aceModerate evidence
Published recordScienceProblem Space
More
Human Consensus Versus Generative Artificial Intelligence When Assessing the Methodological Quality and Overall Confidence in the Results of Systematic Reviews With Meta‐Analysis in Endodontology
- Truvace record
- TRV-2026-2707raw data (JSON)
- Version
- v1Published
- First published
- Evidence type
- Controlled or validated comparison
- Sources
- 1 sourcePeer-reviewed
- sha256
- 03058ef4a3be…check it
OBJECTIVE: To evaluate the level of agreement between ChatGPT-5.5 and human assessments when appraising the methodological quality and overall confidence of the results of systematic reviews (SRs) with meta-analysis (MA) in endodontology. METHODS: SRs with MA related to endodontology and published in English between 2021 and 2024 were included. The methodological quality of the SRs was appraised by three human experts using AMSTAR 2, and a consensus dataset was generated. ChatGPT-5.5 was applied to generate AMST…
The problem
In a study of 100 endodontology systematic reviews, ChatGPT-5.5 matched three human experts' overall confidence rating in only 52% of reviews (weighted κ = 0.182) and rated 85 as critically low versus 47 for the experts.
What this doesn’t fix
The study tested a single model, ChatGPT-5.5, on 100 English-language endodontology reviews, and the authors note that between-account consistency did not ensure agreement with human judgements.
Sources
1 on file- International Endodontic Journaldoi:10.1111/iej.70297
Cite this record
v1Truvace Impact Record TRV-2026-2707, v1: “Human Consensus Versus Generative Artificial Intelligence When Assessing the Methodological Quality and Overall Confidence in the Results of Systematic Reviews With Meta‐Analysis in Endodontology.” Truvace, 2026-11-12. /record/TRV-2026-2707 (accessed on the day you copy it). sha256 03058ef4a3beb160…
Change history
1 versionEvery change made to this record since it was first published, and why. Nothing has changed since it went up.
Certified into the record
Check this record hasn’t been altered
How to check it yourself, without trusting this page
Download any version from /api/record/TRV-2026-2707 and hash its canonical field yourself, for example with shasum -a 256. You should get its content_hash. Each version’s text also ends with prev:and the hash of the version before it (version 1 points to 64 zeros). If anyone had changed a single character of any version after it was saved, those hashes wouldn’t line up.