On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜
Open accessThe DOI matches its Crossref record; 156 of the 159 references it cites were found in a registry or an index, and 3 are not checked yet.
Abstract
The past 3 years of work in NLP have been characterized by the development and deployment of ever larger language models, especially for English. BERT, its variants, GPT-2/3, and others, most recently Switch-C, have pushed the boundaries of the possible both through architectural innovations and through sheer size. Using these pretrained models and the methodology of fine-tuning them for specific tasks, researchers have extended the state of the art on a wide array of tasks as measured by leaderboards on specific benchmarks for English. In this paper, we take a step back and ask: How big is too big? What are the possible risks associated with this technology and what paths are available for mitigating those risks? We provide recommendations including weighing the environmental and financial costs first, investing resources into curating and carefully documenting datasets rather than ingesting everything on the web, carrying out pre-development exercises evaluating how the planned approach fits into research and development goals and supports stakeholder values, and encouraging research directions beyond ever larger language models.
Checked 2026-11-22 · 3 references not checked yet
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The text we found for this paper looks like another document, so the self-check wasn’t run.
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 · cited 5×
“[39] in particular noted that training on a large dataset and fine-tuning for specific tasks leads to strictly increasing results on the GLUE tasks [138] for English as the hyperparameters of the model were increased.”
BackgroundMethodResultInfluentialGlove: Global Vectors for Word Representation
Pennington, J., Socher, R., & Manning, C. (2014) · Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.3115/v1/d14-1162 · Cited by 34,071
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 · cited 4×
“documents linked to from Reddit [25], plus Wikipedia and a collection”
BackgroundMethodInfluentialRoBERTa: 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
Distilling the Knowledge in a Neural Network
Hinton, G., Vinyals, O., & Dean, J. (2015) · arXiv · doi:10.48550/arxiv.1503.02531 · Cited by 14,204 · cited 1×
“[26, 58], quantization [118, 153], factorized embedding parameterization and cross-layer parameter sharing [70], and progressive module replacing [146].”
Deep Contextualized Word Representations
Peters, M., Neumann, M., Iyyer, M., Gardner, M., Clark, C., Lee, K., & Zettlemoyer, L. (2018) · Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers) · doi:10.18653/v1/n18-1202 · Cited by 11,983
Proceedings of the 24th international conference on Machine learning
(2007) · doi:10.1145/1273496 · Cited by 11,630
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
The Mathematical Theory of Communication
Shannon, C. E., Weaver, W., & Wiener, N. (1950) · Physics Today · doi:10.1063/1.3067010 · Cited by 9,801
Research works printed but not linked: 48
- [4] American Psychological Association. 2019. Discrimination: What it is, and how to cope. https://www.apa.org/topics/discrimination (2019). Google Scholar
- 10.18653/v1 · DOI
- 10.18653/v1 · DOI
- [15] Ruha Benjamin. 2019. Race After Technology: Abolitionist Tools for the New Jim Code. Polity Press, Cambridge, UK. Crossref Google Scholar
- [16] Elettra Bietti and Roxana Vatanparast. 2020. Data Waste. Harvard International Law Journal 61 (2020). Google Scholar
- 10.18653/v1 · DOI
- 10.18653/v1 · DOI
- 10.1016/0010-0277(86) · doi:10.1016/0010-0277(86
- [38] Terrance de Vries, Ishan Misra, Changhan Wang, and Laurens van der Maaten. 2019. Does object recognition work for everyone?. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. 52--59. Google Scholar
- 10.18653/v1 · DOI
- [42] Ethan Fast, Tina Vachovsky, and Michael Bernstein. 2016. Shirtless and Dangerous: Quantifying Linguistic Signals of Gender Bias in an Online Fiction Writing Community. In Proceedings of the International AAAI Conference on Web and Social Media, Vol. 10. Google Scholar
- 10.18653/v1 · DOI
- [45] Darja Fišer, Ruihong Huang, Vinodkumar Prabhakaran, Rob Voigt, Zeerak Waseem, and Jacqueline Wernimont (Eds.). 2018. Proceedings of the 2nd Workshop on Abusive Language Online (ALW2). Association for Computational Linguistics, Brussels, Belgium. https://www.aclweb.org/anthology/W18-5100 Google Scholar
- [50] Batya Friedman, Peter H. Kahn, Jr., and Alan Borning. 2006. Value sensitive design and information systems. In Human-Computer Interaction in Management Information Systems: Foundations, P Zhang and D Galletta (Eds.). M. E. Sharpe, Armonk NY, 348--372. Google Scholar
- 10.18653/v1 · DOI
- 10.18653/v1 · DOI
- 10.18653/v1 · DOI
- [67] Brendan Kennedy, Drew Kogon, Kris Coombs, Joseph Hoover, Christina Park, Gwenyth Portillo-Wightman, Aida Mostafazadeh Davani, Mohammad Atari, and Morteza Dehghani. 2018. A typology and coding manual for the study of hate-based rhetoric. PsyArXiv. July 18 (2018). Google Scholar
- [77] Marianna Martindale and Marine Carpuat. 2018. Fluency Over Adequacy: A Pilot Study in Measuring User Trust in Imperfect MT. In Proceedings of the 13th Conference of the Association for Machine Translation in the Americas (Volume 1: Research Track). Association for Machine Translation in the Americas, Boston, MA, 13--25. https://www.aclweb.org/anthology/W18-1803 Google Scholar
- [81] Douglas M McLeod. 2007. News coverage and social protest: How the media's protect paradigm exacerbates social conflict. Journal of Dispute Resolution (2007), 185. Google Scholar
- [87] Robert C. Moore and William Lewis. 2010. Intelligent Selection of Language Model Training Data. In Proceedings of the ACL 2010 Conference Short Papers. Association for Computational Linguistics, Uppsala, Sweden, 220--224. https://www.aclweb.org/anthology/P10-2041 Google Scholar
- 10.18653/v1 · DOI
- [92] Maggie Nelson. 2015. The Argonauts. Graywolf Press, Minneapolis. Google Scholar
- 10.18653/v1 · DOI
- [96] David Ortiz, Daniel Myers, Eugene Walls, and Maria-Elena Diaz. 2005. Where do we stand with newspaper data? Mobilization: An International Quarterly 10, 3 (2005), 397--419. Google Scholar
- 10.3115/v1 · doi:10.3115/v1
- 10.18653/v1 · DOI
- 10.18653/v1 · DOI
- [106] Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. 2019. Language models are unsupervised multitask learners. OpenAI Blog 1, 8 (2019), 9. Google Scholar
- 10.18653/v1 · DOI
- [109] Sarah T. Roberts, Joel Tetreault, Vinodkumar Prabhakaran, and Zeerak Waseem (Eds.). 2019. Proceedings of the Third Workshop on Abusive Language Online. Association for Computational Linguistics, Florence, Italy. https://www.aclweb.org/anthology/W19-3500 Google Scholar
- [112] Corby Rosset. 2020. Turing-NLG: A 17-billion-parameter language model by Microsoft. Microsoft Blog (2020). Google Scholar
- 10.18653/v1 · DOI
- 10.18653/v1 · DOI
- [121] Joonbo Shin, Yoonhyung Lee, and Kyomin Jung. 2019. Effective Sentence Scoring Method Using BERT for Speech Recognition. In Asian Conference on Machine Learning. 1081--1093. Google Scholar
- [124] Karen Spärck Jones. 2004. Language modelling's generative model: Is it rational? Technical Report. Computer Laboratory, University of Cambridge. Google Scholar
- [127] Katrina Srigley and Lorraine Sutherland. 2019. Decolonizing, Indigenizing, and Learning Biskaaybiiyang in the Field: Our Oral History Journey1. The Oral History Review (2019). Google Scholar
- 10.18653/v1 · DOI
- 10.18653/v1 · DOI
- 10.18653/v1 · DOI
- 10.18653/v1 · DOI
- 10.18653/v1 · DOI
- 10.18653/v1 · DOI
- 10.18653/v1 · DOI
- [150] Zhilin Yang, Zihang Dai, Yiming Yang, Jaime Carbonell, Russ R Salakhutdinov, and Quoc V Le. 2019. Xlnet: Generalized autoregressive pretraining for language understanding. In Advances in neural information processing systems. 5753--5763. Google Scholar
- 10.18653/v1 · DOI
- 10.18653/v1 · DOI
- 10.18653/v1 · DOI
Cited by
Items 401–500 of 6,931
From Ethics Washing to Ethics Bashing: A Moral Philosophy View on Tech Ethics
Bietti, E. (2021) · Journal of Social Computing · doi:10.23919/jsc.2021.0031 · Cited by 187
Data-Centric Green Artificial Intelligence: A Survey
Salehi, S., & Schmeink, A. (2023) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2023.3315272 · Cited by 106
“This presents a challenge to the sustainability of AI and highlights the need for environmentally responsible energy generation and consumption methods in the field [10].”
The mechanisms of AI hype and its planetary and social costs
Markelius, A., Wright, C., Kuiper, J., Delille, N., & Kuo, Y.-T. (2024) · AI and Ethics · doi:10.1007/s43681-024-00461-2 · Cited by 76
Kim, T., Shin, D., Kim, Y.-H., & Hong, H. (2024) · Proceedings of the CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3642693 · Cited by 75
“Additionally, it remains challenging to identify and fx biases within LLM outputs consistently [2, 14].”
Liu, V., & Yin, Y. (2024) · Discover Artificial Intelligence · doi:10.1007/s44163-024-00149-w · Cited by 138
Ariai, F., Mackenzie, J., & Demartini, G. (2025) · ACM Computing Surveys · doi:10.1145/3777009 · Cited by 284
Foundation Models and Fair Use
Henderson, P., Li, X., Jurafsky, D., Hashimoto, T., Lemley, M. A., & Liang, P. (2023) · doi:10.2139/ssrn.4404340 · Cited by 57
Activity-difference training of deep neural networks using memristor crossbars
Yi, S.-I., Kendall, J. D., Williams, R. S., & Kumar, S. (2022) · Nature Electronics · doi:10.1038/s41928-022-00869-w · Cited by 75
Agentic Large Language Models, a Survey
Plaat, A., Van Duijn, M., Van Stein, N., Preuss, M., Van der Putten, P., & Batenburg, K. J. (2025) · Journal of Artificial Intelligence Research · doi:10.1613/jair.1.18675 · Cited by 75
Instructors as Innovators: A future-focused approach to new AI learning opportunities, with prompts
Mollick, E., & Mollick, L. (2024) · arXiv · doi:10.48550/arxiv.2407.05181 · Cited by 75
Video PreTraining (VPT): Learning to Act by Watching Unlabeled Online Videos
Baker, B., Akkaya, I., Zhokhov, P., Huizinga, J., Tang, J., Ecoffet, A., Houghton, B., Sampedro, R., & Clune, J. (2022) · arXiv · doi:10.48550/arxiv.2206.11795 · Cited by 75
“[67] Emily M Bender, Timnit Gebru, Angelina McMillan-Major, and Shmargaret Shmitchell.”
BackgroundAgrawal, A., Ali, A., & Boyd, S. (2021) · Foundations and Trends in Machine Learning · doi:10.1561/2200000090 · Cited by 74
A Survey of Race, Racism, and Anti-Racism in NLP
Field, A., Blodgett, S. L., Waseem, Z., & Tsvetkov, Y. (2021) · Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers) · doi:10.18653/v1/2021.acl-long.149 · Cited by 74
“Relatedly, there are numerous racialized effects that NLP research can have that we do not address in this work; for example, Bender et al. (2021) and Strubell et al. (2019) discuss the environmental costs of training large language models, and how global warming disproportionately affects…”
BackgroundMethodSecurity and Privacy on Generative Data in AIGC: A Survey
Wang, T., Zhang, Y., Qi, S., Zhao, R., Xia, Z., & Weng, J. (2024) · ACM Computing Surveys · doi:10.1145/3703626 · Cited by 157
WinoQueer: A community-in-the-loop benchmark for anti-LGBTQ+ bias in large language models
Felkner, V., Chang, H.-C. H., Jang, E., & May, J. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2023.acl-long.507 · Cited by 74
Bojić, L., Zagovora, O., Zelenkauskaitė, A., Vuković, V., Čabarkapa, M., Jerković, S. V., & Jovančević, A. (2025) · Scientific Reports · doi:10.1038/s41598-025-96508-3 · Cited by 173
“However, challenges remain regarding the ethical and practical implications of relying on LLMs. Concerns include model bias, the interpretability of results, and the tendency of LLMs to produce plausible but incorrect or biased outputs 32,33 .”
Creating a large language model of a philosopher
Schwitzgebel, E., Schwitzgebel, D., & Strasser, A. (2024) · Mind & Language · doi:10.1111/mila.12466 · Cited by 73
Evaluating Generative AI in Mental Health: Systematic Review of Capabilities and Limitations
Wang, L., Bhanushali, T., Huang, Z., Yang, J., Badami, S., & Hightow-Weidman, L. (2025) · JMIR Mental Health · doi:10.2196/70014 · Cited by 73
“Current models are largely trained on data from Western, English-speaking sources, which introduce inherent biases into their responses [40,41].”
Kolides, A., Nawaz, A., Rathor, A., Beeman, D., Hashmi, M., Fatima, S., Berdik, D., Al‐Ayyoub, M., & Jararweh, Y. (2023) · Simulation Modelling Practice and Theory · doi:10.1016/j.simpat.2023.102754 · Cited by 105
Spellburst: A Node-based Interface for Exploratory Creative Coding with Natural Language Prompts
Angert, T., Suzara, M., Han, J., Pondoc, C., & Subramonyam, H. (2023) · Proceedings of the 36th Annual ACM Symposium on User Interface Software and Technology · doi:10.1145/3586183.3606719 · Cited by 72
Reducing Conversational Agents’ Overconfidence Through Linguistic Calibration
Mielke, S. J., Szlam, A., Dinan, E., & Boureau, Y.-L. (2022) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00494 · Cited by 72
How Far Are We? The Triumphs and Trials of Generative AI in Learning Software Engineering
Choudhuri, R., Liu, D., Steinmacher, I., Gerosa, M., & Sarma, A. (2024) · Proceedings of the IEEE/ACM 46th International Conference on Software Engineering · doi:10.1145/3597503.3639201 · Cited by 72
Reducing conversational agents' overconfidence through linguistic calibration
Mielke, S. J., Szlam, A., Dinan, E., & Boureau, Y.-L. (2020) · arXiv · doi:10.48550/arxiv.2012.14983 · Cited by 72
Citation: A Key to Building Responsible and Accountable Large Language Models
Huang, J., & Chang, K. (2024) · Findings of the Association for Computational Linguistics: NAACL 2024 · doi:10.18653/v1/2024.findings-naacl.31 · Cited by 72
LaMPost: Design and Evaluation of an AI-assisted Email Writing Prototype for Adults with Dyslexia
Goodman, S. M., Buehler, E., Clary, P., Coenen, A., Donsbach, A., Horne, T. N., Lahav, M., MacDonald, R., Michaels, R. B., Narayanan, A., et al. (2022) · Proceedings of the 24th International ACM SIGACCESS Conference on Computers and Accessibility · doi:10.1145/3517428.3544819 · Cited by 71
“Of particular note is that such models risk generating factually incorrect, ofensive, or stereotyped text since they are trained on content from the internet [5, 65].”
BackgroundExploring the psychology of LLMs’ moral and legal reasoning
Almeida, G. F. C. F., Nunes, J. L., Engelmann, N., Wiegmann, A., & Araújo, M. D. (2024) · Artificial Intelligence · doi:10.1016/j.artint.2024.104145 · Cited by 70
Earth for AI: A Political Ecology of Data-Driven Climate Initiatives
Nost, E., & Colven, E. (2022) · Geoforum · doi:10.1016/j.geoforum.2022.01.016 · Cited by 71
A tutorial on open-source large language models for behavioral science
Hussain, Z., Binz, M., Mata, R., & Wulff, D. U. (2024) · Behavior Research Methods · doi:10.3758/s13428-024-02455-8 · Cited by 141
“…scientific and philosophical literature pertaining to the (evaluation of) language model capabilities (e.g. Mitchell & Krakauer, 2023; Turing, 1950; Bender et al., 2021; Searle, 1980; Günther et al., 2019) – a literature that has by no means reached a consensus – to guard against snap judgments…”
BackgroundStandards, frameworks, and legislation for artificial intelligence (AI) transparency
Lund, B. D., Orhan, Z., Mannuru, N. R., Bevara, R. V. K., Porter, B., Vinaih, M. K., & Bhaskara, P. (2025) · AI and Ethics · doi:10.1007/s43681-025-00661-4 · Cited by 139
“This includes details regarding the data utilized for training, the methodologies employed in system design, and the prospective societal implications [6].”
Cohn, M., Pushkarna, M., Olanubi, G. O., Moran, J. M., Padgett, D., Mengesha, Z., & Heldreth, C. (2024) · Extended Abstracts of the CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613905.3650818 · Cited by 71
“Additionally, advances in neural TTS mean that computer generated voices ‘sound’ increasingly human-like to listeners [11].”
McCoy, R. T., Smolensky, P., Linzen, T., Gao, J., & Celikyilmaz, A. (2023) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00567 · Cited by 70
“This acronym is also a nod to Bender et al.’s (2021) comparison of LMs to another utterance-generating bird, the parrot. to be a major concern in practice.”
The poverty of ethical AI: impact sourcing and AI supply chains
Muldoon, J., Cant, C., Graham, M., & Ustek Spilda, F. (2025) · AI & SOCIETY · doi:10.1007/s00146-023-01824-9 · Cited by 88
“…increasingly popular with many companies, there are currently few regulatory requirements for AI companies to disclose any information concerning the provenance of their datasets or other critical infrastructure related to AI systems (Bender et al. 2021; Gebru et al. 2021; Marshall et al. 2015).”
Impact of Tokenization on Language Models: An Analysis for Turkish
Toraman, Ç., Yılmaz, E. H., Şahinuç, F., & Ozcelik, O. (2023) · ACM Transactions on Asian and Low-Resource Language Information Processing · doi:10.1145/3578707 · Cited by 207
“There is an increasing environmental awareness among the Machine Learning community about responsible training, such as the carbon footprint of extensive model training [Bender et al. 2021; Henderson et al. 2020].”
Irfan, B., Kuoppamäki, S., Hosseini, A., & Skantze, G. (2025) · Autonomous Robots · doi:10.1007/s10514-025-10190-y · Cited by 108
“…the template of the utterance was quite often the same as well (e.g., “I love to... too”, “I liketo... too”, “That sounds likea...”), oftenmirroring the user’s preferences and likes as its own, rarely expressing a different preference or idea, like a ‘stochastic parrot’ (Bender et al., 2021).”
McCoy, R. T., Smolensky, P., Linzen, T., Gao, J., & Celikyilmaz, A. (2021) · arXiv · doi:10.48550/arxiv.2111.09509 · Cited by 70
Probing Toxic Content in Large Pre-Trained Language Models
Ousidhoum, N., Zhao, X., Fang, T., Song, Y., & Yeung, D.-Y. (2021) · Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers) · doi:10.18653/v1/2021.acl-long.329 · Cited by 69
“…the sensitivity of language models towards stereo-typical content involves the construction of assessment benchmarks (Nadeem et al., 2020; Tay et al., 2020; Gehman et al., 2020) in addition to the study of the potential risks associated with the use and deployment of PTLMs (Bender et al., 2021).”
BackgroundBahji, A., Ación, L., Laslett, A., & Adinoff, B. H. (2022) · Nordic Studies on Alcohol and Drugs · doi:10.1177/14550725221102227 · Cited by 94
ChatGPT is no Stochastic Parrot. But it also Claims that 1 is Greater than 1
Arkoudas, K. (2023) · Philosophy & Technology · doi:10.1007/s13347-023-00619-6 · Cited by 80
Extracting Domain Models from Textual Requirements in the Era of Large Language Models
Arulmohan, S., Meurs, M., & Mosser, S. (2023) · ACM/IEEE International Conference on Model Driven Engineering Languages and Systems Companion (MODELS-C) · doi:10.1109/models-c59198.2023.00096 · Cited by 82
Demystifying ChatGPT: An In-depth Survey of OpenAI’s Robust Large Language Models
Bhattacharya, P., Prasad, V. K., Verma, A., Gupta, D. K., Sapsomboon, A., Viriyasitavat, W., & Dhiman, G. (2024) · Archives of Computational Methods in Engineering · doi:10.1007/s11831-024-10115-5 · Cited by 118
“Research has indicated that these models can inadvertently propagate stereotypes or misrepresentations found within their training data, posing serious ethical considerations [60, 61].”
Henriksen, D., Creely, E., Gruber, N., & Leahy, S. (2025) · Journal of Teacher Education · doi:10.1177/00224871251325058 · Cited by 69
“…generative pre-trained transformer (GPT) models have been called “stochastic parrots” because they statistically predict and replicate language patterns without true comprehension or intentionality, often generating convincing yet erroneous, biased, or harmful content (Bender et al., 2021).”
Kanitz, R., Gonzalez, K., Briker, R., & Straatmann, T. (2023) · The Journal of Applied Behavioral Science · doi:10.1177/00218863231168974 · Cited by 69
“Even when GAI systems are programed to refuse discriminant prompts or fi lter out illegal suggestions, the responses might be problematic due to a bias in the underlying text dataset (Bender et al., 2021).”
BackgroundPerturbation augmentation for fairer NLP
Qian, R., Ross, C., Fernandes, J., Smith, E. M., Kiela, D., & Williams, A. (2022) · Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2022.emnlp-main.646 · Cited by 69
Hagendorff, T., Dasgupta, I., Binz, M., Chan, S. C. Y., Lampinen, A., Wang, J. X., Akata, Z., & Schulz, E. (2023) · arXiv · doi:10.48550/arxiv.2303.13988 · Cited by 68
Agentic AI in Education: State of the Art and Future Directions
Kostopoulos, G., Gkamas, V., Rigou, M., & Kotsiantis, S. (2025) · IEEE Access · doi:10.1109/access.2025.3620473 · Cited by 69
“Restrictions apply. unsubstantiated or false claims—commonly referred to as hallucinations—particularly when reasoning about new or ambiguous topics [47].”
How should the advancement of large language models affect the practice of science?
Binz, M., Alaniz, S., Roskies, A., Aczel, B., Bergstrom, C. T., Allen, C., Schad, D., Wulff, D., West, J. D., Zhang, Q., et al. (2025) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2401227121 · Cited by 74
Self-citationAmugongo, L. M., Kriebitz, A., Boch, A., & Lütge, C. (2025) · AI and Ethics · doi:10.1007/s43681-023-00331-3 · Cited by 68
“Lastly, LLMs are trained with large datasets of text from the internet is dominated by hegemonic viewpoints and encode biases and inequalities, disproportionally affecting marginalized communities [82].”
Queer In AI: A Case Study in Community-Led Participatory AI
Queerinai, O. O., Ovalle, A., Subramonian, A., Singh, A., Voelcker, C., Sutherland, D. J., Locatelli, D., Breznik, E., Klubicka, F., Yuan, H., et al. (2023) · 2023 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3593013.3594134 · Cited by 68
Ma, Z., Mei, Y., Long, Y., Su, Z., & Gajos, K. Z. (2024) · Proceedings of the CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3642482 · Cited by 68
“LGBTQ+ users, who are often underrepresented in the training datasets, can encounter unintentional reinforcement of damaging narratives with regard to their identities [9, 34].”
BackgroundInfluentialA survey on multilingual large language models: corpora, alignment, and bias
Xu, Y., Hu, L., Zhao, J., Qiu, Z., Xu, K., Ye, Y., & Gu, H. (2025) · Frontiers of Computer Science · doi:10.1007/s11704-024-40579-4 · Cited by 225
“MLLMs Input O u t p u t O u t pu t”
Instructors as Innovators: a Future-focused Approach to New AI Learning Opportunities, With PromptsPreprint
Mollick, E., & Mollick, L. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4802463 · Cited by 172
The use of generative AI by students with disabilities in higher education
Zhao, X., Cox, A., & Chen, X. (2025) · The Internet and Higher Education · doi:10.1016/j.iheduc.2025.101014 · Cited by 134
Dentella, V., Günther, F., & Leivada, E. (2023) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2309583120 · Cited by 67
Truth Serum: Poisoning Machine Learning Models to Reveal Their Secrets
Tramèr, F., Shokri, R., Joaquin, A. S., Le, H., Jagielski, M., Hong, S., & Carlini, N. (2022) · ACM SIGSAC Conference on Computer and Communications Security (CCS) · doi:10.1145/3548606.3560554 · Cited by 152
“In contrast, the ability to poison a large language model’s training set may be more accessible, especially since these models are typically trained on large minimally curated data sources [5, 60].”
Tabi: An Efficient Multi-Level Inference System for Large Language Models
Wang, Y., Chen, K., Tan, H., & Guo, K. (2023) · European Conference on Computer Systems · doi:10.1145/3552326.3587438 · Cited by 149
“One key to their advancement is the sheer volume of parameters and we are seeing a trend of building ever larger language models (LLMs) to push the limit of NLP performance [9, 34].”
BackgroundEnvisioning Information Access Systems: What Makes for Good Tools and a Healthy Web?
Shah, C., & Bender, E. M. (2024) · ACM Transactions on the Web · doi:10.1145/3649468 · Cited by 66
“In the current narrative around ‘generative AI’ and specifically LLM-based models, several issues have been raised [13, 77, 88] about their potential costs and harms.”
Self-citationJalali, M. S., & Akhavan, A. (2024) · System Dynamics Review · doi:10.1002/sdr.1772 · Cited by 109
Klarin, J., Hoff, E., Larsson, A., & Daukantaitė, D. (2024) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2024.1415782 · Cited by 67
Stakeholder Perspectives on the Ethics of AI in Distance-Based Higher Education
Holmes, W., Iniesto, F., Anastopoulou, S., & Boticario, J. G. (2023) · The International Review of Research in Open and Distributed Learning · doi:10.19173/irrodl.v24i2.6089 · Cited by 67
“AI has also been challenged for the hyperbole and the many myths surrounding it (e.g., Bender et al., 2021).”
“They only care to show us the wheelchair”: disability representation in text-to-image AI models
Mack, K. A., Qadri, R., Denton, R., Kane, S. K., & Bennett, C. L. (2024) · Proceedings of the CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3642166 · Cited by 67
Rahimzadeh, V., Kostick-Quenet, K., Blumenthal Barby, J., & McGuire, A. L. (2023) · The American Journal of Bioethics · doi:10.1080/15265161.2023.2233358 · Cited by 66
Smolensky, P., McCoy, R. T., Fernandez, R., Goldrick, M., & Gao, J. (2022) · AI Magazine · doi:10.1002/aaai.12065 · Cited by 68
AI for the public. How public interest theory shifts the discourse on AI
Züger, T., & Asghari, H. (2022) · AI & Society · doi:10.1007/s00146-022-01480-5 · Cited by 108
“As Bender et al. (2021) argue models need to add enough public value to warrant their additional computational and environmental costs.”
BackgroundHow Beginning Programmers and Code LLMs (Mis)read Each Other
Nguyen, S., Babe, H. M., Zi, Y., Guha, A., Anderson, C. J., & Feldman, M. Q. (2024) · Proceedings of the CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3642706 · Cited by 66
Software Engineering Education Must Adapt and Evolve for an LLM Environment
Kirova, V. D., Ku, C. S., Laracy, J. R., & Marlowe, T. J. (2024) · Proceedings of the 55th ACM Technical Symposium on Computer Science Education V. 1 · doi:10.1145/3626252.3630927 · Cited by 67
“It is broadly recognized that these models, while powerful, are far from perfect, with notable concerns surrounding their susceptibility to biases, misuse, and their “black box” nature [8].”
BackgroundHard choices in artificial intelligence
Dobbe, R., Krendl Gilbert, T., & Mintz, Y. (2021) · Artificial Intelligence · doi:10.1016/j.artint.2021.103555 · Cited by 65
“…various scholars have advocated about the dangers of abstraction in AI systems [Selbst et al., 2019], and pointed to the dangers of how imposing such abstractions can reify inequities resulting from institutional racism [Benjamin, 2019b] and harm marginalized communities [Bender et al., 2021].”
BackgroundCodexDB: synthesizing code for query processing from natural language instructions using GPT-3 codex
Trummer, I. (2022) · Proceedings of the VLDB Endowment · doi:10.14778/3551793.3551841 · Cited by 65
Critical AI: A Field in Formation
Raley, R., & Rhee, J. (2023) · American Literature · doi:10.1215/00029831-10575021 · Cited by 104
“2021), and considered attempts to draw attention to encoded bias (Bender et al. 2021).”
BackgroundInfluentialChatGPT in society: emerging issues
Farina, M., & Lavazza, A. (2023) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2023.1130913 · Cited by 66
“Despite many experts in the field have already labeled these models as mere stochastic parrots (Bender et al., 2021), it is undeniable that the generative performances of ChatGPT (and presumably of evenmore advanced future models) are extraordinary in many ways (Doshi et al.”
BackgroundTraining data influence analysis and estimation: a survey
Hammoudeh, Z., & Lowd, D. (2024) · Machine Learning · doi:10.1007/s10994-023-06495-7 · Cited by 233
Do large language models have a legal duty to tell the truth?
Wachter, S., Mittelstadt, B., & Russell, C. (2024) · Royal Society Open Science · doi:10.1098/rsos.240197 · Cited by 66
Initial policy considerations for generative artificial intelligence
(2023) · OECD Artificial Intelligence Papers · doi:10.1787/fae2d1e6-en · Cited by 36
How Does Narrow AI Impact Human Creativity?
Marrone, R., Cropley, D., & Medeiros, K. (2026) · Creativity Research Journal · doi:10.1080/10400419.2024.2378264 · Cited by 65
“The likening, by Bender et al. (2021), of large models such as GPT-3 to a “stochastic parrot” may therefore be a little unfair.”
Research can help to tackle AI-generated disinformation
Feuerriegel, S., DiResta, R., Goldstein, J. A., Kumar, S., Lorenz-Spreen, P., Tomz, M., & Pröllochs, N. (2023) · Nature Human Behaviour · doi:10.1038/s41562-023-01726-2 · Cited by 64
Confronting Power and Corporate Capture at the FAccT Conference
Young, M., Katell, M., & Krafft, P. M. (2022) · 2022 ACM Conference on Fairness Accountability and Transparency · doi:10.1145/3531146.3533194 · Cited by 63
Beyond Traditional Assessment: Exploring the Impact of Large Language Models on Grading Practices
Fagbohun, O., Iduwe, N. P., Abdullahi, M., Ifaturoti, A., & Nwanna, M. (2024) · Journal of Artificial Intelligence Machine Learning and Data Science · doi:10.51219/jaimld/oluwole-fagbohun/19 · Cited by 112
“For instance, LLMs predominantly trained on data from native English speakers may inadequately address the nuances of responses from English language learners 62,67-69 .”
Beyond Discrimination: Generative AI Applications and Ethical Challenges in Forensic Psychiatry
Tortora, L. (2024) · Frontiers in Psychiatry · doi:10.3389/fpsyt.2024.1346059 · Cited by 91
Almuhanna, M. A. (2024) · Education and Information Technologies · doi:10.1007/s10639-024-13257-y · Cited by 137
“…previously to produce particular answers, but the reality is constantly evolving faster than the updates of these tools and systems and their programming to provide updated answers (Walczak & Cellary, 2023), especially as training generative AI models require high cost (Bender et al., 2021).”
Mitigating Language-Dependent Ethnic Bias in BERT
Ahn, J., & Oh, A. (2021) · Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2021.emnlp-main.42 · Cited by 64
“Moreover, since the data used in training language models are mainly based on the texts from the Internet, language models are more likely to represent and reflect only the skewed population of the language users (Bender et al., 2021).”
Jang, J., Moharana, S., Carrington, P., & Begel, A. (2024) · Proceedings of the CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3642894 · Cited by 64
“Despite these modern advances, it remains the case that the risks high-lighted by Bender et al. [9], of exposure to encoded biases and discriminatory viewpoints, lack of grounded communication, and overreliance on automated systems likely still apply with current and future human-LLM interactions, potentially exposing…”
BackgroundSpasić, A. J., & Jankovic, D. S. (2023) · International Scientific Conference on Information, Communication and Energy Systems and Technologies (ICEST) · doi:10.1109/icest58410.2023.10187269 · Cited by 77
ChatGPT Isn't Magic: The Hype and Hypocrisy of Generative Artificial Intelligence (AI) Rhetoric
Leaver, T., & Srdarov, S. (2023) · M/C Journal · doi:10.5204/mcj.3004 · Cited by 62
Barman, K. G., Wood, N. G., & Pawłowski, P. (2024) · Ethics and Information Technology · doi:10.1007/s10676-024-09778-2 · Cited by 86
“There are widely recognized issues such as bias (Abid et al., 2021; Ferrara, 2023), inaccuracies (Bender et al., 2021), plagiarism (Lee et al., 2023), and the spreading of misinformation (Pan et al., 2023), but there are far more common and potentially impactful issues as well.”
Large Process Models: A Vision for Business Process Management in the Age of Generative AI
Kampik, T., Warmuth, C., Rebmann, A., Agam, R., Egger, L. N. P., Gerber, A., Hoffart, J., Kolk, J., Herzig, P., Decker, G., et al. (2024) · Künstliche Intell. · doi:10.1007/s13218-024-00863-8 · Cited by 102
“In contrast, both Step 2 and Step 3 pose substantial challenges regarding the management of ML models, such as a potential process execution traces-based foundation model, as well as regarding the variability of results such as predictions and action recommendations.”
Transitioning from MLOps to LLMOps: Navigating the Unique Challenges of Large Language Models
Pahune, S. A., & Akhtar, Z. (2025) · Information · doi:10.3390/info16020087 · Cited by 116
Writer-Defined AI Personas for On-Demand Feedback Generation
Benharrak, K., Zindulka, T., Lehmann, F., Heuer, H., & Buschek, D. (2024) · Proceedings of the CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3642406 · Cited by 63
Updating “The Future of Coding”: Qualitative Coding with Generative Large Language Models
Than, N., Fan, L., Law, T., Nelson, L. K., & McCall, L. (2025) · Sociological Methods & Research · doi:10.1177/00491241251339188 · Cited by 63
“This collaborative environment accelerates innovation but also invites the kind of broader scrutiny needed to mitigate well-documented biases in language models (Bender et al. 2021; Ollion et al. 2024; Spirling 2023).”
A review of LLMs and their applications in the architecture, engineering and construction industry
Kampelopoulos, D., Tsanousa, A., Vrochidis, S., & Kompatsiaris, I. (2025) · Artificial Intelligence Review · doi:10.1007/s10462-025-11241-7 · Cited by 69
“As a result, the outputs may contain social and cultural preju - dices making the model behavior biased and less neutral (Bender et al. 2021 ).”
Ulfsnes, R., Moe, N. B., Stray, V., & Skarpen, M. (2024) · Generative AI for Effective Software Development · doi:10.1007/978-3-031-55642-5_10 · Cited by 62
Multilingualism and AI: The Regimentation of Language in the Age of Digital Capitalism
Schneider, B. (2022) · Signs and Society · doi:10.1086/721757 · Cited by 114
“It could be shown 63,000 texts used for training GPT3 were from “ subreddits ” that had been banned due to their problematic content (Bender et al. 2021, 613).”
InfluentialCulturally Responsive Assessment: Provisional Principles
Walker, M. E., Olivera‐Aguilar, M., Lehman, B., Laitusis, C., Guzman‐Orth, D., & Gholson, M. (2023) · ETS Research Report Series · doi:10.1002/ets2.12374 · Cited by 62
“Research on artificial intelligence has flagged the underrepresentation of certain subgroups in the data used to train the models, resulting in algorithms that reproduce existing biases (Bender et al., 2021; Nordling, 2019).”
Evaluating large language models in analysing classroom dialogue
Long, Y., Luo, H., & Zhang, Y. (2024) · npj Science of Learning · doi:10.1038/s41539-024-00273-3 · Cited by 117
“This substantial time ef fi ciency offered by ChatGPT emphasises its potential to signi fi cantly expedite the coding process in educational research, aligning with fi ndings from previous studies that highlight the ef fi ciency of AI in data analysis tasks 38,39 .”
MethodCritical GenAI Literacy: Postdigital Configurations
Rapanta, C., Bhatt, I., Bozkurt, A., Chubb, L. A., Erb, C., Forsler, I., Gravett, K., Koole, M., Lintner, T., Örtegren, A., et al. (2025) · Postdigital Science and Education · doi:10.1007/s42438-025-00573-w · Cited by 77
“These biases are not just technical issues; they represent deeper forms of epistemic injustice, where dominant Western knowledge is privileged while alternative perspectives are sidelined (Bender et al. 2021; Lee et al. 2024).”
Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Task
Li, K., Hopkins, A. K., Bau, D., Viégas, F., Pfister, H., & Wattenberg, M. (2022) · arXiv · doi:10.48550/arxiv.2210.13382 · Cited by 61
“This issue is of practical concern, since relying on spurious correlations may lead to problems on out-of-distribution data (Bender et al., 2021; Floridi & Chiriatti, 2020).”
BackgroundExploring TensorRT to Improve Real-Time Inference for Deep Learning
Zhou, Y., & Yang, K. (2022) · 2022 IEEE 24th Int Conf on High Performance Computing & Communications; 8th Int Conf on Data Science & Systems; 20th Int Conf on Smart City; 8th Int Conf on Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys) · doi:10.1109/hpcc-dss-smartcity-dependsys57074.2022.00299 · Cited by 61
Evaluation Gaps in Machine Learning Practice
Hutchinson, B., Rostamzadeh, N., Greer, C., Heller, K., & Prabhakaran, V. (2022) · 2022 ACM Conference on Fairness Accountability and Transparency · doi:10.1145/3531146.3533233 · Cited by 62
A Medical Ethics Framework for Conversational Artificial Intelligence
Fournier-Tombs, E., & McHardy, J. (2023) · Journal of Medical Internet Research · doi:10.2196/43068 · Cited by 62
“However, conversational AI is unique in that it also features specific risks related to large language models [18].”
Democratising AI: Multiple Meanings, Goals, and Methods
Seger, E., Ovadya, A., Siddarth, D., Garfinkel, B., & Dafoe, A. (2023) · Proceedings of the 2023 AAAI/ACM Conference on AI, Ethics, and Society · doi:10.1145/3600211.3604693 · Cited by 73
Lindroth, H. L., Nalaie, K., Raghu, R., Ayala, I. N., Busch, C., Bhattacharyya, A., Franco, P. M., Diedrich, D. A., Pickering, B. W., & Herasevich, V. (2024) · Journal of Imaging · doi:10.3390/jimaging10040081 · Cited by 87
Reconciling the contrasting narratives on the environmental impact of large language models
Ren, S., Tomlinson, B., Black, R. W., & Torrance, A. W. (2024) · Scientific Reports · doi:10.1038/s41598-024-76682-6 · Cited by 128
Metadata
- Authors
- Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S.
- Affiliations
- University of Washington (US)
- Black in AI, Palo Alto, CA, USA
- The Aether
- Venue
- Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency
- Publisher
- ACM
- Published
- 2021-03
- Pages
- 610-623
- OpenAlex
- W3133702157
- Cited by
- 6,988
- References
- 117
- Citations by year
- 2026: 2484 · 2025: 1887 · 2024: 1153 · 2023: 842 · 2022: 384 · 2021: 229
- FWCI
- 300.72
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- ACM Conference on Fairness, Accountability, and Transparency (FAccT) (publishedVersion, cc-by)
- Creighton Digital Repository (CDR) (submittedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Keywords
- large language models, model size, AI risks, financial costs, data curation, dataset documentation, stakeholder value, BERT, GPT, fine-tuning
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
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.