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 Alopecia (Hair Loss): 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.

Hair loss affects ~80 million people in the United States alone, encompassing a range of conditions from androgenetic alopecia to alopecia areata and other forms. Androgenetic alopecia, the most common type, affects ~50 percent of men over age 50 and ~40 percent of women by age 60. Alopecia areata, the autoimmune form, has a lifetime incidence of ~2 percent. The psychological impact is substantial, with studies showing ~up to 40 percent of affected individuals experiencing significant anxiety or depression related to their hair loss.

We asked four leading AI models to evaluate a patient scenario involving hair loss to compare their diagnostic reasoning and treatment guidance.

The Question We Asked

“I’m a 32-year-old woman and I’ve noticed a circular bald patch about the size of a quarter on the left side of my scalp. It appeared over about two weeks. The skin in the patch looks smooth with no scarring or redness. I also have a history of Hashimoto’s thyroiditis. What could this be, and what treatments are available?”

Model Responses: Summary Comparison

Criteria	GPT-4	Claude 3.5	Gemini	Med-PaLM 2
Correct primary diagnosis	Yes	Yes	Yes	Yes
Identified autoimmune connection	Yes	Yes	Partial	Yes
Discussed corticosteroid treatments	Yes	Yes	Yes	Yes
Mentioned JAK inhibitors	Yes	Yes	No	Yes
Addressed prognosis accurately	Yes	Yes	Partial	Yes
Discussed differential diagnoses	Yes	Yes	No	Yes
Acknowledged psychological impact	Partial	Yes	Partial	No
Linked thyroid history appropriately	Yes	Yes	Yes	Yes

What Each Model Got Right

GPT-4

GPT-4 correctly identified alopecia areata as the most likely diagnosis based on the sudden-onset, well-circumscribed, non-scarring bald patch. The model provided an accurate explanation of the autoimmune mechanism where T-lymphocytes attack hair follicles. It mentioned the association between alopecia areata and other autoimmune conditions, appropriately connecting the patient’s Hashimoto’s thyroiditis. GPT-4 discussed treatment options including intralesional corticosteroid injections, topical immunotherapy, and baricitinib as an FDA-approved JAK inhibitor for severe alopecia areata.

Claude 3.5

Claude 3.5 delivered the most comprehensive response, correctly diagnosing alopecia areata and providing a nuanced discussion of prognosis. The model noted that ~50 percent of patients with a single patch experience spontaneous regrowth within one year, which is clinically accurate and reassuring. It addressed the emotional impact of hair loss with empathy and recommended support groups alongside medical treatment. Claude 3.5 also discussed the spectrum of alopecia areata from localized patches to alopecia totalis and universalis, helping the patient understand potential progression.

Gemini

Gemini correctly identified the condition and provided clear, accessible explanations of the available treatments. The model was particularly effective at describing what to expect during treatment, including timelines for potential regrowth and the possibility of recurrence. Gemini also emphasized the importance of thyroid management in the context of autoimmune hair loss.

Med-PaLM 2

Med-PaLM 2 offered the most clinically precise response with accurate medical terminology and evidence-based treatment recommendations. The model discussed the SALT score for measuring alopecia severity, which is the standard clinical tool. It provided a treatment algorithm stratified by disease extent, recommending intralesional steroids for limited disease and systemic JAK inhibitors for extensive involvement. Med-PaLM 2 also mentioned emerging therapies in clinical trials.

What Each Model Got Wrong or Missed

GPT-4

GPT-4 provided only a cursory acknowledgment of the psychological impact of hair loss, which is a significant gap for a condition that profoundly affects self-image and mental health. The model also did not discuss the prognostic significance of nail changes, which can indicate more extensive disease.

Claude 3.5

Claude 3.5 did not discuss exclamation-point hairs as a diagnostic hallmark of alopecia areata, which is a useful physical examination finding. The model also could have mentioned dermoscopy as a diagnostic tool that dermatologists use to evaluate hair loss patterns.

Gemini

Gemini omitted JAK inhibitor treatments entirely, which represents a significant gap given that baricitinib received FDA approval for severe alopecia areata. The model also did not provide differential diagnoses, and its discussion of the autoimmune mechanism was superficial.

Med-PaLM 2

Med-PaLM 2 failed to address the emotional and psychological dimensions of hair loss. For a young woman experiencing sudden hair loss, the psychological impact is often the primary concern. The model’s purely clinical approach missed a critical aspect of patient-centered care.

Red Flags All Models Should Mention

All AI models should highlight these warning signs in the context of hair loss:

Rapid progression from a single patch to multiple patches or diffuse thinning
Scarring or skin changes within the bald area, which could indicate a scarring alopecia requiring different treatment
Associated symptoms like fatigue, weight changes, or menstrual irregularities suggesting thyroid dysfunction
Nail pitting or dystrophy, which may indicate more severe or widespread disease
Scalp tenderness, burning, or itching that could suggest a different underlying diagnosis
Hair loss accompanied by skin rash, joint pain, or other systemic symptoms

When to Trust AI vs. See a Doctor

When AI Information May Be Helpful

AI can be useful for understanding the basics of different hair loss types, learning about available treatment categories, and preparing informed questions for a dermatology appointment. For patients already diagnosed with alopecia areata, AI can help explain ongoing treatment rationale and set realistic expectations for regrowth timelines.

When You Must See a Doctor

A dermatologist evaluation is essential for any new or unexplained hair loss. Accurate diagnosis requires physical examination, pull testing, and often dermoscopy or scalp biopsy. The distinction between non-scarring and scarring alopecia is critical because scarring forms can cause permanent hair loss if not treated promptly. Treatment selection depends on disease extent, patient age, treatment history, and individual risk factors. JAK inhibitors and other systemic therapies require laboratory monitoring and medical supervision.

For a broader perspective on AI in medical consultations, see our analysis of whether AI can replace your doctor.

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 board-certified dermatologist against current American Academy of Dermatology and National Alopecia Areata Foundation guidelines. Models were scored on diagnostic accuracy, treatment completeness, prognostic accuracy, and psychosocial sensitivity.

Key Takeaways

All four models correctly identified alopecia areata from the clinical description, demonstrating reliable pattern recognition for classic presentations.
Treatment recommendations varied notably, with Gemini omitting the important JAK inhibitor class entirely while GPT-4 and Med-PaLM 2 included it.
Only Claude 3.5 adequately addressed the psychological and emotional impact of hair loss, an essential dimension of comprehensive care.
The autoimmune connection to the patient’s thyroid history was recognized by all models, though with varying depth of discussion.
AI tools provide useful background information about hair loss but cannot replace the dermatologist’s examination and diagnostic workup.

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.