AI safety on whose terms?
Open accessAbstract
Rapid, widespread adoption of the latest large language models has sparked both excitement and concern about advanced artificial intelligence (AI). In response, many are looking to the field of AI safety for answers. Major AI companies are purportedly investing heavily in this young research program, even as they cut “trust and safety” teams addressing harms from current systems. Governments are taking notice too. The United Kingdom just invested £100 million in a new “Foundation Model Taskforce” and plans an AI safety summit this year. And yet, as research priorities are being set, it is already clear that the prevailing technical agenda for AI safety is inadequate to address critical questions. Only a sociotechnical approach can truly limit current and potential dangers of advanced AI.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Most AI tools are trained using data created by humans, and frequently exhibit a broad range of prejudice and cognitive errors accordingly.
“Most AI tools are trained using data created by humans, and frequently exhibit a broad range of prejudice and cognitive errors accordingly (1, 57–60).”
From Can Generative AI improve social science? (2024) · cites this paper as [58]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cited by
Sorted by citations
Artificial intelligence and illusions of understanding in scientific research
Messeri, L., & Crockett, M. J. (2024) · Nature · doi:10.1038/s41586-024-07146-0 · Cited by 707 · see citing works
“Rather, they require sociotechnical approaches that account for the inseparability of social and technical dynamics 21,174,187 .”
Can Generative AI improve social science?
Bail, C. A. (2024) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2314021121 · Cited by 369 · see citing works
Mapping the ethics of generative AI: a comprehensive scoping review
Hagendorff, T. (2024) · Minds and Machines · doi:10.1007/s11023-024-09694-w · Cited by 141 · see citing works
“Sources also analyze challenges of the AI research community to ensure workforce diversity (Lazar & Nelson, 2023).”
Generative AI and the politics of visibility
Gillespie, T. (2024) · Big Data & Society · doi:10.1177/20539517241252131 · Cited by 122 · see citing works
“But, and here ’ s the second lesson, the more profound problem is that a handful of design teams at a handful companies get to calibrate these tools to approximate fairness, according to their own corporate sensibilities (Lazar and Nelson, 2023).”
Why human–AI relationships need socioaffective alignment
Kirk, H. R., Gabriel, I., Summerfield, C., Vidgen, B., & Hale, S. A. (2025) · Humanities and Social Sciences Communications · doi:10.1057/s41599-025-04532-5 · Cited by 146 · see citing works
“…that many of the outstanding challenges in AI alignment extend beyond purely ‘ technical ’ issues with the model or its training data (Lazar and Nelson, 2023; Weidinger et al. 2023) — and will continue to persist even if we develop effective techniques for steering the behaviour of…”
Black-Box Access is Insufficient for Rigorous AI Audits
Casper, S., Ezell, C., Siegmann, C., Kolt, N., Curtis, T. L., Bucknall, B., Haupt, A. A., Wei, K., Scheurer, J., Hobbhahn, M., et al. (2024) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3630106.3659037 · Cited by 104 · see citing works
Building machines that learn and think with people
Collins, K. M., Sucholutsky, I., Bhatt, U., Chandra, K., Wong, L., Lee, M., Zhang, C. E., Zhi‐Xuan, T., Ho, M. K., Mansinghka, V. K., et al. (2024) · Nature Human Behaviour · doi:10.1038/s41562-024-01991-9 · Cited by 93 · see citing works
Managing workplace AI risks and the future of work
Howard, J., & Schulte, P. A. (2024) · American Journal of Industrial Medicine · doi:10.1002/ajim.23653 · Cited by 54 · see citing works
Advances in the Application of AI Robots in Critical Care: Scoping Review
Li, Y., Wang, M., Wang, L., Cao, Y. Y., Liu, Y., Zhao, Y., Rui, Y., Yang, M., Lu, S., Sun, Z., et al. (2024) · Journal of Medical Internet Research · doi:10.2196/54095 · Cited by 48 · see citing works
“Narrow technical approaches are insufficient to ensure the safety of AI robots, which must be considered within the broader sociotechnical context in which they operate [130].”
Gyevnár, B., & Kasirzadeh, A. (2025) · Nature Machine Intelligence · doi:10.1038/s42256-025-01020-y · Cited by 40 · see citing works
“As the relationship between AI safety and these normative theories and movements is quite complex, the association between AI safety and the conventional conception of AI existential risk is commonly presented as a defining characteristic [2, 3, 10–12].”
Beyond Preferences in AI Alignment
Zhi-Xuan, T., Carroll, M., Franklin, M., & Ashton, H. (2024) · Philosophical Studies · doi:10.1007/s11098-024-02249-w · Cited by 85 · see citing works
“Insofar as these AI systems are used to exercise power over others, they should also act in accordance with standards that are not just fair, but legitimate (Lazar, 2023; Stone and Mittelstadt, 2024).”
BackgroundFuture Shock: Generative AI and the International AI Policy and Governance Crisis
Leslie, D., & Perini, A. M. (2024) · Harvard Data Science Review · doi:10.1162/99608f92.88b4cc98 · Cited by 33 · see citing works
Li, B., & Gilbert, S. (2024) · npj Digital Medicine · doi:10.1038/s41746-024-01345-9 · Cited by 34 · see citing works
Competing narratives in AI ethics: a defense of sociotechnical pragmatism
Watson, D., Mökander, J., & Floridi, L. (2024) · AI & Society · doi:10.1007/s00146-024-02128-2 · Cited by 36 · see citing works
“11 In applied contexts, algorithms inevitably form part of larger sociotechnical systems that encapsulate other artifacts, people, and organizations (Lazar and Nelson 2023).”
Should Users Trust Advanced AI Assistants? Justified Trust As a Function of Competence and Alignment
Manzini, A., Keeling, G., Marchal, N., McKee, K. R., Rieser, V., & Gabriel, I. (2024) · The 2024 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3630106.3658964 · Cited by 28 · see citing works
The necessity of AI audit standards boards
Manheim, D., Martin, S., Bailey, M., Samin, M., & Greutzmacher, R. (2025) · AI & Society · doi:10.1007/s00146-025-02320-y · Cited by 45 · see citing works
“This shows that, while sociotechnical approaches and involving broader stakeholder groups are critical (Lazar and Nelson 2023), those approaches need far more than static evaluation to be meaningfully effective in reducing harm.”
Manning, L., & Grant, J. H. (2024) · Trends in Food Science & Technology · doi:10.1016/j.tifs.2024.104811 · Cited by 18 · see citing works
Mapping the individual, social and biospheric impacts of Foundation Models
Hernández, A. D., Krishna, S., Perini, A., Katell, M. A., Bennett, S., Borda, A., Hashem, Y., Hadjiloizou, S., Mahomed, S., Jayadeva, S., et al. (2024) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3630106.3658939 · Cited by 20 · see citing works
Arzani, S., Karimi, A., Iranmanesh, P., Yazdi, M., Sabeti, M. A., Nekoofar, M. H., Kolahi, J., Bang, H., & Dummer, P. M. H. (2025) · PLoS ONE · doi:10.1371/journal.pone.0329986 · Cited by 16 · see citing works
“The dental research community, journal editors, clinicians, and policymakers should be aware and be attentive of the significant and emerging concerns regarding AI safety [79] and ethical worries regarding the use of AI systems in biomedical research and clinical practice [80,81].”
Trustworthy AI for Whom? GenAI Detection Techniques of Trust Through Decentralized Web3 Ecosystems
Calzada, I., Németh, G., & Al-Radhi, M. S. (2025) · Big Data and Cognitive Computing · doi:10.3390/bdcc9030062 · Cited by 16 · see citing works
Lessons from complex systems science for AI governance
Kolt, N., Shur‐Ofry, M., & Reuven Cohen, C. (2025) · Patterns · doi:10.1016/j.patter.2025.101341 · Cited by 22 · see citing works
“3 The discourse is, to a large extent, influenced by conceptual framing.”
Navigating AI conformity: A design framework to assess fairness, explainability, and performance
von Zahn, M., Zacharias, J., Lowin, M., Chen, J., & Hinz, O. (2025) · Electronic Markets · doi:10.1007/s12525-025-00770-2 · Cited by 12 · see citing works
“For instance, privacy can have widely different interpretations depending on the context (Chua et al., 2021), and best practices for AI safety and cybersecurity are continually evolving (Lazar & Nelson, 2023).”
Towards Guaranteed Safe AI: A Framework for Ensuring Robust and Reliable AI Systems
Dalrymple, D. "., Skalse, J., Bengio, Y., Russell, S., Tegmark, M., Seshia, S., Omohundro, S., Szegedy, C., Goldhaber, B., Ammann, N., et al. (2024) · arXiv · doi:10.48550/arxiv.2405.06624 · Cited by 12 · see citing works
Intersectional biases in narratives produced by open-ended prompting of generative language models
Shieh, E., Vassel, F.-M., Sugimoto, C. R., & Monroe‐White, T. (2026) · Nature Communications · doi:10.1038/s41467-025-68004-9 · Cited by 14 · see citing works
“Such framings give greater attention to malicious users, national security concerns, or future existential risks at the expense of safeguarding fundamental human rights 91,59 .”
Limiting bias in AI models for improved and equitable cancer care
Ghassemi, M., & Gusev, A. (2024) · Nature reviews. Cancer · doi:10.1038/s41568-024-00739-x · Cited by 8 · see citing works
Mariñas, K. A., Saflor, C. S., Alvarado, P., Uminga, J. M., & Verde, N. A. (2025) · Computers and Education Artificial Intelligence · doi:10.1016/j.caeai.2025.100435 · Cited by 14 · see citing works
Ethical and Epistemological Reflections on Autonomous AI-powered Agents (AAIAs)
Farina, M., Wang, Y., & Kladko, S. (2025) · Topoi · doi:10.1007/s11245-025-10218-z · Cited by 9 · see citing works
How public involvement can improve the science of AI
Matias, J. N., & Price, M. (2025) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2421111122 · Cited by 9 · see citing works
Hernández, A. D., Perini, A., Hadjiloizou, S., Borda, A., Mahomed, S., & Leslie, D. (2025) · AI and Ethics · doi:10.1007/s43681-025-00902-6 · Cited by 11 · see citing works
“Moreover, an evaluation produced by the Coun - cil of Europe’s Ad hoc Committee on Artificial Intelligence It is clear that the current landscape of global AI ini - tiatives has a disadvantaged representation of the global majority [ 17 , 104 ].”
A Framework for Evaluating Global AI Governance Initiatives
Roberts, H. R., Taddeo, M., & Floridi, L. (2026) · Global Policy · doi:10.1111/1758-5899.70164 · Cited by 7 · see citing works
Governing Artificial Intelligence Through a Sociotechnical Lens
Filgueiras, F. D. B., Mendonça, R. F., & Almeida, V. A. F. (2023) · IEEE Internet Computing · doi:10.1109/mic.2023.3310110 · Cited by 7 · see citing works
“Second, algorithms hold a central role in AI as they establish the rules, standards, and common approaches that institutionalize new practices, thereby upending the conventional manners in which industries, governments, and societies foster coexistence and initiatives.(1,3) The delineation between these two scenarios…”
BackgroundAI red-teaming is a sociotechnical problem: on values, labor, and harms
Gillespie, T., Shaw, R., Gray, M. L., & Suh, J. (2024) · arXiv · doi:10.48550/arxiv.2412.09751 · Cited by 6
Butler, J. M., Doubleday, A., Sattar, U., Nies, M. A., Jeppesen, A., Wright, M. C., Reese, T., Kawamoto, K., Fiol, G. D., & Madaras‐Kelly, K. (2025) · Applied Clinical Informatics · doi:10.1055/a-2505-7743 · Cited by 5 · see citing works
Public trust in AI: A dynamic social media view
Thaiduong, N. (2026) · Social Sciences & Humanities Open · doi:10.1016/j.ssaho.2026.102457 · Cited by 6 · see citing works
Computational hermeneutics: evaluating generative AI as a cultural technology
Kommers, C., Ahnert, R., Antoniak, M., Benetos, E., Benford, S. A., Bunz, M., Caramiaux, B., Concannon, S. J., Disley, M., Dobson, J. E., et al. (2026) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2026.1753041 · Cited by 19 · see citing works
“Each of these already exists in one form or another in contemporary AI (Akbulut et al., 2025; Lazar and Nelson, 2023; Sorensen et al., 2024).”
Strategic insights from simulation gaming of AI race dynamics
Gruetzemacher, R., Avin, S., Fox, J., & Saeri, A. K. (2025) · Futures · doi:10.1016/j.futures.2025.103563 · Cited by 10 · see citing works
Quantifying emergent complexity
Hoel, E. P. (2026) · Patterns · doi:10.1016/j.patter.2025.101472 · Cited by 5 · see citing works
“…networks 56 and how the causal structure changes following learning in artificial neural networks, 57 CE 2.0 may also be primed to contribute to the growing field of AI interpretability 58 and AI safety, 59 such as analyzing the multi-scale structure of deep neural networks and their concepts.”
InfluentialTowards friendly AI: a comprehensive review and new perspectives on human-AI alignment
Sun, Q., Li, Y., Alturki, E., Murthy, S. M. K., & Schuller, B. W. (2026) · AI and Ethics · doi:10.1007/s43681-026-01070-x · Cited by 12 · see citing works
Historical Methods for AI Evaluations, Assessments, and Audits
Sandoval, . C. B., & Jing, F. S. (2025) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3715275.3732093 · Cited by 4 · see citing works
Beyond accidents and misuse: decoding the structural risk dynamics of artificial intelligence
Kilian, K. A. (2025) · AI & Society · doi:10.1007/s00146-025-02419-2 · Cited by 4 · see citing works
“Thus, researchers are increasingly evaluating AI through a complex systems lens, focusing on how its structure, function, and relations with parallel systems influence risk dynamics (Lazar and Nelson 2023; Weidinger et al. 2023; Kilian et al. 2023; Pilditch 2024a, b).”
Race and Privacy in Broadcast Police Communications
Venkit, P. N., Graziul, C., Goodman, M., Kenny, S. N., & Wilson, S. (2024) · Proc. ACM Hum. Comput. Interact. · doi:10.1145/3686921 · Cited by 6 · see citing works
How Might Artificial Intelligence Influence Human Evolution?
Brooks, R. C. (2024) · The Quarterly Review of Biology · doi:10.1086/733290 · Cited by 4 · see citing works
Pulsar’s Application in Energy Systems: Review of Current Status, Challenges, and Opportunities
Zhang, Y., Wu, Y., Liu, Y., Sun, B., Yang, J., Thotakura, N. L., Ahmed, K. I., Liu, Q., Shen, C., & Liu, Y. (2025) · Energies · doi:10.3390/en18040828 · Cited by 4 · see citing works
AI Safety: A Poisoned Chalice?
Nissenbaum, H. F. (2024) · IEEE Security & Privacy · doi:10.1109/msec.2024.3356848 · Cited by 3 · see citing works
What is AI safety? What do we want it to be?
Harding, J., & Kirk‐Giannini, C. D. (2025) · Philosophical Studies · doi:10.1007/s11098-025-02367-z · Cited by 16 · see citing works
“These suggestions, we anticipate, will strike many readers as obviously sensible, continuous with many proposals to broaden AI safety’s tent in recent years (Lazar and Nelson 2023).”
Rask, M., & Shimizu, K. (2024) · International Conference on Artificial Intelligence, Computer, Data Sciences and Applications (ACDSA) · doi:10.1109/acdsa59508.2024.10467341 · Cited by 5 · see citing works
“The social sciences are recognized as being vital for understanding and managing the risks associated with LLMs, yet detailed studies are scarce [5].”
BackgroundEstep, P. W. (2024) · AI & Society · doi:10.1007/s00146-024-02134-4 · Cited by 3 · see citing works
Pan-Evo: The Evolution of Information and Biology’s Part in This
Sherwin, W. B. (2024) · Biology · doi:10.3390/biology13070507 · Cited by 3 · see citing works
Sandoval, . C. B., & Jing, F. S. (2025) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3715275.3732062 · Cited by 2 · see citing works
Catch Me If You Can: Rogue AI Detection and Correction at Scale
Stodt, F., Stodt, J., Alshawki, M., Sartakhti, J. S., & Reich, C. (2025) · Electronics · doi:10.3390/electronics14204122 · Cited by 2 · see citing works
Exploring expert and public perceptions of answerability and trustworthy autonomous systems
Hatherall, L., & Sethi, N. (2025) · Journal of Responsible Technology · doi:10.1016/j.jrt.2025.100106 · Cited by 3 · see citing works
AI Trustworthy: Ethical Challenges and Strategies
Liu, J., Sandjaja, I. N., & Wunsch, D. C. (2024) · IEEE International Conference on Service Operations and Logistics, and Informatics (SOLI) · doi:10.1109/soli63266.2024.10956105 · Cited by 2 · see citing works
“Broader social participation in AI development, including diverse perspectives, is crucial for addressing AI’s risks [70-71].”
Estimating divergent moral and diversity preferences between AI builders and AI users
Purcell, Z. A., Charbit, L., Borst, G., & Nußberger, A.-M. (2025) · Cognition · doi:10.1016/j.cognition.2025.106198 · Cited by 2 · see citing works
“We focus on utilitarianism due to its strong presence in philosophical movements popular among AI developers (Bordelon, 2023; Clarke, 2023; Lazar & Nelson, 2023; McMillan & Seetharaman, 2023).”
Roadmap on Incentive Compatibility for AI Alignment and Governance in Sociotechnical Systems
Zhang, Z., Bai, F., Wang, M., Ye, H., Ma, C., & Yang, Y. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-00800-8_33 · Cited by 3 · see citing works
“Some studies have indicated that relying solely on technical means will result in a sociotechnical gap between the model’s development context and its actual deployment context [64,45,79], which is also detrimental to further social governance.”
Human–AI social cooperation: convergences, divergences, and challenges
Yao, S., Shan, J., Hu, L., & lu, X. (2025) · Chinese Science Bulletin (Chinese Version) · doi:10.1360/csb-2025-0662 · Cited by 1 · see citing works
Zheng, E. L., Cho, M. K., & Lee, S. S. (2026) · Cambridge Forum on AI Culture and Society · doi:10.1017/cfc.2026.10030 · Cited by 1 · see citing works
Do Citizens Agree with the EU AI Act? Public Perspectives on Risk and Regulation of AI Systems
Lima, G., Gasiola, G. G., Zufall, F., & Zou, Y. (2026) · International Conference on Human Factors in Computing Systems · doi:10.1145/3772318.3790535 · Cited by 1 · see citing works
IEEE Pervasive Computing - Call for Articles
(2024) · IEEE Security & Privacy · doi:10.1109/msec.2024.3369438
Research Agenda for Sociotechnical Approaches to AI SafetyPreprint
Curtis, S., Iyer, R., Kirk‐Giannini, C. D., Krakovna, V., Krueger, D., Lambert, N., Marnette, B., McKenzie, C., Michael, J., Miyazono, E., et al. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5097286
Leirós, C. B. (2025) · Convivium Germanistisches Jahrbuch Polen · doi:10.18778/2196-8403.s2.08
Be Aware the Perils of Solutionism in AI SafetyPreprint
Pi, Y. (2025) · Qeios · doi:10.32388/tjkh93
Science and Causality in Technology Litigation
Matias, J. N., & Penney, J. W. (2025) · Journal of Online Trust and Safety · doi:10.54501/jots.v2i5.233
Adamu, M. S., Lazem, S., Nkwo, M. S., & Peters, A. N. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5347484
A Study of Suppression Strategies for AI Hallucinations in Personal Learning Scenarios
Man, X. (2025) · International Conference on Image Processing, Intelligent Control and Computer Engineering (IPICE) · doi:10.1145/3768184.3768227
“[ 7 ][ 8 ]”
Sustainable Urban Mobility: Co-Designing a Responsible AI Recommender System
Patelli, A., Ekárt, A., Chli, M., Ferariu, L.-E., Hamilton, J. R., Kanyi, M., Lee, R., Lewis, P., Lumsden, J., & Owen, S. (2025) · International Conference on Information Technology for Social Good (GoodIT) · doi:10.1145/3748699.3749783
Trustworthy AI for Whom in Smart Cities?
Calzada, I. (2025) · Studies in digital politics and governance · doi:10.1007/978-3-032-11888-2_3
Systematic Review on AI Safety and Future Research Agenda
Zhang, S., & Lu, H. (2025) · International Conference on System Reliability and Safety Engineering (SRSE) · doi:10.1109/srse67406.2025.11357293
“The concept of AI safety, therefore, has emerged as a paramount concern, encompassing a wide array of considerations aimed at mitigating potential harms, ensuring human control, and fostering public trust in these powerful technologies [4].”
Artificial intelligence and medicine: an epistemological inquiry
Abastado, P. (2026) · Medical Humanities · doi:10.1136/medhum-2025-013509
Ebo, T. O., Osunmakinde, A., Asaolu, A. J., Ebo, D. M., Egbon, E., & Olawade, D. B. (2026) · AI & Society · doi:10.1007/s00146-026-02930-0 · Cited by 1 · see citing works
Explaining the Biased Agent: Coherent Bayesian Updating under Worldview DistortionsPreprint
Cirillo, P. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6140488
Future Concepts of Artificial Intelligence
ŞAHİN, H. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/406016
Navigating the Political, Ethical, and Epistemological Landscape of Foundation Models
Durán, J. M., Maas, J., & Jonker, C. M. (2026) · Springer eBooks · doi:10.1007/978-3-031-61050-9_29-1
Lee, C. J. (2026) · Social Epistemology · doi:10.1080/02691728.2026.2669578
“…and financial priorities of the most powerful and well-resourced, including AI tool creators, should not be entrenched (Birhane et al. 2022; Lazar and Nelson 2023; Selbst et al. 2019) in ways that thwart norms fostering critical discursive engagement, including perspectival diversity and…”
Yu, W.-N., Hou, X.-Y., & Zhang, X.-R. (2026) · International Journal of Contemporary Hospitality Management · doi:10.1108/ijchm-11-2025-1679
Chan, H., & Erduran, S. (2026) · Science & Education · doi:10.1007/s11191-026-00770-4
Artificial Intelligence and democracy: A Delphi Study Into Opportunities and Threats
Kars, J., Hillebrandt, M. Z., & Meijer, A. J. (2026) · Information Polity · doi:10.1177/15701255261492066
- Authors
- Lazar, S., & Nelson, A.
- Affiliations
- Australian National University (AU)
- Institute for Advanced Study (US)
- Venue
- Science
- Published
- 2023-07-14
- Volume
- 381
- Issue
- 6654
- Pages
- 138-138
- ISSN
- 0036-8075 (ISSN-L)
- 1095-9203
- OpenAlex
- W4384155145
- Cited by
- 77
- References
- 0
- Citations by year
- 2026: 19 · 2025: 32 · 2024: 25 · 2023: 1
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- Science (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Social Sciences › Safety Research › Ethics and Social Impacts of AI
- Physical Sciences › Engineering › Safety, Risk, Reliability and Quality › Safety Systems Engineering in Autonomy
- Social Sciences › Law › Digitalization, Law, and Regulation
- Keywords
- AI safety, large language models, research priorities, socio-technical approach, foundation models, government AI policy, AI governance
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
