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 6,501–6,600 of 6,931
Using Artificial Intelligence as Homework Support in Higher Education
Çoruk, H., & ERDEMİR, T. (2026) · Van Yüzüncü Yıl Üniversitesi Eğitim Fakültesi dergisi/Yüzüncü Yıl Üniversitesi Eğitim Fakültesi dergisi · doi:10.33711/yyuefd.1843076
GS-Chaff: Multi-Agent Prompt-Level Semantic Chaffing for Privacy-Preserving LLM Inference
Zhou, Q., Wang, Z., Yue, Z., Cai, L., Liu, K., & Qin, C. (2026) · Sensors · doi:10.3390/s26175385 · Cited by 1
“For high-stakes information, such as legal, healthcare, and financial content, this level of utility may pose serious risks to privacy and data protection [5].”
Banda, L., & Feukeu, E. A. (2026) · Telecom · doi:10.3390/telecom7050107
Large Language Models as Distributional Baselines for Language Tasks
Trott, S., Michaelov, J. A., Jones, C., Chang, T. A., & Bergen, B. K. (2026) · Open Mind · doi:10.1162/opmi.a.373
“The training data for most language models consists primarily of written text from the Internet (Bender et al., 2021; Grieve et al., 2025), which is not representative of all language varieties that constitute the broader language or languages in question (Grieve et al., 2025).”
From symbols to geometry: an indexical epistemology of generative AI
Levin, I. (2026) · Synthese · doi:10.1007/s11229-026-05778-5 · Cited by 1
“Bender et al. (2021) sharpen this critique into the stochastic parrots thesis: generative systems merely recombine statistical patterns, and any appearance of genuine novelty is illusory.”
The TRACE Model for Graduate Research Ethics, Publication Integrity, and Global Academic Contexts
KALECİ, F. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-2600-1232-1.ch004 · Cited by 1
Yagouni-Bouzar, H. (2026) · Papers in Applied Geography · doi:10.1080/23754931.2026.2718099
Maturo, F., Micozzi, A., & Porreca, A. (2026) · SN Social Sciences · doi:10.1007/s43545-026-01683-3 · Cited by 2
“Large language models can generate fluent and apparently meaningful text without seman - tic awareness, intentionality, or accountable agency (Bender et al. 2021; Floridi 2023 ).”
Machado, D. (2026) · Scientometrics · doi:10.1007/s11192-026-05792-x · Cited by 4
“Rare, unusual, or unprecedented combinations of ideas are precisely those for which the model has the weakest signal and toward which it has the least pull (Bender et al., 2021; Birhane et al., 2022).”
Gamage, D., Nanayakkara, I., & Sasahara, K. (2026) · Computers in Human Behavior Reports · doi:10.2139/ssrn.7360004
Elega, A. A., Mohammed, A., & Babatunde, A. (2026) · Media and Communication · doi:10.17645/mac.12417 · Cited by 1
“(R1) Research on generative AI and the ethics of automation has warned that LLMs can confidently fabricate facts, like stochastic parrots, producing fluent but ungrounded assertions that can be socially toxic when passed off as authority (Bender et al., 2021).”
Kılıçkaya, F. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-2600-1232-1.ch011
Liu, S.-W., Chu, S., Wang, H.-C., & Lee, M.-S. (2026) · JMIR Formative Research · doi:10.2196/98343
Automated and Personalized Employee Performance Evaluations Using Natural Language Generation
Belarache, O., Ounacer, S., & Azzouazi, M. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-21200-9_30
Gallardo, M. R., & Carbonell, M. M. (2026) · Arte y Políticas de Identidad · doi:10.6018/reapi.704601
Zhorzholiani, T. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-2600-1232-1.ch009
Containing AI: from responsible to critical AI governance
Schütze, P., & Brevini, B. (2026) · AI & Society · doi:10.1007/s00146-026-03337-7 · Cited by 1
Rajkumar, S., Kesavan, V., & Kaleem, K. H. A. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-2600-0970-3.ch012
Standards, Regulation, and Governance for Agentic AI
Piroumian, V. (2026) · Studies in computational intelligence · doi:10.1007/978-3-032-18585-3_13
Yang, J. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.7357559
Socio-Culturally Responsive VR Training for Enhancing English Teachers’ Pragmatic Competence
Mahmoudi-Dehaki, M., & Nasr-Esfahani, N. (2026) · ECNU Review of Education · doi:10.1177/20965311261472506 · Cited by 1
“Beyond the widely discussed Western-centric bias (Bender et al., 2021), additional concerns, such as socio-economic, gendered, and linguistic bias, were considered.”
Libraries as necessary sites for refusing faux fables of technofascism and imagining otherwise
Appedu, S. (2026) · AI and Ethics · doi:10.1007/s43681-026-01314-w
Algorithmic Semiosis: Emojis, Memes, and Meaning-Making in Platformed Digital Communication
Prasantham, P., & Zhou, B. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-2600-1242-0.ch010
BAYRAKLI, V. K., Demircioğlu, A., & Ertek, Z. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-2600-1232-1.ch002
Said, N. A. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-2600-1232-1.ch003
Global Trends in Research Ethics in the Context of Artificial Intelligence
Resulbegoviq, H. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-2600-1232-1.ch012
Acharya, S., & Anutariya, C. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-36547-7_18
Teaching What the Machine Told Us: Critical Agency and Epistemic Trust in the Algorithmic Classroom
Genç, L. G., & Uygun, E. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-2600-1232-1.ch014
Is Living With Intelligent Machines Bringing an End to Teaching?
Çetinkaya, M. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-2600-1232-1.ch010
Sustainable Agentic Intelligence: A Taxonomy, Evidence-Based Analysis, and Research AgendaPreprint
Poongodi, K., Fameen, R., & Meenakshi, K. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.7358627
Duration and Externalized Intelligence: A Bergsonian Critique of Large Language Models
Hamamra, B. (2026) · Human Arenas · doi:10.1007/s42087-026-00692-1
Zhu, H., Liu, W., & Shi, Q. (2026) · Frontiers in Education · doi:10.3389/feduc.2026.1874510
Autodesign? Large Language Model Evaluations for UX Assistants: Review and Guidelines
Santos, W. K. B., Souza, F. D. C., da Silva, B. C., Gimenez, P. P., Mendes, A. K. V., Paixão, G. D. S. L., Correia, W. F. M., & de Siqueira, A. G. (2026) · DAT Journal · doi:10.29147/datjournal.v11i2.1034
Towards equitable hydrological data in the age of Earth observations and artificial intelligence
Baldassarre, G. D., Brandimarte, L., de Brito, M. M., Mukherjee, J., & Rusca, M. (2026) · Nature Water · doi:10.1038/s44221-026-00706-w · Cited by 1
“However, because LLMs are trained mainly on English-language, Global-North-centric, and often uncurated datasets, they can inherit stereotypes and misrepresent already-vulnerable and marginalized communities 7 .”
Disruptive Horizons: Emerging AI Technologies Poised to Reshape the Global Media Industry
Vukanović, Z. (2026) · Springer eBooks · doi:10.1007/978-3-032-28082-4_29
The Principia of Generative-AI Pedagogy: Three Theories of Generative Motion
Ebron, J. P. G., Gamez, J. N., Amores, C. C., et al. (2026) · International Conference on Education and Training Technologies (ICETT) · doi:10.1109/icett70124.2026.11665848
Can Machines Write? Recursive Re-Embodiment and Distributed Agency in Human–AI WritingPreprint
Abdelal, G. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.7366889
Mo, C., Manchaiah, V. K. C., Wasmann, J.-W. A., & Wang, S. (2026) · International Journal of Audiology · doi:10.1080/14992027.2026.2720841 · Cited by 1
Train and deploy: Expropriation and accumulation in the generative AI economy
Ravishankar, V., & Kienscherf, M. (2026) · Capital & Class · doi:10.1177/03098168261478525 · Cited by 1
Zhou, C.-H. (2026) · International Journal of Heritage Studies · doi:10.1080/13527258.2026.2723869
An Analysis of Artificial Intelligence Governance Policy Models
Gao, H., Xie, Y., Guo, S., et al. (2026) · International Conference on Computer Communication and the Internet (ICCCI) · doi:10.1109/iccci70321.2026.11666523
“However, while reshaping the economy, this technological evolution is systematically accompanied by various security risks, including algorithmic opacity, data privacy concerns, deepfake disinformation, and the potential risks of autonomous systems losing control [1].”
Xu, J., & Zhang, Z. (2026) · International Conference on Distance Education and Learning (ICDEL) · doi:10.1109/icdel70250.2026.11663645
“These models have demonstrated formidable capabilities not only in natural language understanding and generation but also in complex reasoning, knowledge synthesis, and contextual adaptation [1].”
Alves, J. N., & Tseng, H. (2026) · AI and Ethics · doi:10.1007/s43681-026-01343-5
“Unlike traditional automation, which executes pre-defined human rules, AI systems exercise a form of discretion through inference processes whose outputs are not fully predictable from their specifications and whose internal logic remains opaque to designers and operators alike [ 10, 35].”
Silva, A., & Lino, M. (2026) · Nursing Inquiry · doi:10.1111/nin.70162
“The global nursing workforce is under severe and sustained pressure, and artificial intelligence (AI) is increasingly drawn upon to inform the policy, education, and system‐level responses intended to address it (Bender et al. 2021).”
Chakraborty, R. (2026) · British Journal of Contemporary Research · doi:10.67693/bjcr-y674x85r
张, 绍. (2026) · Modern Linguistics · doi:10.12677/ml.2026.148887
Haroz, E. E., Bradshaw, C. P., & Feaster, D. J. (2026) · Prevention Science · doi:10.1007/s11121-026-01971-7
“Their paper engages questions of validity, reliability, prompt transparency, and the limits of LLMs as analytic instruments, drawing on the broader literature on LLMs in social and behavioral science (Bender et al., 2021; Demszky et al., 2023).”
Bridging AI and sociology: a comparative analysis of AI integration in China and the US
Zeng, F., & Fan, X. (2026) · Theory and Society · doi:10.1007/s11186-026-09743-6
“Meanwhile, while LLMs offer unprecedented capabilities for sociological analysis, several limitations including training data biases (Bender et al., 2021), the “black-box” nature of proprietary models which undermines methodological transparency (Ollion et al., 2024), and privacy and potential…”
Campos, J., Chimenti, P. C. P. D. S., & da Fonseca, A. L. A. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.7382911
TEI, Taxonomies, and Semantic Data Governance for Documented LLM-Assisted Literary Analytics
Giménez, D. (2026) · Analytics · doi:10.3390/analytics5030033
“The model does not usually expose a stable chain of textual evidence, and its internal reasoning remains opaque even when the prompt and output are visible [1–3].”
Generative Language Models (GLMs)
Parson, V. (2026) · Springer eBooks · doi:10.1007/978-3-031-52643-5_268-1
Intentionality All-Stars Redux: Do Language Models Know What They Are Talking About?
Browning, J. (2026) · Oxford University Press eBooks · doi:10.1093/9780191998317.003.0002
Learning Energy for Sustainable Learning with Artificial Intelligence in Higher EducationPreprint
Nguyen, V. H. (2026) · Research Square · doi:10.21203/rs.3.rs-10708878/v1
Critical Reflections on Technological Singularity: Revisiting Ray Kurzweil’s Predictions
Boas, M. A. S. V., & Bliacheriene, A. C. (2026) · Revista Brasileira de Direito · doi:10.18256/2238-0604.2025.v21i3.5275
Picó, J. F. I., Mas, C. A., Lozano, À. M., et al. (2026) · Systems · doi:10.3390/systems14091060
A Talking Cure for Autonomy Traps: How to Share Our World with Chatbots
Rini, R. A. (2026) · Oxford University Press eBooks · doi:10.1093/9780191998317.003.0017
Chaudhary, B., & Ojha, A. (2026) · Philosophy of Translation · doi:10.1080/29984750.2026.2726216
A review of bias detection and fairness auditing techniques in LLMs
Kaveti, N. K., & Pattanshetti, T. R. (2026) · Discover Artificial Intelligence · doi:10.1007/s44163-026-02121-2 · Cited by 1
Shakaraliyeva, Z. A. (2026) · Proceedings of the ... International Conference on Business Excellence · doi:10.2478/picbe-2026-0465
Sebti, R. (2026) · Journal of Information Communication and Ethics in Society · doi:10.1108/jices-12-2025-0413
Garcia, M. P. (2025) · Simpósio Brasileiro de Computação Musical (SBCM) · doi:10.5753/sbcm.2025.13065
“A much-debated article on this question was published by Bender et al. [15].”
A contamination model for regression bias and residual-variance compression in LLM-generated data
Mankad, S. (2026) · Statistics & Probability Letters · doi:10.2139/ssrn.7398614
Carbon Chat: A Virtual Reality Game for Creating Awareness of the Environment Impact of LLMs
Hurst, W., Sallou, J., & Krampe, C. (2026) · International Conference on Virtual Reality · doi:10.1109/icvr70658.2026.11672397
Do Large Language Models Overvalue AI Narratives?Preprint
Qin, Q., & Chen, Y. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.7397642
Semantic-Fidelity Risk in AI-Mediated Revision of Multilingual Academic Abstracts
Almashour, M., Aldamen, H. A. K., Al‐Deaibes, M., & Alsharefeen, R. (2026) · IEEE Access · doi:10.1109/access.2026.3730340
“Writers and institutions remain responsible for verifying the source-output relation, while tool providers control opaque model deployment, updates, defaults, and nondeterministic generation conditions [29], [30], [31], [39].”
Kumar, G., & Sharma, S. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-9200-4.ch001
Digital linguistic sovereignty and Kazakh language policy in the digital environment
Zharkynbekova, S. K., Ayupova, G., Galiyeva, B., Shakhputova, Z. K., & Zabrodskaja, A. (2026) · Frontiers in Psychology · doi:10.3389/fpsyg.2026.1858650
“Research indicates that modern language technologies are not neutral, as they reproduce structural inequalities by reinforcing the dominance of English-language content and displacing low-resource languages (Bender et al., 2021).”
de Portugal, I. N.-R. C., Rumayor, M., & Génova, G. (2026) · Philosophies · doi:10.3390/philosophies11050156
“Bender, Gebru, McMillan-Major and Shmitchell [3] were right that the coherence perceived in language model output is “in the eye of the beholder” [3] (pp. 616–617).”
InfluentialDecision substitution and the reorganisation of judgement in AI-mediated society
Ibitoye, A., Ajenifuja, O. F., Kareem, R. S. A., Adeyemo, D. D., & Famutimi, R. F. (2026) · AI & Society · doi:10.1007/s00146-026-03349-3
Sawad, A. B., & Binsawad, M. H. (2026) · Applied Sciences · doi:10.3390/app16178779
Silva, K. L. C. D., & Oliveira, V. M. D. (2026) · Revista de Geopolítica · doi:10.56238/revgeov17n9-026
Mühlhoff, R. (2026) · AI & Society · doi:10.1007/s00146-026-03281-6 · Cited by 4
Harry, B. (2026) · IGI Global eBooks · doi:10.4018/979-8-3373-8835-9.ch001
Zarabi, H. (2026) · Current Issues in Language Planning · doi:10.1080/14664208.2026.2728062 · Cited by 1
Beyond the autonomy illusion: hybrid intelligent systems and the architecture of AI accountability
Medsker, L., & Virmani, S. (2026) · AI and Ethics · doi:10.1007/s43681-026-01327-5
“The documented bias remains, and a future output in a different context may reproduce it in a form not yet detected [4, 15].”
Digital linguistic imperialism: can we decolonize English when the AI writing tools are colonial?
Hosseinpur, R. M. (2026) · Applied Linguistics · doi:10.1093/applin/amag073 · Cited by 1
Almeida, J. G. R. T., Pérez, E. F. P., Carapia, A. A. Q., Sánchez, J. G. A., & Soto‐Pinto, L. (2026) · Cureus Journal of AI-Augmented Research · doi:10.7759/s44568-026-00263-6
Perry, A., & Cameron, C. D. (2026) · Cambridge University Press eBooks · doi:10.1017/9781009639057.002
Models Match or Surpass Objective Human Performance on Various Tasks
Goldstein, A., & Stanovsky, G. (2026) · Cambridge University Press eBooks · doi:10.1017/9781009639057.006
Courchesne, S., Powell, W. A., & Zeidler, D. L. (2026) · Disciplinary and Interdisciplinary Science Education Research · doi:10.1186/s43031-026-00170-2 · Cited by 1
“Large language models (LLM) have demonstrated a capacity to produce coherent and relevant text (Bender et al., 2021) and have performed well on a variety of exams and benchmarks (Griot et al., 2025; Katz et al., 2024; Sabri et al., 2024).”
TAN, K. H. (2026) · International Journal of Law Management & Humanities · doi:10.63108/ijlmh.12783
AI Psychosis is a Measured Risk of Capitalist Technology
Ayres, S. C. (2026) · ACM SIGCAS Computers and Society · doi:10.1145/3842677.3842679
A Reader’s Workbench: Reading With Friends
Bernstein, M. (2026) · ACM Conference on Hypertext and Social Media (HT) · doi:10.1145/3800935.3830862
“It is difficult to reconcile this dialogue with the supposition that Claude is a mere stochastic parrot [10].”
Kumar, A. (2026) · Journal of Knowledge Management · doi:10.1108/jkm-03-2026-0627 · Cited by 1
Hedberg, A. (2026) · AI & Society · doi:10.1007/s00146-026-03343-9
Cognitive Enhancement or Social Atrophy? Generative AI, Neurodiversity, and Human Flourishing
Casanovas, Á. P., Castellví, P., & Lidón‐Moyano, C. (2026) · Human Affairs · doi:10.1515/humaff-2025-0108
“” (Bender et al. 2021, 616) Such biases had already appeared in other systems trained on narrow datasets; paradigmatically, algorithmic bias became a central topic in philosophy of science and technology in 2018, with Amazon ’ s experimental AI recruiting tool, which systematically discriminated…”
허대찬, & 허대찬 (2026) · Archives of Design Research · doi:10.15187/adr.2026.08.39.3.365
Garvey, M. D. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.7428998
Zhadko, N. V. (2026) · Problems of Modern Education (Problemy Sovremennogo Obrazovaniya) · doi:10.31862/2218-8711-2026-2-167-179
Sakurai, T., Aizawa, D., Okawa, K., Kofuchi, R., Hirota, M., Gotouda, H., Yamamoto, T., & Ohkubo, C. (2026) · Journal of Medical Education and Curricular Development · doi:10.1177/23821205261483568
“27,40,41 A balanced assessment system should evaluate both independent competence and the ability to use AI critically, ethically, and responsibly.”
The Treatment of Absence and Null Results in Materials Machine Learning Literature: A Review Study
Huy, N. T., Minh, P. Q., & Bich, L. T. (2022) · Journal of Artificial Intelligence for Materials Science · doi:10.68159/c772411799
Teachers as Cyber-Social Designers of Literacy Across the Curriculum in the Age of GenAI
Smala, S., & Spinelli, F. (2026) · Emerald Group Publishing Limited eBooks · doi:10.1108/978-1-80686-243-620261002
Traduire sans corps ? Fragmentation, fidélité et illusions sémiotiques à l’ère de l’IA générative
Duteil, C. (2026) · Signata · doi:10.4000/16qm8
Multimodal Learning: Integrating Text, Images, Video, and Audio for Cross-Modal Reasoning
Nayak, L. S. A., Aparna, P., Rashmi, K., Phonsa, G., Tiwari, S. P., & Raju, M. V. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-27691-9_27
William, E. (2026) · Journalism and Media · doi:10.3390/journalmedia7030182
L’IA générative ne serait-elle qu’une technologie peu explicative ?
Desclés, J.-P. (2026) · Signata · doi:10.4000/16qme
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.