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 5,901–6,000 of 6,931
Warren, G., Sun, J., Shklovski, I., & Augenstein, I. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3812358 · Cited by 5
“At the same time, the linguistic fluency and persuasive qualities of LLM-generated text [52] can convince people to erroneously accept their outputs [6].”
Closed Rooms and Oracles: the poetics of vector space in Sam Riviere’s Conflicted Copy
Farrier, D. S. (2026) · Journal of British and Irish innovative poetry · doi:10.16995/bip.25843
Red teaming with faith leaders: expanding digital safety and accountability to frontiers of care
Lutz, N., & Weyl, E. G. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3812362 · Cited by 1
“We framed LLMs as non-neutral technologies that incorporate biases from their human makers [56, 60] and as "stochastic parrots" associating probablistic text based on (sometimes biased) data labeling [8].”
Uniformity and Diversity at the Same Time?
Mair, C. (2026) · Cambridge University Press eBooks · doi:10.1017/9781009569699.012
Samal, J. (2025) · International Journal of Health & Allied Sciences · doi:10.4103/ijhas.ijhas_64_25
Patra, S. C., Maheswari, B. U., & Naik, G. R. (2026) · Journal of Computational Design and Engineering · doi:10.1093/jcde/qwag059 · Cited by 2
Dewi, A. N., Sørensen, B., & Thellefsen, M. (2026) · Journal of Documentation · doi:10.1108/jd-04-2026-0185
trAIt: Species-by-Trait Data Retrieval using Large Language ModelsPreprint
Balaji, S., Martinson, K. A., Schellenberger, J. S., Koley, J., Inman, C. M., Hofmann, H. A., Young, R. L., & Harpak, A. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.06.19.732660 · Cited by 1
Hlaimi, S., & Maloney, J. (2026) · Citizenship Social and Economics Education · doi:10.1177/14788047261461204 · Cited by 1
Davey, G. (2026) · Advances in Artificial Intelligence and Robotics Research · doi:10.4236/airr.2026.22004
Usage Variation amid Social Diversity and Political Change
Smith, A., & Collins, P. C. (2026) · Cambridge University Press eBooks · doi:10.1017/9781009569699.007
Between Promise and Uncertainty
Moreira, F. T. T., Martins, S. L., & Figueiredo, A. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-9883-9.ch011
Navarro, M. G. J. (2026) · El FARO Revista Digital de Docencia Universitaria · doi:10.63790/7stn5510
Selyshcheva, I. A. (2026) · CTE Workshop Proceedings · doi:10.55056/cte.1438
Muslim, A., Asri, N. A. M., & Lyons, D. (2026) · Journal of Medical Internet Research · doi:10.2196/105453
Testing Framework Migration with Large Language Models
Alves, A., Montandon, J. E., & Hora, A. (2026) · doi:10.1145/3793654.3793749 · Cited by 4
Kiang, Y. W., Li, Z., Lee, A. T. L., Tam, V. W. L., Lam, K. H., & Lee, V. C. S. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-29791-4_26
Wang, F., Guo, L. P., & Zhang, Y. R. (2026) · Education and Information Technologies · doi:10.1007/s10639-026-14060-7
“At the data level, training corpora and existing benchmarks are often dominated by well-structured, adult-action-cen-tered narratives, leaving children’s perspectives and developmental characteristics underrepresented (Bender et al., 2021).”
Zheng, C., & Chen, G. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-29791-4_8
Leveraging AI Flaws for Human Learning
Sinha, T. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-29791-4_32
MISHRA, A. R. (2026) · IGI Global eBooks · doi:10.4018/979-8-3373-8424-5.ch012
Harnessing ChatGPT for educational scientific research: A qualitative analysis approach
Moussa, M. A., Saleh, M. M., & Al-Adamat, O. A. (2026) · Journal of Digital Educational Technology · doi:10.29333/jdet/18855 · Cited by 1
“Furthermore, while scholarly commentary has raised concerns about plagiarism, authorship, and academic integrity (Bender et al., 2021; Cotton et al., 2023; Perkins, 2023; Van Dis et al., 2023), little is known about how researchers themselves navigate these ethical tensions in practice, how they…”
InfluentialKansa, E. C. (2026) · Internet Archaeology · doi:10.11141/ia.71.15 · Cited by 1
AI-powered chatbots in education: Supporting students and enhancing learning experiences
Kasimova, Z. S., Mirzakarimova, Z., Qosimova, S., et al. (2026) · AIP conference proceedings · doi:10.1063/5.0330154
The case for humanities-based studies of AI production and effects
Asscher, O. (2026) · AI & Society · doi:10.1007/s00146-026-03170-y
Lutostański, B. (2026) · Forum Poetyki · doi:10.14746/fp.2026.43.3 · Cited by 1
Algorithmic accommodation: linguistic alignment in human-AI relational engagement
Li, H., & Zhang, R. (2026) · Communication and Change · doi:10.1007/s44382-026-00032-5 · Cited by 7
“…as word embeddings, enable LLMs to produce semantically coherent responses, not because they share meaning with the user, but because they are trained to predict linguistically probable output that are contextually appropriate and aligned with patterns of human conversations (Bender et al., 2021).”
Generative AI, Architecture, and Mathematics: Supporting Sustainable Development Goals
Henriques, D. P., & Ostwald, M. J. (2026) · Springer eBooks · doi:10.1007/978-3-032-02376-6_20-1
Thomas, C. D., Busch, A. H., & Carlo, S. M. (2025) · Journal of Education Finance and Law · doi:10.5406/30678560.50.3.08
“Instead, they rely on advanced pattern recognition to generate responses based on statistical correlations found in the data they have been trained on (Bender et al., 2021).”
InfluentialSpeculative Design in the Age of AI
KAYABAŞI, Y. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-2600-2711-0.ch007
Hassen, M. Z. (2026) · Science Journal of Education · doi:10.11648/j.sjedu.20261403.14
Formation ethics: a cross-cultural framework for AI governance
Kim, S. (2026) · AI & Society · doi:10.1007/s00146-026-03186-4
“…formation failures recur at four sites: interpretive fragility, when fluent outputs are treated as authoritative despite their lack of grounding (Bender et al. 2021); organizational haste, when speed becomes a proxy for competence; epistemic narrowing, when what is measurable crowds out what is…”
Community Futures: Hybrid and Situated Critical AI Literacy
Pait, S., Janicki, S., Hou, Y., Nitsche, M. A., & Howell, N. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-29773-0_18 · Cited by 1
Chen, W. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6903078
Sociodemographic Biases in Educational Counselling by Large Language Models
Adamczyk, T., Mieleszczenko-Kowszewicz, W., Bajcar, B., Chodak, G., Szczęsny, A., Markiewicz, M., Ostrowska, K., Sawczuk, A., & Kazienko, P. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-29773-0_40
Göktepe, O. (2026) · Lectio Socialis · doi:10.47478/lectio.1849650
Kurshan, E. (2026) · ACM International Conference on Computing Frontiers (CF) · doi:10.1145/3801487.3802805
A Persona-Augmented Multi-Agent System for Varied Narrative Generation
Sillano, A., Arbore, G., & Russis, L. D. (2026) · ACM SIGCHI Symposium on Engineering Interactive Computing Systems (EICS) · doi:10.1145/3807968.3810931
Yıldız, S. (2026) · Anadolu Üniversitesi Sosyal Bilimler Dergisi · doi:10.18037/ausbd.1807814
Beyond Black-Box Agents: Explainable and Validatable Generative ABMs
Tang, X., & Törnberg, P. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-29915-4_39
“First, LLMs are stochastic systems whose outputs are sensitive to prompts, sampling parameters, and contextual framing, complicating replication and systematic comparison across simulations [8][9].”
Ratnani, S. J., Li, L., Lou, Y., Li, M., & Hua, K. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3806487
Sakib, A. N., Chisty, M. M. O., & Zhang, Z. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3812342
Brainrot: Deskilling and Addiction are Overlooked AI Risks
Chalkidis, I., & Søgaard, A. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3812306 · Cited by 6
Generative AI in Engineering Education: Equity-Enhancing Tool or Digital Divide Deepener?
Biswas, R. (2026) · IEEE Global Engineering Education Conference (EDUCON) · doi:10.1109/educon67543.2026.11574391 · Cited by 1
Hundertmark, S. (2026) · Informatik-Spektrum · doi:10.1007/s00287-026-01608-7
Haq, I., & Saldías-Fuentes, B. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3812419 · Cited by 9
Willem, T., & Piraud, M. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3812295
“…valuing AI emissions-feedback tools as awareness-raising and as an incentive and learning prompt for more energy efficient coding substantiate calls for greater transparency in reporting the energy use and carbon footprint of machine learning research [3, 10, 13, 22, 23, 39, 41, 46, 57, 59].”
Safety Theater in Anthropomorphic AI: Why We Need to Shift from Safety to Accountability Narratives
Maeda, T., & Buening, R. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3812282 · Cited by 1
3. An Introduction to the Philosophy of Artificial Intelligence
Torbjørnsen, R. R., & Kyselo, M. (2026) · Universitetsforlaget eBooks · doi:10.18261/9788294167234-26-03 · Cited by 2
Enhancing situation awareness for disaster policymaking via LLM-generated narratives
Shao, R., & Fan, C. (2026) · Technology in Society · doi:10.1016/j.techsoc.2026.103441 · Cited by 1
Havens, L., Terras, M., Bach, B., & Alex, B. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3812283
“Critiques of predominant approaches to AI research and practice have noted an emphasis on quantity in relation to dataset size [122], model architecture [15], and performance evaluations [155].”
Does It Lead to Prejudice or Discrimination? An Assessment of Stereotypes in LLM Decision-Making
Aoshima, T., & Akiyama, M. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3812286 · Cited by 1
“Drawing on the taxonomy of representational harm [28], we also consider expressions that imply the person being evaluated is non-typical [22] as concrete examples of problematic uses of generics (Section 2.2).”
InfluentialResponsible NLP: Why performance is no longer enough
Lamsiyah, S. (2026) · Natural language processing. · doi:10.1017/nlp.2026.10030 · Cited by 1
“Bias in language technologies has been documented in word embeddings (Bolukbasi et al. 2016), language models (Bender et al. 2021), and stereotype-focused evaluations (Mostafazadeh Davani et al. 2025).”
Density-Guided Response Optimization: Community-Grounded Alignment via Implicit Acceptance Signals
Gerard, P., & Volkova, S. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3806449 · Cited by 1
“Ethical NLP work further argues for embedding social values and community perspectives into model design [7, 28, 35], yet little work has explored how such norms can be learned operationally from naturally occurring community behavior.”
Helpful, Harmless, Honest? RLHF as Survey Design and Content Moderation
D'Alonzo, S., Kreuter, F., & Booth, S. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3806444 · Cited by 4
“Other steps of the LLM pipeline, such as dataset curation [10], have similar design decisions embedded within them.”
Hoel, A., & Wilde, L. R. A. (2026) · Universitetsforlaget eBooks · doi:10.18261/9788294167234-26-07 · Cited by 2
Kivity, S., Barilan, Y. M., Noham, R., & Saban, M. (2026) · Journal of Medical Internet Research · doi:10.2196/88677 · Cited by 4
The Impact of Large Language Models on Content Quality in Social Media
Shahbazi, Z., & Johnsson, M. (2026) · Electronics · doi:10.3390/electronics15132820
Generative AI-Supported Assessment of Engineering Capstone Proposals: A South African Case Study
Klopper, B., Blackie, H., Dyk, L. V., & Marais, H. (2026) · IEEE Global Engineering Education Conference (EDUCON) · doi:10.1109/educon67543.2026.11574460
“However, some scholars caution that despite these advancements, concerns remain regarding the validity and consistency of assessments generated with the assistance of LLMs [10]– [12].”
Wilde, L. R. A. (2026) · Universitetsforlaget eBooks · doi:10.18261/9788294167234-26-08
Can AI Chatbots Provide Coaching in Engineering? Beyond Information Processing Toward Mastery
Qadir, J., Attique, M. A., Shoaib, S., & Ghaznavi, I. (2026) · IEEE Global Engineering Education Conference (EDUCON) · doi:10.1109/educon67543.2026.11574214 · Cited by 2
“Those who understand how large language models (LLMs) work (as statistical pattern matchers trained on text rather than as reasoning agents with genuine understanding) may be particularly cautious about relying on AI for moral guidance or complex judgment calls [34].”
Dilin Doğuştanlığı ve Uyaran Eksikliği Argümanı
Kerimoğlu, C. (2026) · BABUR Research · doi:10.66119/babur.1939913 · Cited by 1
4. Foundational Conceptual Issues in the Ethics of Artificial Intelligence
Torbjørnsen, R. R., & Kyselo, M. (2026) · Universitetsforlaget eBooks · doi:10.18261/9788294167234-26-04 · Cited by 2
Gao, Z., Ajibola, Y., & Thebault-Spieker, J. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3812214
Auditing the Space Between Languages: Character Representation in Bilingual Young Adult Literature
Nagar, T., Maheshwary, P., & Karumbaiah, S. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3806471
Ladd, J. (2026) · English Language Notes · doi:10.1215/00138282-12301864
Shahin, A., Masry, B., Hamzeh, R., Aljabali, A., & Hussain, N. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.7023071
Gutiérrez, R., Pretolesi, D., Regal, G., & Tscheligi, M. (2026) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3816763
“Third, remote LLMs like OpenAI’s GPT and Google’s Gemini have significant ethical issues from data misuse and civilian surveillance and environmental impacts due to their high energy consumption and the extensive resources required for AI training and operation [8].”
Terms of (Ab)Use: An Analysis of GenAI Services
Pandit, H. J., Blankvoort, D. A. H., Shaaban, A., Luccioni, A. S., & Birhane, A. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3806809 · Cited by 6
Li, S., & Hartono, P. (2026) · Proceedings of the ACM on Computer Graphics and Interactive Techniques · doi:10.1145/3816086
“Although large language models can generate fluent text with apparent personality or stance, their outputs are not grounded in understanding, intention, or contextual awareness, but in the prediction and recombination of statistical language patterns [Bender et al. 2021].”
Yapay Zekâ Destekli Anlatı Sistemlerinde Çocukluk, Etik ve Bilgi Üretimi
Karademir, N. (2026) · Artuklu Humanities · doi:10.46628/ahu.1777111
Nagaraj, P., Muneeswaran, V., Amer, A., Hafez, M., Muthiah, R., Islam, M. T., & Ijaz, M. F. (2025) · Asian Conference on Communication and Networks (ASIANComNet) · doi:10.1109/asiancomnet68615.2025.11579569
The biased co-counsel: analyzing gendered defense strategies in AI-assisted criminal litigation
Voutyrakou, D. A., & Voutyrakos, N. (2026) · AI and Ethics · doi:10.1007/s43681-026-01221-0
Pathways, Limitations, and Future Directions for Embedding Values in Large Models
Zhou, Y., & Teng, Y. (2026) · Philosophy of engineering and technology · doi:10.1007/978-981-95-8023-1_21 · Cited by 1
Lin, H., & Xie, L. (2026) · Frontiers in Psychology · doi:10.3389/fpsyg.2026.1839565 · Cited by 1
“Bender et al. (2021) noted that outputs from large language models are generated from patterns learned from existing data, which may constrain the production of highly novel ideas.”
Narrativa transmedia en la era de la IA
Robledo-Dioses, K. (2026) · SOCIAL Review International Social Sciences Review / Revista Internacional de Ciencias Sociales · doi:10.62701/revsocial.v14.5533
Gebremeskel, H. G., Feng, C., & Abera, A. M. (2026) · Communications in computer and information science · doi:10.1007/978-981-92-0199-0_1
Графин, А. Д., Золин, А. А., & Shekhovtsov, O. I. (2026) · Современные наукоемкие технологии (Modern High Technologies) · doi:10.17513/snt.40839
Wang, G., & Cao, Y. (2026) · Philosophy of engineering and technology · doi:10.1007/978-981-95-8023-1_5 · Cited by 1
Lu, J. (2026) · SHS Web of Conferences · doi:10.1051/shsconf/202623505012 · Cited by 1
The intention turn: AI and the first democratization of understanding
Gazit, L. (2026) · AI and Ethics · doi:10.1007/s43681-026-01208-x
Five Educational-Psychology Lenses for Training and Evaluating AI Models
Wang, P., & PAAS, F. G. W. C. (2026) · Educational Psychology Review · doi:10.1007/s10648-026-10195-8
“An LLM trained on open internet text, for instance, learns myriad patterns that reflect societal biases, stereotypes, and power structures present in the data (Bender et al., 2021; Bolukbasi et al., 2016; Gebru et al., 2021).”
da Silva, R. L. G., & da Cruz, E. J. (2026) · Revista da Emeron · doi:10.62009/emeron.2764.9679.v2.2026.558.p107-122
AI-Powered Service Innovation and Hyper Personalization in Modern Product Ecosystems
Bhardwaj, A. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-5067-7.ch004
ChatGPT-5.4 in health education: inter-generation stability and persistent readability challenges
Han, X., Li, D., Xing, R., & Song, X. (2026) · Frontiers in Public Health · doi:10.3389/fpubh.2026.1864894
Dong, M., Liu, X., Luo, Y., Zeng, Y., Fei, F., & Du, L. C. (2026) · Research Integrity and Peer Review · doi:10.1186/s41073-026-00231-0
Güder, Ş. C. (2026) · The Turkish Online Journal of Design Art and Communication · doi:10.7456/tojdac.1929108 · Cited by 2
AI Model for Automated Content Moderation: Integrating ML and Image Processing
Madhusudhan, S., Gooliyavar, K., Budihalamath, K., Lavanya, T., & Nishitha, M. R. (2026) · International Conference on Inventive Computing and Informatics (ICICI) · doi:10.1109/icici68773.2026.11580906
“[30].”
LLMs, Epistemic Responsibility, and the Prospects for AGI
Wolf, M. P. (2026) · Minds and Machines · doi:10.1007/s11023-026-09790-z
The Compassion Algorithm: Large Language Models and the Future of Medicine
Periyakoil, V. S. (2026) · Journal of Palliative Medicine · doi:10.1177/10966218261465070
“…should include structured exposure to the conceptual foundations of LLMs: how LLM transformer architectures generate outputs, 5 how training data shapes and biases model behavior, 13 and why the appearance of empathic or insightful outputs does not re fl ect genuine understanding or compassion.”
The Word without a Speaker? Scripture, Its Meaning, and the Challenge of Large Language Models
Dosoudil, J. (2026) · Caritas et Veritas · doi:10.32725/cetv.2026.005
Luo, J., Li, S., Wang, M., Cao, X., & Li, H. (2026) · IEEE Transactions on Learning Technologies · doi:10.1109/tlt.2026.3708984
“GenAI simply cannot replace the material and affective dimensions of human experience [29], as it is trained on string prediction tasks and its training data contains embedded biases that have already been recognized as harmful [30].”
Grounding large language models in an anthropological knowledge graph: A neuro-symbolic approach
Artz, M. (2026) · Human Organization · doi:10.1080/00187259.2026.2674887
Salvador, I., Furno, A., & Derrible, S. (2026) · Transportation Research Record Journal of the Transportation Research Board · doi:10.1177/03611981261456273 · Cited by 1
Predicting Adolescent Mental Health from Social, Academic & Health Data
Nakirekanti, M., Reddy, K. G. R., Balusa, B. C., et al. (2026) · doi:10.1109/ci2a69097.2026.11576773
“[7].”
Saikia, D., Basumatary, P., & Saikia, A. (2026) · AI & Society · doi:10.1007/s00146-026-03181-9
Razak, M. R. A., Nasrul, M. A. D., & Abbas, I. (2026) · Millah Journal of Religious Studies · doi:10.20885/millah.vol25.iss1.art2 · Cited by 3
“Conclusions drawn today about generative AI risks or capabilities may need updating within a short time frame (Bender et al., 2021).”
Alves, A. C. V., Martignago, G. M., Correa, M. R. D. S., Neto, M. P. P., Pillon, M. A., Miers, C. C., Correia, P. H. B., Hayashi, V. T., & Junior, M. A. S. (2026) · Lecture notes on data engineering and communications technologies · doi:10.1007/978-3-032-23311-0_7
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.