments. A chatbot can respond at any hour, but it can also misinterpret images or provide inaccurate guidance. Housman’s own description contains both sides: he says doctors often agreed with the answers, while also recounting a plainly incorrect pregnancy interpretation.
As an affiliate, we earn on qualifying purchases.
Six Years of Unexplained Infertility
Housman frames the MRI as a development in a long fertility journey, not an isolated consultation. He says the couple underwent multiple scans, tests, biopsies, and embryo transfers over approximately six years. Despite being able to create embryos, they had not achieved a pregnancy through the transfers described in the essay. Their clinicians ultimately used the label unexplained infertility, meaning that the prior evaluation had not identified a clear explanation, as Housman presents it.
The couple began asking GPT-4 medical questions after its release, according to the essay. Housman says the chatbot initially resisted some health questions, and that he framed prompts as scenes from a medical television drama to get responses. He describes using it to discuss test results and possible investigations, then checking answers with human doctors. This is his recollection of using the service; it is not an account of a medically supervised AI program or an assessment of the chatbot’s performance across patients.
The essay is a personal report by a guest author, not a clinical study. It offers no medical records, independent review of the imaging, or comparison showing what would have happened without the chatbot’s suggestion. Its central factual claim about the scan and fibroid is attributed to Housman’s account, while the broader conclusions about AI’s usefulness remain untested by the material provided.
““That’s when our LLM doctor stepped in.””
— Drew Housman, guest author
fertility testing ultrasound device
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
What the MRI Cannot Establish
The essay does not establish whether the fibroid caused the failed embryo transfers or whether removing it improved the couple’s chances of pregnancy. It also does not say whether the procedure took place, when it happened, or what the outcome was. The couple’s fertility status after the discovery is not provided.
Other details are also unavailable: the essay does not provide the MRI images or full medical records, the specific earlier tests and their results, or an independent account from the fertility clinic. Housman says the doctor initially tried to dissuade them from the MRI but does not know why. It is unclear whether the clinician had a specific medical rationale, considered the test unlikely to help, or had another reason.
Finally, a single personal account cannot determine how often chatbots suggest appropriate tests, how often they make harmful or misleading suggestions, or whether their use improves patient outcomes. Housman reports that many responses aligned with his doctors’ views, but that comparison is his own and is not a systematic accuracy assessment.
As an affiliate, we earn on qualifying purchases.
The Couple’s Treatment Outcome
The supplied report ends with the MRI finding and the agreement between Housman’s fertility doctor and the chatbot that the fibroid needed removal. It does not state what treatment followed or whether the couple later achieved a pregnancy. No further clinical milestone or follow-up date is given, so the result of the intervention remains unknown.
For readers, the immediate takeaway is limited: Housman says an AI suggestion helped prompt a discussion about further imaging, but any testing or treatment decision should be made with a qualified clinician familiar with the patient’s history. The account provides no basis for treating chatbot output as a substitute for medical diagnosis.
fertility health monitoring devices
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What did ChatGPT suggest in Housman’s account?
Housman says ChatGPT suggested an MRI for his wife during their fertility investigation. The scan revealed a 6-centimeter submucosal fibroid, according to his account.
Did the MRI prove the fibroid caused the infertility?
No. The essay reports that the fibroid was found and that the fertility doctor said it needed removal. It does not establish that the fibroid caused the failed embryo transfers or describe the result of treatment.
Was the chatbot consistently accurate?
Housman says many answers matched what the couple’s doctors told them, but he also recounts ChatGPT incorrectly interpreting an ultrasound as showing a pregnancy. The essay is not a formal accuracy study.
Does this account show that AI can diagnose infertility?
No. It describes one couple’s experience using a chatbot as a supplement to fertility care. It does not show that AI can reliably diagnose infertility or recommend tests for other patients.
Source: rss
Evergreen bestsellers Picks
bestsellers
As an affiliate, we earn on qualifying purchases.
