A meta-analysis of 69 studies reports generative AI improves academic performance with g = 0.7, but the article argues this likely reflects immediate task success rather than learning¶
Claim · All claims
Evidence · 1 study · theoreticalr3·q2·i2medium
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person
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 i2 A meta-analysis of 69 studies of ChatGPT and other generative AI tools in K-12 and tertiary education reported improved student academic performance with Hedge's g = 0.7. → Yan 2025
q1 i? The article argues this large effect probably reflects immediate task success rather than learning, reflecting a conflation of the two concepts. → Yan 2025
Evidence¶
Yan 2025¶
Yan, L., Greiff, S., Lodge, J. M., & Gašević, D. (2025). Distinguishing performance gains from learning when using generative AI. Nature Reviews Psychology, 4, 435–436. https://doi.org/10.1038/s44159-025-00467-5
q2 · i2 · theoretical · r3
The article reports, citing Deng et al., a meta-analysis of 69 experimental studies in K-12 and tertiary education reporting improved academic performance with "an effect of (Hedge’s)g= 0.7". The commentary attributes the performance-versus-learning conflation to that literature, not to new data.
"a meta-analysis of 69 studies aimed to investigate the impact of ChatGPT and other generative AI tools on “student learning” in K-12 and tertiary education. Yet, the meta-analysis reported that these technologies improve student “academic performance”, with an effect of (Hedge’s)g= 0.7"
Discussion¶
Related Claims¶
- AI severs the correlation between artefact quality and cognitive engagement, breaking the artefact-as-proxy for learning — the general claim this finding bears on, with the studies behind it
- Product-outcome studies report GenAI-related gains on product-quality measures but do not assess durable, transferable learning — a broader claim this one bears on
- ChatGPT-assisted students scored substantially higher on the coding assessment than No AI students (89% vs 69%) — related
- Reviewed literature reports risks of overreliance, AI errors in complex tasks, and students' uncritical acceptance of AI-generated outputs — related
- Many current GenAI applications in STEM education conflate learning with performance and feedback, neglecting active, reflective, and embodied learning processes — related
- Engagement remains an observable and evidential proxy for learning that AI has not broken, though it is not invulnerable — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
genai-meta-analysis-g-0-7-performance-not-learning |
title |
A meta-analysis of 69 studies reports generative AI improves academic performance with g = 0.7, but the article argues this likely reflects immediate task success rather than learning |
description |
A meta-analysis of 69 studies reports generative AI improves academic performance with g = 0.7, but the article argues this likely reflects immediate task success rather than learning |
status |
review |
evidence_strength |
weak |
newest_evidence |
2025 |
generated |
by process:wiki-ingest on 2026-10-10 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — yan-2025 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.