AIThis post was created with the assistance of artificial intelligence (AI).

TL;DR

Prime Big Deal Days · Oct 6–7Offer from Amazon

Get school and study supplies 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
Start your free Prime trial Free trial for eligible customers · Cancel anytime
As an affiliate, we earn on qualifying purchases.

In a guest report for Astral Codex Ten, Drew Housman says he and his wife used ChatGPT as a supplementary source of fertility information during six years of unexplained infertility. He says the chatbot suggested an MRI that revealed a 6-centimeter uterine fibroid, which their fertility doctor recommended removing; the account does not establish that AI can diagnose infertility or that the fibroid caused the failed transfers.

ChatGPT suggested an MRI that uncovered a previously undetected 6-centimeter uterine fibroid during a couple’s years-long struggle with infertility, according to a guest report by Drew Housman published by Astral Codex Ten. Housman says the finding led the couple’s fertility doctor to recommend removing the growth, though the account does not show that the chatbot diagnosed the cause of their failed embryo transfers or that the fibroid’s removal resulted in a pregnancy.

Housman writes that he and his wife could create embryos through in vitro fertilization, but repeated embryo transfers did not result in pregnancy. After numerous consultations and tests over roughly six years, they were described as having unexplained infertility, a label used when a cause has not been identified. Housman says they felt their doctors had few new avenues to explore and that appointment time was limited.

The couple began using ChatGPT after GPT-4 became available, asking it about scans, hormone levels, follicle counts, procedures and other aspects of treatment. Housman says they compared its answers with those from their clinicians and found them generally consistent. He also describes errors, including an occasion when the chatbot misread an ultrasound and told them his wife was pregnant. The account is a personal report, not a clinical study or an independently verified assessment of the chatbot’s accuracy.

According to Housman, the chatbot proposed that his wife get an MRI. Their fertility doctor was initially reluctant, he says, and the couple had to press for the scan. The MRI identified a 6-centimeter submucosal fibroid in the uterine wall. Housman says the doctor reviewed the images and recommended removal. The supplied report does not provide the procedure’s outcome or establish whether the fibroid explained the couple’s infertility.

At a glance
reportWhen: Published in an Astral Codex Ten guest…
The developmentA guest writer says ChatGPT suggested an MRI that found a previously undetected uterine fibroid during his wife’s infertility treatment.

An AI Prompt Led to Further Imaging

The account illustrates a possible role for AI tools as patient question aids: they can help people organize concerns, ask about options and prepare to discuss further testing with clinicians. In this case, Housman says the chatbot raised an imaging option that had not been pursued during years of treatment, prompting the couple to return to their doctor with a specific request.

That is different from proving that an AI system can diagnose infertility or reliably identify which test a patient needs. The MRI finding and the doctor’s recommendation are reported by Housman; there is no medical record, independent expert review or comparison with other patients in the source material. The case matters as an example of how patients may use chatbots to advocate for further evaluation, while underscoring that clinical decisions and interpretation remain with medical professionals.

Amazon

uterine fibroid removal kit

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Six Years of Unexplained Infertility

Housman describes a treatment journey involving multiple fertility doctors, scans, tests and biopsies. Despite being able to create embryos, the couple experienced repeated failed transfers. Their situation was eventually categorized as unexplained infertility, which the report presents as a point at which the couple felt their medical team had no clear next diagnostic step.

The couple used ChatGPT alongside—not instead of—their clinicians, according to the account. Housman says they asked questions between appointments and checked responses with doctors. He portrays the chatbot as patient and readily available, but also notes that it sometimes made mistakes and could refuse medical questions. The report does not identify the exact model version used for the MRI suggestion, provide the scan date, or include a response from the treating clinic.

“That’s when our LLM doctor stepped in.”

— Drew Housman, guest writer at Astral Codex Ten

Amazon

MRI scan for uterine fibroid

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

What the Case Cannot Establish

The source is a single couple’s account, and the supplied material does not include medical records or independent confirmation. It remains unclear when the MRI was performed, how the fibroid affected the couple’s fertility, whether it was removed, and what happened after the doctor recommended removal. The report does not state that the couple later achieved a pregnancy.

It is also unclear why the fertility doctor initially opposed or questioned the MRI request; Housman says he never learned the reason. The account cannot show whether the chatbot’s suggestion was medically appropriate in other cases, whether a clinician would have recommended the same test with different information, or how often similar AI advice is accurate. Housman himself reports at least one major chatbot misreading, reinforcing that its responses were not consistently reliable.

Amazon

fertility tracking monitor

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Treatment Outcome Is Not Reported

The immediate next step described in the report is the fertility doctor’s recommendation to remove the fibroid. The supplied account does not say whether the couple proceeded with surgery, whether further treatment followed, or whether the finding changed the outcome of their IVF attempts. Without those details, the medical significance of the discovery for this couple remains unresolved.

The report offers no broader study or clinical guidance on using chatbots during fertility treatment. For now, its documented development is that a patient says an AI suggestion helped prompt a scan, which found a growth his clinicians had not previously detected. Any diagnosis, decision to order imaging, or treatment plan must be evaluated by the patient’s medical team.

Amazon

ultrasound pregnancy test

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Key Questions

What did ChatGPT suggest?

According to Drew Housman, the chatbot suggested that his wife get an MRI after years of infertility treatment and repeated failed embryo transfers.

What did the MRI find?

Housman says the scan revealed a 6-centimeter submucosal uterine fibroid that had not been detected during the couple’s earlier testing.

Did the chatbot diagnose the cause of infertility?

The report does not establish that. It says the chatbot suggested an MRI; the scan found a fibroid, and the fertility doctor recommended removing it. The source does not confirm that the fibroid caused the failed transfers.

Did the couple have the fibroid removed or later become pregnant?

The supplied account says the doctor recommended removal but does not report whether the procedure happened or provide a later pregnancy outcome.

Does this report show that AI can replace a fertility doctor?

No. It is a personal account, not a clinical study. Housman also describes a chatbot error, and the report does not validate its medical accuracy. The treating doctor reviewed the MRI and made the treatment recommendation.

Source: rss

HALLOWEEN

Halloween Picks

As an affiliate, we earn on qualifying purchases.

You May Also Like

LeMario: Training a JEPA World Model on Super Mario Bros

LeMario has trained a JEPA World Model on Super Mario Bros, marking progress in AI game understanding. Details on capabilities and implications are emerging.

Show HN: Learn By Rebuilding Redis, Git, A Database From Scratch

A developer shares a project to learn by reconstructing Redis, Git, and a database from the ground up, highlighting practical learning methods for system design.

Research Brent Baker Surges In Global Coverage

Brent Baker’s research has seen a surge in international coverage, with mentions increasing 46-fold in recent reporting, highlighting growing interest.

Can AI Management Styles Be Trusted? A Live Experiment Reveals Surprising Behaviors

A live experiment pits four AI management models against real crises, revealing their decision-making personalities, honesty, and ability to close deals—insights vital for responsible automation.