The DOI matches its Crossref record; 81 of the 81 references it cites were found in a registry or an index; 30 of the 53 sources its claims cite were checked for support.
Deanthropomorphising NLP: Can a language model be conscious?
Open accessAbstract
This work is intended as a voice in the discussion over previous claims that a pretrained large language model (LLM) based on the Transformer model architecture can be sentient. Such claims have been made concerning the LaMDA model and also concerning the current wave of LLM-powered chatbots, such as ChatGPT. This claim, if confirmed, would have serious ramifications in the Natural Language Processing (NLP) community due to wide-spread use of similar models. However, here we take the position that such a large language model cannot be conscious, and that LaMDA in particular exhibits no advances over other similar models that would qualify it. We justify this by analysing the Transformer architecture through Integrated Information Theory of consciousness. We see the claims of sentience as part of a wider tendency to use anthropomorphic language in NLP reporting. Regardless of the veracity of the claims, we consider this an opportune moment to take stock of progress in language modelling and consider the ethical implications of the task. In order to make this work helpful for readers outside the NLP community, we also present the necessary background in language modelling.
Overview
CoChat Summary
TL;DR
The paper argues that Transformer-based LLMs such as LaMDA cannot be conscious under Integrated Information Theory, and that claims of sentience arise from anthropomorphic language in NLP and media reporting. It also documents anthropomorphic terminology in ACL abstracts and offers recommendations for developers, users, and reporters.
Key findings
- Applying IIT, LaMDA and other Transformer-based LLMs cannot possess consciousness for three reasons: they execute as a sequence of Transformer blocks with extremely limited inter-block information exchange; the blocks are feed-forward networks with no recurrent connections; and they are implemented on modular, low-connectivity computer hardware.
- In LaMDA, the link between modules corresponds to a single word represented by 8192 numbers, while within-module links are at least the number of non-embedding weights, 137 billion in the most advanced LaMDA; within-module links outnumber between-module links by at least 8 orders of magnitude.
- LaMDA has 64 layers, 128 attention heads, 137 billion parameters, and was trained on 1.56 trillion words; the paper argues there is no breakthrough innovation in such models that would give rise to consciousness, and that apparent consciousness is a simulated language style produced by prompting.
- In an analysis of the first 100 abstracts from the 2022 ACL Annual Meeting, 90 referred to language modelling technology and 36 used anthropomorphic terminology; media reporting of LaMDA was found to use explicit anthropomorphism.
Methods
- Conceptual and argumentative analysis: the paper reviews language modelling, Transformer-based LLMs, and LaMDA, then assesses them against Integrated Information Theory as presented by Tononi and Koch, including recent LLM developments such as larger models, larger context windows, non-Transformer architectures, and retrieval-augmented generation.
- Qualitative audit: the authors analysed the first 100 abstracts in order of appearance from the proceedings of the 2022 ACL Annual Meeting, identified anthropomorphic terminology, and proposed a three-level framework—non-anthropomorphic, ambiguous anthropomorphism, and explicit anthropomorphism—which they applied to the LaMDA paper and selected media reporting.
- No empirical consciousness test or direct experimental manipulation of LaMDA is reported; the study is an analysis of architecture, theory, and language use.
Limitations
- Stated: the identification of anthropomorphic language relies on the subjective application of the framework by one author, and others may reach different conclusions; the authors acknowledge their inherent bias, namely non-belief in the model’s consciousness, and cannot claim to have fully overcome it.
- Stated: the study presupposes that distinguishing consciousness from non-consciousness in LLMs is valuable; it uses IIT because it appears most applicable to neural networks, but other approaches are possible; conclusions are limited to English-language models and examples, and generalisability is argued for Transformer-based models rather than all possible architectures.
- Evident: the media analysis is limited to selected sources around LaMDA, and the anthropomorphism audit is not presented with inter-rater reliability or a full quantitative coding protocol; the paper does not directly test LaMDA or other LLMs for consciousness.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures
Verification
Claims the paper makes
22 Valid · 13 Not valid · 12 Not checked · 68 Own results and interpretations
Every source the paper cites for this claim was checked and supports it.
In the paper
“weakly connected aggregates are not conscious [9].”
§ Deanthropomorphising NLP: Can a language model be conscious? > Can an LLM be conscious? > Transformer-based LLMs in IIT
What the sources say
- Supportedcites [9] G Tononi, 2015, Consciousness: here, there and everywhere?, Philosophical Transactions of the Royal Society B: Biological Sciences source
In “(f) Aggregates are not conscious,” the paper says that interacting smaller complexes can form a larger system without forming a superordinate conscious entity. It explicitly concludes that aggregates have no consciousness; this does not mean their constituent complexes cannot be conscious. (LLM, conf=0.93)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“And computer hardware is unlikely to be conscious itself because of the way it is designed: as a collection of modules, which are tightly integrated internally, but with limited connection between them [9].”
§ Deanthropomorphising NLP: Can a language model be conscious? > Can an LLM be conscious? > Transformer-based LLMs in IIT
What the sources say
- Partly supportedcites [9] G Tononi, 2015, Consciousness: here, there and everywhere?, Philosophical Transactions of the Royal Society B: Biological Sciences sourceExperimental — accuracy not yet measured
The section ‘Simulations of conscious neural systems can be unconscious’ says a digital computer would likely break into many low-integration mini-complexes because its circuitry has limited fan-in and fan-out. That supports the claim’s central point, but the paper does not describe the components as tightly integrated internally, and its Endnotes allow that differently designed neuromorphic hardware could potentially approach human consciousness. (LLM, conf=0.84)
hedgedstates a possible cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“Models of consciousness from neurology rely on stochastic variation [65] arising from noise caused by varied biological sources [66].”
§ Deanthropomorphising NLP: Can a language model be conscious? > Can an LLM be conscious? > Challenges for consciousness beyond IIT
What the sources say
- Supportedcites [65] HA Braun, 2021, Stochasticity Versus Determinacy in Neurobiology: From Ion Channels to the Question of the “Free Will”, Frontiers in Systems Neuroscience source
This source is cited for the noise clause [66], not the preceding claim about models of consciousness [65]. In “The Need of Appropriately Adjusted System Dynamics” (page not specified), the paper describes noise associated with synaptic connections, ion-channel stochasticity, and irregular neuronal firing patterns. (LLM, conf=0.88)
- Supportedcites [66] P Kraikivski, 2021, Implications of Noise on Neural Correlates of Consciousness: A Computational Analysis of Stochastic Systems of Mutually Connected Processes, Entropy source
The citation [66] is attached to the claim that neuronal noise arises from varied biological sources. The Introduction identifies voltage-gated channel, synaptic, and sensory-motor noise as sources in neuronal systems. (LLM, conf=0.91)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“According to IIT, this transition does not preserve consciousness, even if a simulated system can perform the same functions as the original one.”
§ Deanthropomorphising NLP: Can a language model be conscious? > Can an LLM be conscious? > Transformer-based LLMs in IIT
What the sources say
- Supportedcites [9] G Tononi, 2015, Consciousness: here, there and everywhere?, Philosophical Transactions of the Royal Society B: Biological Sciences source
The section “Simulations of conscious neural systems can be unconscious” says that faithfully replicating a brain’s functions in software does not justify granting the simulation the brain’s consciousness. It states that a simulation of a conscious brain will not have consciousness because IIT requires the relevant physical cause-and-effect power. (LLM, conf=0.94)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“Tononi and Koch [9] show why such a system cannot be conscious.”
- Also stated in: § Transformer-based LLMs in IIT · plain
“Following this procedure recursively eliminates the whole network, showing that a feed-forward network cannot be conscious according to IIT.”
§ Deanthropomorphising NLP: Can a language model be conscious? > Can an LLM be conscious? > Transformer-based LLMs in IIT
What the sources say
- Supportedcites [9] G Tononi, 2015, Consciousness: here, there and everywhere?, Philosophical Transactions of the Royal Society B: Biological Sciences source
In “Multiple consciousnesses,” Tononi and Koch explain that a purely feed-forward network has no recurrent connections and, according to IIT, carries out tasks unconsciously. The paper also explicitly predicts that feed-forward networks are not conscious. (LLM, conf=0.95)
boosted- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § Transformer-based LLMs in IIT · plain
50 more claims · 9 with their cited sources checked
Create a free account to see all 115 claims and to verify the 3 references still open.
- Probably not a substantial claim. 52
- It may state two separate claims as one. 21
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Deanthropomorphising NLP: Can a language model be conscious? > Can an LLM be conscious? > Challenges for consciousness beyond IIT
“The susceptibility to prompting has been also observed in ChatGPT, which started to provide abusive and threatening responses after being asked about its ‘shadow self’ [67].”
§ Deanthropomorphising NLP: Can a language model be conscious? > References
“67. Roose K. Why a Conversation with Bing’s Chatbot Left Me Deeply Unsettled; 2023.”
Citations
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325
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
RoBERTa: A Robustly Optimized BERT Pretraining Approach
Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., Levy, O., Lewis, M., Zettlemoyer, L., & Stoyanov, V. (2019) · arXiv · doi:10.48550/arxiv.1907.11692 · Cited by 16,515
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W., & Liu, P. J. (2019) · arXiv · doi:10.48550/arxiv.1910.10683 · Cited by 10,929 · cited 1×
“This means that our conclusions, especially regarding consciousness, apply to other models such as BERT (Devlin et al, 2019), GPT-3 (Brown et al, 2020a), T5 (Raffel et al, 2020) and future models using the same, or similar, architectures.”
I.—COMPUTING MACHINERY AND INTELLIGENCE
TURING, A. M. (1950) · Mind · doi:10.1093/mind/lix.236.433 · Cited by 10,290
Lewis, M., Liu, Y., Goyal, N., Ghazvininejad, M., Mohamed, A., Levy, O., Stoyanov, V., & Zettlemoyer, L. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.703 · Cited by 8,821 · cited 1×
“Among the most popular are XLNet (Yang et al, 2019), RoBERTa (Liu et al, 2019), BART (Lewis et al, 2020) and ELEC-TRA (Clark et al, 2020).”
Nagel, T. (1974) · The Philosophical Review · doi:10.2307/2183914 · Cited by 7,440
On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜
Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021) · Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3442188.3445922 · Cited by 6,988
Searle, J. R. (1980) · Behavioral and Brain Sciences · doi:10.1017/s0140525x00005756 · Cited by 6,954
Research works printed but not linked: 13
- Brown T, Mann B, Ryder N, Subbiah M, Kaplan JD, Dhariwal P, et al. Language Models are Few-Shot Learners. In: Larochelle H, Ranzato M, Hadsell R, Balcan MF, Lin H, editors. Advances in Neural Information Processing Systems. vol. 33. Curran Associates, Inc.; 2020. p. 1877–1901. Available from: https://proceedings.neurips.cc/paper/2020/file/1457c0d6bfcb4967418bfb8ac142f64a-Paper.pdf.
- Thoppilan R, De Freitas D, Hall J, Shazeer N, Kulshreshtha A, Cheng HT, et al. Lamda: Language models for dialog applications. arXiv preprint arXiv:220108239. 2022;.
- J Austen, 1813, Pride and Prejudice
- Mikolov T, Sutskever I, Chen K, Corrado GS, Dean J. Distributed Representations of Words and Phrases and their Compositionality. In: Burges CJ, Bottou L, Welling M, Ghahramani Z, Weinberger KQ, editors. Advances in Neural Information Processing Systems. vol. 26. Curran Associates, Inc.; 2013.Available from: https://proceedings.neurips.cc/paper/2013/file/9aa42b31882ec039965f3c4923ce901b-Paper.pdf.
- Yang Z, Dai Z, Yang Y, Carbonell J, Salakhutdinov R, Le QV. XLNet: Generalized Autoregressive Pretraining for Language Understanding. arXiv: 190608237. 2019;.
- Clark K, Luong T, Le QV, Manning C. ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators. In: Proceddings of ICLR; 2020.Available from: https://openreview.net/pdf?id=r1xMH1BtvB.
- Zhu F, Lei W, Wang C, Zheng J, Poria S, Chua TS. Retrieving and reading: A comprehensive survey on open-domain question answering. arXiv preprint arXiv:210100774. 2021;.
- Chowdhery A, Narang S, Devlin J, Bosma M, Mishra G, Roberts A, et al. Palm: Scaling language modeling with pathways. arXiv preprint arXiv:220402311. 2022;.
- Zhang S, Roller S, Goyal N, Artetxe M, Chen M, Chen S, et al. Opt: Open pre-trained transformer language models. arXiv preprint arXiv:220501068. 2022;.
- Beltagy I, Peters ME, Cohan A. Longformer: The long-document transformer. arXiv preprint arXiv:200405150. 2020;.
- Khandelwal U, Levy O, Jurafsky D, Zettlemoyer L, Lewis M. Generalization through Memorization: Nearest Neighbor Language Models. In: 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020. OpenReview.net; 2020.Available from: https://openreview.net/forum?id=HklBjCEKvH.
- Fu DY, Dao T, Saab KK, Thomas AW, Rudra A, Re C. Hungry Hungry Hippos: Towards Language Modeling with State Space Models. In: The Eleventh International Conference on Learning Representations; 2022.
- Liang PP, Wu C, Morency LP, Salakhutdinov R. Towards Understanding and Mitigating Social Biases in Language Models. In: Meila M, Zhang T, editors. Proceedings of the 38th International Conference on Machine Learning. vol. 139 of Proceedings of Machine Learning Research. PMLR; 2021. p. 6565–6576. Available from: https://proceedings.mlr.press/v139/liang21a.html.
Cited by
Language Model Behavior: A Comprehensive Survey
Chang, T. A., & Bergen, B. K. (2024) · Computational Linguistics · doi:10.1162/coli_a_00492 · Cited by 105
Folk Psychological Attributions of Consciousness to Large Language Models
Colombatto, C., & Fleming, S. M. (2023) · doi:10.31234/osf.io/5cnrv · Cited by 96
Folk psychological attributions of consciousness to large language models
Colombatto, C., & Fleming, S. M. (2024) · Neuroscience of Consciousness · doi:10.1093/nc/niae013 · Cited by 96
The influence of mental state attributions on trust in large language models
Colombatto, C., Birch, J., & Fleming, S. M. (2025) · Communications Psychology · doi:10.1038/s44271-025-00262-1 · Cited by 34
“Thetendency toassignmentalstatestoAIsystemsisindependentfromwhetherthese systemstrulypossessthem — and indeed, the extent to which current AI systems possess consciousness remains contentious in scienti fi c analysis 1 – 5 .”
Six Fallacies in Substituting Large Language Models for Human Participants
Lin, Z. (2025) · Advances in Methods and Practices in Psychological Science · doi:10.1177/25152459251357566 · Cited by 64
“Explicitly acknowledge in research designs and reports that current LLMs do not possess mental states or consciousness (Shardlow & Przybyla, 2024).”
Metadata
- Authors
- Shardlow, M., & Przybyła, P.
- Affiliations
- Manchester Metropolitan University (GB)
- Universitat Pompeu Fabra (ES)
- Corresponding author
- Shardlow, Matthew
- Venue
- PLoS ONE
- Publisher
- Public Library of Science (PLoS)
- Published
- 2024-12-04
- Volume
- 19
- Issue
- 12
- Pages
- e0307521
- ISSN
- 1932-6203 (ISSN-L)
- PMC
- PMC11617003
- OpenAlex
- W4405017998
- Cited by
- 16
- References
- 80
- Citations by year
- 2026: 7 · 2025: 6 · 2024: 2
- FWCI
- 6.11
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- PLoS ONE (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Repositori digital de la UPF (Universitat Pompeu Fabra) (publishedVersion, cc-by)
- e-space (Manchester Metropolitan University) (acceptedVersion)
- arXiv (Cornell University) (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- arXiv (Cornell University)
- Funders
- HORIZON EUROPE Framework Programme (101060930)
- Topics
- Social Sciences › Arts and Humanities › Philosophy › Philosophy and Theoretical Science
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Social Sciences › Arts and Humanities › History and Philosophy of Science › Cybernetics and Technology in Society
- Keywords
- large language models, sentience, consciousness, LaMDA, transformer models, integrated information theory, anthropomorphism, chatbots, ChatGPT, AI ethics, language modeling
- MeSH terms
- Consciousness, Humans, Language, Models, Theoretical, Natural Language Processing
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
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: Deanthropomorphising NLP: Can a language model be conscious? https://app.cochat.ai/papers/10.1371/journal.pone.0307521
- The paper
- Shardlow, M., & Przybyła, P. (2024). Deanthropomorphising NLP: Can a language model be conscious? PLoS ONE. https://doi.org/10.1371/journal.pone.0307521
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.
