LLMs, particularly GPT-4, can match or exceed medical students' examination performance¶
Claim · All claims
Evidence · 1 study · quant-synthesisr3·q2
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person · recheck by 2028: tool and model accuracy changes quickly
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q2 i? The review concludes that LLMs, particularly GPT-4, can match or exceed medical students' examination performance, citing prior studies in ophthalmology and general medicine. → Nouri Haniyeh 2025
Evidence¶
Nouri Haniyeh 2025¶
Nouri Haniyeh, Mahdavi Abdollah, Abedi Ali, Mohammadnia Alireza, Hamedan Mahnaz, Amanzadeh Masoud. (2025). Performance of large language models in medical licensing examinations: a systematic review and meta-analysis. Journal of Educational Evaluation for Health Professions. https://doi.org/10.3352/jeehp.2025.22.36
q2 · i? · quant-synthesis · r3
The review attributes this to prior work (citing Wu et al. [11] and Liu et al. [9]); it is a second-hand citation narrated by the review, not a comparison computed within this meta-analysis. The review reports GPT-4 "performs at or above the level of medical students."
"For example, studies in ophthalmology and general medicine have shown that GPT-4 performs at or above the level of medical students when responding to board-style questions [11]."
Discussion¶
Related Claims¶
- GPT-4 achieves passing accuracy on USMLE Steps 1–3 (mean 86%), significantly exceeding GPT-3.5 (mean 56%) — a narrower finding that bears on this claim
- GPT-4 exceeds pass marks on India NEET-PG and UK PLAB 1 but falls below pass on clinical reasoning and procedural items — related
- GPT-4 examination accuracy is moderated by item type: multi-step vignettes (61–71%) and image-dependent items (48–63%) fall well below single-fact recall (83–88%) — related
- In an RCT reviewed by this study, an LLM-based digital patient improved ophthalmology medical history-taking skills relative to control — related
- Case-based learning benefits were reported particularly when GenAI was embedded in faculty-guided teaching, while complex cases and team workflows showed null or unfavorable effects — related
Page metadata
| Field | Value |
|---|---|
status |
review |
generated |
by process:wiki-ingest on 2026-10-31 |
trust |
unverified — no one has confirmed this page's content |
Frontmatter is the page's machine-readable record — see the Schema & Guide.