The dangers of using proprietary LLMs for research
Open accessThe claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Reproducibility of results is a key concern in the application of LLMs in any research context.
“reproducibility of results is a key concern in the application of LLMs in any research context, as highlighted by Ollion et al. (2024) [48]”
From Sentiment Analysis in the Age of Generative AI (2024) · cites this paper as Ollion et al. (2024) [48]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
The citation lists haven’t been fetched yet
References 10 · Cited by 62
Research works printed but not linked: 1
- Gilardi, F., Alizadeh, M., & Kubli, M. (2023). PNAS 120, 30 (2023).
- Authors
- Ollion, É., Shen, R., Macanovic, A., & Chatelain, A.
- Affiliations
- Centre National de la Recherche Scientifique (FR)
- Utrecht University (NL)
- Corresponding author
- Ollion, Étienne
- Venue
- Nature Machine Intelligence
- Published
- 2024-01-17
- Volume
- 6
- Issue
- 1
- Pages
- 4-5
- ISSN
- 2522-5839 (ISSN-L)
- OpenAlex
- W4390947766
- Cited by
- 62
- References
- 10
- Citations by year
- 2026: 25 · 2025: 24 · 2024: 13
- FWCI
- 19.98
- Type
- journal article
- Language
- en
- Open access
- green
- hal.science
- Licence
- Locations
- Nature Machine Intelligence (publishedVersion)
- HAL (submittedVersion, other-oa)
- Utrecht University Repository (Utrecht University) (submittedVersion, other-oa)
- Radboud Repository (Radboud University) (submittedVersion)
- Funders
- Agence Nationale de la Recherche (ANR-19-P3IA-0001)
- Vetenskapsrådet (Mining for Meaning 2018-0517)
- Topics
- Social Sciences › Safety Research › Ethics and Social Impacts of AI
- Social Sciences › General Social Sciences › Computational and Text Analysis Methods
- Physical Sciences › Computer Science › Artificial Intelligence › Sentiment Analysis and Opinion Mining
- Keywords
- research ethics, generative AI in research, algorithmic bias, research reproducibility
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
