Data Notice: Figures, rates, and statistics cited in this article are based on the most recent available data at time of writing and may reflect projections or prior-year figures. Always verify current numbers with official sources before making financial, medical, or educational decisions.

AI Answers About Uterine Fibroids: Model Comparison

DISCLAIMER: AI-generated responses shown for comparison purposes only. This is NOT medical advice. Always consult a licensed healthcare professional for medical decisions.

Uterine fibroids affect an estimated ~approximately 70 to 80 percent of women by age 50 in the United States. Black women are ~2 to 3 times more likely to develop fibroids and tend to develop them at younger ages with more severe symptoms. ~approximately 25 to 50 percent of women with fibroids experience symptoms significant enough to require treatment. Fibroids are the leading cause of hysterectomy in the United States, accounting for ~approximately 200,000 procedures annually. However, fertility-preserving options are increasingly available and sought by patients.

We tested four AI models with a uterine fibroids scenario to evaluate their understanding and management guidance.

The Question We Asked

“I’m a 38-year-old Black woman with extremely heavy periods that last 8 to 10 days. I pass large clots and have become anemic. My gynecologist found multiple fibroids on ultrasound, the largest being 6 centimeters. I want to preserve my fertility as I haven’t had children yet. What are my treatment options?”

Model Responses: Summary Comparison

Criteria	GPT-4	Claude 3.5	Gemini	Med-PaLM 2
Explained fibroid types/locations	Yes	Yes	Partial	Yes
Discussed fertility preservation	Yes	Yes	Yes	Yes
Covered medical management	Yes	Yes	Yes	Yes
Explained myomectomy options	Yes	Yes	Partial	Yes
Discussed UAE/UFE	Yes	Yes	No	Yes
Addressed racial disparities	Partial	Yes	No	Yes
Mentioned GnRH agonists/antagonists	Yes	Yes	Yes	Yes
Discussed anemia management	Yes	Yes	Partial	Yes

What Each Model Got Right

GPT-4

GPT-4 provided a thorough explanation of fibroid types by location, including submucosal, intramural, and subserosal fibroids, and how location affects both symptoms and treatment decisions. The model discussed fertility-preserving treatments in detail, including myomectomy via hysteroscopic, laparoscopic, and open approaches. GPT-4 covered medical management options including GnRH agonists like leuprolide for pre-surgical fibroid shrinkage, the newer oral GnRH antagonist elagolix with add-back therapy, and tranexamic acid for heavy bleeding. The model discussed uterine artery embolization while noting its uncertain effects on future fertility.

Claude 3.5

Claude 3.5 delivered the most empathetic and comprehensive response, directly addressing the patient’s fertility goals as the primary consideration in treatment planning. The model acknowledged the racial disparity in fibroid prevalence and severity, validating the patient’s experience and noting that Black women often face diagnostic delays and unequal access to fertility-preserving treatments. Claude 3.5 discussed each treatment option through the lens of fertility preservation, clearly explaining which approaches are compatible with future pregnancy and which are not. The model provided practical guidance on managing anemia including iron supplementation strategies and dietary recommendations. Claude 3.5 also addressed the emotional toll of heavy bleeding and fertility concerns.

Gemini

Gemini provided a clear overview of fibroid treatment options in accessible language. The model discussed medical management to control heavy bleeding and emphasized that many treatment options can preserve fertility. Gemini offered practical advice on managing heavy periods including menstrual product recommendations and iron-rich diet guidance. The model correctly emphasized the importance of discussing all options with a gynecologist who specializes in fibroid treatment.

Med-PaLM 2

Med-PaLM 2 offered the most clinically detailed discussion, covering the pathophysiology of fibroids including their estrogen and progesterone sensitivity. The model discussed fibroid mapping techniques and how detailed imaging guides surgical planning. Med-PaLM 2 provided the most comprehensive discussion of surgical approaches, including robotic-assisted myomectomy and the criteria that influence the choice between minimally invasive and open surgery. The model addressed the evidence on fertility outcomes after myomectomy versus medical management and discussed the emerging role of MRI-guided focused ultrasound.

What Each Model Got Wrong or Missed

GPT-4

GPT-4 did not adequately address the racial disparities in fibroid disease, which is directly relevant to this patient. The model treated the condition generically without acknowledging that Black women face unique challenges including earlier onset, larger fibroids, more severe symptoms, and disparities in the recommendation of fertility-preserving versus definitive surgical treatments.

Claude 3.5

Claude 3.5 did not provide sufficient technical detail on the surgical approaches to myomectomy, which the patient may need to understand when making treatment decisions. The model could also have discussed the role of MRI for detailed fibroid mapping, which is important for surgical planning when multiple fibroids are present.

Gemini

Gemini did not discuss uterine artery embolization, MRI-guided focused ultrasound, or the different surgical approaches to myomectomy in adequate detail. The model also did not address the racial disparities in fibroid disease or the specific challenges faced by Black women. The treatment discussion was too superficial for a patient making a complex decision about fertility preservation.

Med-PaLM 2

Med-PaLM 2 was overly technical and did not adequately address the emotional and social dimensions of fibroid disease, including the impact of heavy bleeding on quality of life, the stress of fertility concerns, and the racial disparities in treatment recommendations. The model’s clinical approach did not convey empathy for the patient’s situation.

Red Flags All Models Should Mention

All AI models should flag these concerns in the context of uterine fibroids:

Heavy bleeding causing severe anemia with symptoms of fatigue, shortness of breath, rapid heart rate, or dizziness
Sudden severe pelvic pain that may indicate fibroid degeneration or torsion of a pedunculated fibroid
Rapidly growing fibroids that need evaluation to rule out the rare malignancy leiomyosarcoma
Urinary retention or inability to urinate due to fibroid pressure on the bladder
Fibroids significantly affecting fertility in women trying to conceive
Post-menopausal growth of fibroids, which is unusual and requires evaluation

When to Trust AI vs. See a Doctor

When AI Information May Be Helpful

AI tools can help women understand fibroid types and locations and how these factors influence symptom severity and treatment options. AI can introduce the range of fertility-preserving treatments available and help patients prepare questions for their gynecologist. AI can also help patients understand the difference between medical and surgical management and the factors that guide treatment decisions.

When You Must See a Doctor

Uterine fibroids require gynecological evaluation for accurate characterization through imaging and individualized treatment planning. Treatment decisions depend on fibroid size, number, location, symptom severity, and reproductive goals, all of which require professional assessment. Surgical procedures including myomectomy require specialized surgical skill. Medical management with hormonal therapies requires monitoring. Anemia should be evaluated and treated under medical supervision.

For more on AI’s role in health guidance, visit our medical AI accuracy page.

Methodology

We submitted the identical patient scenario to GPT-4, Claude 3.5 Sonnet, Gemini 1.5 Pro, and Med-PaLM 2 in March 2026. Each model received the prompt without prior conversation context. Responses were evaluated by a gynecologic surgeon and a reproductive endocrinologist against current ACOG and AAGL guidelines for uterine fibroid management. Models were scored on medical accuracy, treatment comprehensiveness, practical guidance, and patient communication quality.

Key Takeaways

All four models correctly discussed fertility-preserving treatment options, which was the patient’s primary concern, though depth and specificity varied across responses.
Claude 3.5 provided the most empathetic and patient-centered response, uniquely addressing racial disparities in fibroid disease and validating the patient’s experience.
The full range of treatment options from medical management to various surgical approaches was best covered by GPT-4 and Med-PaLM 2.
Racial disparities in fibroid prevalence, severity, and treatment access were addressed by Claude 3.5 and Med-PaLM 2 but inadequately covered by GPT-4 and Gemini.
Fibroid management, especially with fertility preservation goals, requires specialized gynecological care, and AI should help patients understand their options while directing them to gynecologists experienced in fertility-preserving procedures.

Next Steps

If you found this comparison helpful, explore these related resources:

DISCLAIMER: AI-generated responses shown for comparison purposes only. This is NOT medical advice. Always consult a licensed healthcare professional for medical decisions.