The exact text we saved

This is the version exactly as it was saved, character for character. The SHA-256 hash below is a 64-character code worked out from that text. Run the same text through SHA-256 and you get the same code; change a single letter and the code comes out completely different. Each version also stores the previous version’s hash, which chains them together so no version can be swapped out quietly.

TRUVACE RECORD VERSION
record: TRV-2026-2783
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-11-12T17:23:22.189426Z
status: published
lens: g_space
sector: education
headline: Generative AI in higher education: a meta-analysis of intellectual and social–emotional outcomes
dek: Background Generative AI has rapidly entered higher education, yet evidence on its effects across learning domains remains fragmented. This study synthesized experimental evidence on whether generative AI improves intellectual outcomes and social emotional outcomes among higher education students. Methods A systematic review and meta-analysis was conducted on 33 experimental and quasi-experimental studies involving 3,394 participants. The included studies were published between 2021 and 2024, and the literature…
gain_title: A meta-analysis of 33 experimental studies with 3,394 university students found generative AI raised intellectual outcomes (Hedges' g = 1.096) and social-emotional outcomes (g = 0.301) compared with control conditions.
problem_title: (none)
trace_subject: (none)
gain_reading: A meta-analysis of 33 experimental studies with 3,394 university students found generative AI raised intellectual outcomes (Hedges' g = 1.096) and social-emotional outcomes (g = 0.301) compared with control conditions.
gain_evidence: Generative AI demonstrated a significant positive effect on intellectual outcomes (Hedges’ g = 1.096, 95% CI 0.087 to 2.104, p = 0.033) | a smaller but significant positive effect on social emotional outcomes (Hedges’ g = 0.301, 95% CI 0.048 to 0.553, p = 0.020)
problem_reading: (none)
problem_evidence: (none)
quick_read: A systematic review and meta-analysis pooled 33 experimental and quasi-experimental studies covering 3,394 higher education students. It found generative AI had a significant positive effect on intellectual outcomes (Hedges' g = 1.096) and a smaller one on social-emotional outcomes (g = 0.301).

The intellectual-outcome estimate has a very wide confidence interval (0.087 to 2.104), and the gap between the two outcome domains was not statistically significant (p = 0.252). The authors call generative AI a promising resource, particularly for intellectual outcomes.

Confidence in the intellectual result is weaker: publication bias was detected for those outcomes and the results were sensitive to individual studies. The social-emotional findings were relatively stable.
limitation: The authors say the evidence base is heterogeneous and methodologically uneven, publication bias was detected for intellectual outcomes, and those results were sensitive to individual studies.
tag: Evidence-backed gain
key_points: The review pooled 33 experimental and quasi-experimental studies with 3,394 participants, published between 2021 and 2024. | Intellectual outcomes showed a pooled effect of Hedges' g = 1.096 (95% CI 0.087 to 2.104, p = 0.033). | Social-emotional outcomes showed a smaller pooled effect of g = 0.301 (95% CI 0.048 to 0.553, p = 0.020). | The difference between the two outcome domains was not statistically significant (coefficient = -0.739, p = 0.252). | Publication bias was detected for intellectual outcomes but not for social-emotional outcomes. | Functional type, intervention duration and knowledge domain did not significantly explain differences between studies.
rundown: The authors searched databases from inception to January 26, 2026, and included experimental and quasi-experimental studies published between 2021 and 2024. They used random-effects models in Stata 18 and registered the review in PROSPERO (CRD420261361622).

Beyond the pooled effects, the team ran publication bias, sensitivity and moderator analyses. Social-emotional findings held up relatively well, while intellectual outcomes shifted depending on which individual studies were included.

Moderator analyses found that the type of AI function, how long the intervention lasted and the subject area did not significantly explain differences between studies at the omnibus level. The authors describe the evidence base as heterogeneous and methodologically uneven.
sources:
- peer_reviewed | Frontiers in Psychology | https://doi.org/10.3389/fpsyg.2026.1848745 | 2026-06-08
prev: 0000000000000000000000000000000000000000000000000000000000000000

SHA-256 hash

e151f018ae27a16f5a11a3a678b081f07ed83671713d209b41380c34aec6e3f5

Previous version’s hash

0000000000000000000000000000000000000000000000000000000000000000