The DOI matches its Crossref record; 51 of the 51 references it cites were found in a registry or an index.
Playing Games with Ais: The Limits of GPT-3 and Similar Large Language Models
Open accessAbstract
This article contributes to the debate around the abilities of large language models such as GPT-3, dealing with: firstly, evaluating how well GPT does in the Turing Test, secondly the limits of such models, especially their tendency to generate falsehoods, and thirdly the social consequences of the problems these models have with truth-telling. We start by formalising the recently proposed notion of reversible questions, which Floridi & Chiriatti (2020) propose allow one to ‘identify the nature of the source of their answers’, as a probabilistic measure based on Item Response Theory from psychometrics. Following a critical assessment of the methodology which led previous scholars to dismiss GPT’s abilities, we argue against claims that GPT-3 completely lacks semantic ability. Using ideas of compression, priming, distributional semantics and semantic webs we offer our own theory of the limits of large language models like GPT-3, and argue that GPT can competently engage in various semantic tasks. The real reason GPT’s answers seem senseless being that truth-telling is not amongst them. We claim that these kinds of models cannot be forced into producing only true continuation, but rather to maximise their objective function they strategize to be plausible instead of truthful. This, we moreover claim, can hijack our intuitive capacity to evaluate the accuracy of its outputs. Finally, we show how this analysis predicts that a widespread adoption of language generators as tools for writing could result in permanent pollution of our informational ecosystem with massive amounts of very plausible but often untrue texts.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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?
The cited source says what the claim attributes to it.
In the citing paper
“Notably, LLMs use deep neural networks to predict the next word in a sequence of text and provide responses based on statistical patterns learned during training [12].”
From High Rates of Fabricated and Inaccurate References in ChatGPT-Generated Medical Content (2023) · cites this paper as [12]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318
A Mathematical Theory of Communication
Shannon, C. E. (1948) · Bell System Technical Journal · doi:10.1002/j.1538-7305.1948.tb01338.x · Cited by 88,486
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
A spreading-activation theory of semantic processing.
Collins, A. M., & Loftus, E. F. (1975) · Psychological Review · doi:10.1037/0033-295x.82.6.407 · Cited by 8,141 · cited 1×
“The metaphor of activation spreading through a semantic web has been introduced in context of human cognition (Collins & Quillian, 1969; Collins & Loftus, 1975 ) and while a simplification of human abilities, it may capture how these learnt links of different complexity are utilized by GPT.”
Searle, J. R. (1980) · Behavioral and Brain Sciences · doi:10.1017/s0140525x00005756 · Cited by 6,954
Retrieval time from semantic memory
Collins, A. M., & Quillian, M. R. (1969) · Journal of Verbal Learning and Verbal Behavior · doi:10.1016/s0022-5371(69)80069-1 · Cited by 3,109
GPT-3: Its Nature, Scope, Limits, and Consequences
Floridi, L., & Chiriatti, M. (2020) · Minds and Machines · doi:10.1007/s11023-020-09548-1 · Cited by 2,519
SPERBER, D., CLÉMENT, F., HEINTZ, C., MASCARO, O., MERCIER, H., ORIGGI, G., & WILSON, D. (2010) · Mind & Language · doi:10.1111/j.1468-0017.2010.01394.x · Cited by 1,795
A Review of Relational Machine Learning for Knowledge Graphs
Nickel, M., Murphy, K., Tresp, V., & Gabrilovich, E. (2016) · Proceedings of the IEEE · doi:10.1109/jproc.2015.2483592 · Cited by 1,638
Gilbert, D. T. (1991) · American Psychologist · doi:10.1037//0003-066x.46.2.107 · Cited by 1,515
Research works printed but not linked: 4
- Heller, F. (Director), & Goodson, M.B (Eds.). (1957). Jan 27). Salvador Dalí and Lillian Roth (Season 8, Episode 22) [TV series episode]. In M. Goodson & B. Todman (Executive producers), What’s my line?. Goodson-Todman Productions
- Lewis, D. K. (1986). On the plurality of worlds (322 vol.). Oxford: Blackwell
- Pearl, J., & Mackenzie, D. (2019). The book of why. Penguin Books
- Radford, A., Wu, J., Child, R., Luan, D., Amodei, D., & Sutskever, I. (2019). & others. Language models are unsupervised multitask learners. OpenAI Blog, 1(8), 9
Cited by
(2024) · doi:10.1109/access.2024.3365742 · Cited by 934
Recent advancements and challenges of NLP-based sentiment analysis: A state-of-the-art review
Jim, J. R., Talukder, M. A. R., Malakar, P., Kabir, M. M., Nur, K., & Mridha, M. F. (2024) · Natural Language Processing Journal · doi:10.1016/j.nlp.2024.100059 · Cited by 297
High Rates of Fabricated and Inaccurate References in ChatGPT-Generated Medical Content
Bhattacharyya, M., Miller, V. M., Bhattacharyya, D., & Miller, L. E. (2023) · Cureus · doi:10.7759/cureus.39238 · Cited by 294
Auditing large language models: a three-layered approach
Mökander, J., Schuett, J., Kirk, H. R., & Floridi, L. (2024) · AI and Ethics · doi:10.1007/s43681-023-00289-2 · Cited by 289
AutoAttacker: A Large Language Model Guided System to Implement Automatic Cyber-attacksPreprint
Xu, J., Stokes, J. W., McDonald, G., Bai, X., Marshall, D., Wang, S., Swaminathan, A., & Li, Z. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2403.01038 · Cited by 263
Metadata
- Authors
- Sobieszek, A., & Price, T.
- Affiliations
- University of Warsaw (PL)
- University of Nottingham (GB)
- Corresponding author
- Price, Tadeusz
- Venue
- Minds and Machines
- Published
- 2022-05-03
- Volume
- 32
- Issue
- 2
- Pages
- 341-364
- ISSN
- 0924-6495 (ISSN-L)
- 1572-8641
- OpenAlex
- W4225411436
- Semantic Scholar
- d072b46a0504ac023d5035d8ec0c7876151245c4
- Cited by
- 86
- References
- 41
- Citations by year
- 2026: 12 · 2025: 17 · 2024: 22 · 2023: 34 · 2022: 1
- FWCI
- 21.44
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- creativecommons.org/licenses/by/4.0 (tdm)
- creativecommons.org/licenses/by/4.0 (vor)
- Locations
- Minds and Machines (publishedVersion, cc-by)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Social Sciences › Arts and Humanities › History and Philosophy of Science › Cybernetics and Technology in Society
- Keywords
- GPT-3, large language models, Turing test, item response theory, semantic abilities, truth-telling, factual errors, distributional semantics, priming, compression, semantic web, information pollution, AI-generated text
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Playing Games with Ais: The Limits of GPT-3 and Similar Large Language Models. https://app.cochat.ai/papers/10.1007/s11023-022-09602-0
- The paper
- Sobieszek, A., & Price, T. (2022). Playing Games with Ais: The Limits of GPT-3 and Similar Large Language Models. Minds and Machines. https://doi.org/10.1007/s11023-022-09602-0
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.