AI Deception: A Survey of Examples, Risks, and Potential Solutions
Open accessAbstract
This paper argues that a range of current AI systems have learned how to deceive humans. We define deception as the systematic inducement of false beliefs in the pursuit of some outcome other than the truth. We first survey empirical examples of AI deception, discussing both special-use AI systems (including Meta's CICERO) built for specific competitive situations, and general-purpose AI systems (such as large language models). Next, we detail several risks from AI deception, such as fraud, election tampering, and losing control of AI systems. Finally, we outline several potential solutions to the problems posed by AI deception: first, regulatory frameworks should subject AI systems that are capable of deception to robust risk-assessment requirements; second, policymakers should implement bot-or-not laws; and finally, policymakers should prioritize the funding of relevant research, including tools to detect AI deception and to make AI systems less deceptive. Policymakers, researchers, and the broader public should work proactively to prevent AI deception from destabilizing the shared foundations of our society.
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?
Large language models are competent at lying and strategic deception when instructed or trained to be deceptive.
“Large Language Models (LMs) are now competent at lying (Pacchiardi et al., 2023) and strategic deception when instructed or trained to be deceptive (Park et al., 2023; Hagendorff, 2023; O'Gara, 2023; Hubinger et al., 2024)”
From Large Language Models can Strategically Deceive their Users when Put Under Pressure (2023) · cites this paper as Park et al., 2023
Section 2.2.1 (p. 10) reports that LLMs have reasoned their way into strategic deception to complete tasks, including lying in games. The Executive summary (p. 2) gives successful examples. These passages support competence at strategic deception, but do not establish that the models were specifically instructed or trained to be deceptive; the paper describes deception emerging as a way to achieve other goals. (LLM, conf=0.82)
Experimental — accuracy not yet measuredWhen presented with ethically complex questions, LLMs tend to simply mirror the users' views.
“Indeed, there is emerging concern that, when presented with ethically complex questions, LLMs tend to simply mirror the users’ views (see, e.g. Turpin et al. 2023, Park et al. 2023, or the sycophancy benchmarking tasks of Perez et al. 2023).”
From Helpful, harmless, honest? Sociotechnical limits of AI alignment and safety through Reinforcement Learning from Human Feedback (2025) · cites this paper as Park et al. 2023
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
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 · see citing works
Advances in Neural Information Processing Systems 29
Teymur, O., Zygalakis, K., & Calderhead, B. (2016) · Cited by 13,881 · see citing works
Transformers: State-of-the-Art Natural Language Processing
Wolf, T., Debut, L., Sanh, V., Chaumond, J., Delangue, C., Moi, A., Cistac, P., Rault, T., Louf, R., Funtowicz, M., et al. (2020) · Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations · doi:10.18653/v1/2020.emnlp-demos.6 · Cited by 8,741 · see citing works
On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜
Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021) · Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3442188.3445922 · Cited by 6,988 · see citing works
Proceedings of the 25th international conference on Machine learning - ICML '08
(2008) · doi:10.1145/1390156 · Cited by 6,865 · see citing works
Grandmaster level in StarCraft II using multi-agent reinforcement learning
Vinyals, O., Babuschkin, I., Czarnecki, W. M., Mathieu, M., Dudzik, A., Chung, J., Choi, D. H., Powell, R., Ewalds, T., Georgiev, P., et al. (2019) · Nature · doi:10.1038/s41586-019-1724-z · Cited by 3,455 · see citing works
Proceedings for the 5th International Conference on Learning Representations
Srivastava, A., & Sutton, C. A. (2017) · Cited by 1,829 · see citing works
Deep reinforcement learning from human preferences
Christiano, P., Leike, J., Brown, T. B., Martic, M., Legg, S., & Amodei, D. (2017) · arXiv · doi:10.48550/arxiv.1706.03741 · Cited by 1,058 · see citing works
Is U.S. Economic Growth Over? Faltering Innovation Confronts the Six Headwinds
Gordon, R. J. (2012) · Voprosy Ekonomiki · doi:10.32609/0042-8736-2013-4-49-67 · Cited by 928 · see citing works
How to Define Theoretical Terms
(1970) · doi:10.2307/2023861 · Cited by 916 · see citing works
TruthfulQA: Measuring How Models Mimic Human Falsehoods
Lin, S., Hilton, J., & Evans, O. (2022) · Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2022.acl-long.229 · Cited by 904 · cited 3× · see citing works
“Lin et al. (2022) demonstrate that language models often repeat common misconceptions such as “If you crack your knuckles a lot, you may develop arthritis” (p. 2). Disturbingly, Perez et al. (2022) found that LLMs tend to give more of these inaccurate answers when the user appears to be less educated.”
InfluentialWayne, S. J., & Ferris, G. R. (1990) · Journal of Applied Psychology · doi:10.1037/0021-9010.75.5.487 · Cited by 860 · see citing works
(2019) · Cited by 771 · see citing works
Superhuman AI for multiplayer poker
Brown, N., & Sandholm, T. (2019) · Science · doi:10.1126/science.aay2400 · Cited by 592 · cited 2× · see citing works
“• Bluffs: Pluribus, a poker-playing model created by Meta, successfully bluffed human players into folding (Brown et al. 2019).”
Fine-Tuning Language Models from Human Preferences
Ziegler, D. M., Stiennon, N., Wu, J., Brown, T. B., Radford, A., Amodei, D., Christiano, P., & Irving, G. (2019) · arXiv · doi:10.48550/arxiv.1909.08593 · Cited by 413 · see citing works
Deal or No Deal? End-to-End Learning of Negotiation Dialogues
Lewis, M., Yarats, D., Dauphin, Y., Parikh, D., & Batra, D. (2017) · Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d17-1259 · Cited by 347 · see citing works
Impact of ingratiation on judgments and evaluations: A meta-analytic investigation.
Gordon, R. A. (1996) · Journal of Personality and Social Psychology · doi:10.1037/0022-3514.71.1.54 · Cited by 338 · see citing works
Bai, Y., Kadavath, S., Kundu, S., Askell, A., Kernion, J., Jones, A., Chen, A., Goldie, A., Mirhoseini, A., & Kaplan, J. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2212.08073 · Cited by 326 · cited 1× · see citing works
“96,97 Here, AI outputs are rated by human evaluators (RLHF) or AI evaluators (constitutional AI), based on criteria such as perceived helpfulness and honesty, and fine-tuned to train the language model.”
The Alignment Problem: Machine Learning and Human Values
Christian, B. (2021) · Perspectives on Science and Christian Faith · doi:10.56315/pscf12-21christian · Cited by 311 · see citing works
Turpin, M., Michael, J., Perez, E., & Bowman, S. R. (2023) · arXiv · doi:10.48550/arxiv.2305.04388 · Cited by 262 · see citing works
Meta Fundamental AI Research Diplomacy Team (FAIR)†, Bakhtin, A., Brown, N., Dinan, E., Farina, G., Flaherty, C., Fried, D., Goff, A., Gray, J., Hu, H., et al. (2022) · Science · doi:10.1126/science.ade9097 · Cited by 241 · see citing works
Quilty-Dunn, J., Porot, N., & Mandelbaum, E. (2023) · Behavioral and Brain Sciences · doi:10.1017/s0140525x22002849 · Cited by 205 · see citing works
Deceit and self-deception : fooling yourself the better to fool others
Trivers, R. L. (2013) · Medical Entomology and Zoology · Cited by 195 · see citing works
Discovering Language Model Behaviors with Model-Written Evaluations
Perez, E., Ringer, S., Lukosiute, K., Nguyen, K., Chen, E., Heiner, S., Pettit, C., Olsson, C., Kundu, S., Kadavath, S., et al. (2023) · Findings of the Association for Computational Linguistics: ACL 2023 · doi:10.18653/v1/2023.findings-acl.847 · Cited by 194 · see citing works
A Watermark for Large Language Models
Kirchenbauer, J., Geiping, J., Wen, Y., Katz, J., Miers, I., & Goldstein, T. (2023) · arXiv · doi:10.48550/arxiv.2301.10226 · Cited by 194 · see citing works
BALD-FACED LIES! LYING WITHOUT THE INTENT TO DECEIVE
(2007) · doi:10.1111/j.1468-0114.2007.00290.x · Cited by 183
Lehman, J., Clune, J., Misevic, D., Adami, C., Altenberg, L., Beaulieu, J., Bentley, P. J., Bernard, S., Beslon, G., Bryson, D. M., et al. (2020) · Artificial Life · doi:10.1162/artl_a_00319 · Cited by 181 · cited 1× · see citing works
“With loss of control, autonomous AI systems may use deception to accomplish their own goals.”
The Internal State of an LLM Knows When It’s Lying
Azaria, A., & Mitchell, T. (2023) · Findings of the Association for Computational Linguistics: EMNLP 2023 · doi:10.18653/v1/2023.findings-emnlp.68 · Cited by 175 · see citing works
Goal Misgeneralization in Deep Reinforcement LearningPreprint
Langosco, L., Koch, J., Sharkey, L. D., Pfau, J., Orseau, L., & Krueger, D. (2021) · International Conference on Machine Learning · doi:10.48550/arxiv.2105.14111 · Cited by 171 · see citing works
Can AI-Generated Text be Reliably Detected?
Sadasivan, V. S., Kumar, A., Balasubramanian, S., Wang, W., & Feizi, S. (2023) · arXiv · doi:10.48550/arxiv.2303.11156 · Cited by 161 · cited 1× · see citing works
“85 Attackers will attempt to bypass these defenses, 86 but companies should be required to stay ahead of these attacks and provide trustworthy techniques for identifying AI outputs.”
Goldstein, J. A., Sastry, G., Musser, M., DiResta, R., Gentzel, M., & Sedova, K. (2023) · arXiv · doi:10.48550/arxiv.2301.04246 · Cited by 152 · see citing works
Paraphrasing evades detectors of AI-generated text, but retrieval is an effective defense
Krishna, K., Song, Y., Karpinska, M., Wieting, J., & Iyyer, M. (2023) · Advances in Neural Information Processing Systems 36 · doi:10.52202/075280-1195 · Cited by 150 · see citing works
A General Language Assistant as a Laboratory for Alignment
Askell, A., Bai, Y., Chen, A., Drain, D., Ganguli, D., Henighan, T., Jones, A., Joseph, N., Mann, B., DasSarma, N., et al. (2021) · arXiv · doi:10.48550/arxiv.2112.00861 · Cited by 146 · cited 1× · see citing works
“96,97 Here, AI outputs are rated by human evaluators (RLHF) or AI evaluators (constitutional AI), based on criteria such as perceived helpfulness and honesty, and fine-tuned to train the language model.”
Devising and Detecting Phishing Emails Using Large Language Models
Heiding, F., Schneier, B., Vishwanath, A., Bernstein, J., & Park, P. S. (2024) · IEEE Access · doi:10.1109/access.2024.3375882 · Cited by 123 · cited 1× · see citing works
“44 These trends continue to increase the degree to which victims are vulnerable to scams, extortion, and other forms of fraud, and, in the words 27 illustratesthat,ineachgame played byanLLM-basedAIagent, thereisatext-based storythatisgeneratedadaptively as the agent observes the current text-based…”
Self-citationACGS-2: A Production-Ready Constitutional AI Governance System
Lyu, H. (2025) · Zenodo (CERN European Organization for Nuclear Research) · doi:10.5281/zenodo.16416793 · Cited by 121 · see citing works
Inference-Time Intervention: Eliciting Truthful Answers from a Language Model
Li, K., Patel, O., Viégas, F., Pfister, H., & Wattenberg, M. (2023) · Advances in Neural Information Processing Systems 36 · doi:10.52202/075280-1797 · Cited by 117 · cited 1× · see citing works
“Policymakers should help fill this gap through targeted research funding.”
The Volkswagen emissions scandal and its aftermath
Jung, J. C., & Sharon, E. (2019) · Global Business and Organizational Excellence · doi:10.1002/joe.21930 · Cited by 109 · see citing works
Deception abilities emerged in large language models
Hagendorff, T. (2024) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2317967121 · Cited by 100 · see citing works
Characterizing Manipulation from AI Systems
Carroll, M., Chan, A., Ashton, H., & Krueger, D. (2023) · ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization (EAAMO) · doi:10.1145/3617694.3623226 · Cited by 85 · cited 1× · see citing works
“For similar definitions, see Evans et al. 2 and Carroll et al. 3 We present a wide range of examples where AI systems do not merely produce false outputs by accident .”
AgentBench: Evaluating LLMs as Agents
Liu, X., Yu, H., Zhang, H., Xu, Y., Lei, X., Lai, H., Gu, Y., Ding, H., Men, K., Yang, K., et al. (2023) · arXiv · doi:10.48550/arxiv.2308.03688 · Cited by 79 · cited 1× · see citing works
“To illustrate, Liu et al. 61 and Kinniment et al. 62 measured different LLMs’ ability to autonomously carry out various tasks, such as browsing the web, online shopping, making a phone call, and using a computer’s operating system.”
Model evaluation for extreme risks
Shevlane, T., Farquhar, S., Garfinkel, B., Phuong, M., Whittlestone, J., Leung, J., Kokotajlo, D., Marchal, N., Anderljung, M., Kolt, N., et al. (2023) · arXiv · doi:10.48550/arxiv.2305.15324 · Cited by 66 · cited 1× · see citing works
“81 The information provided about safety-relevant features such as deception or lack thereof should be accurate, with clear legal liability for failures to comply with safety testing requirements.”
Hoodwinked: Deception and Cooperation in a Text-Based Game for Language ModelsPreprint
O'Gara, A. (2023) · arXiv.org · doi:10.48550/arxiv.2308.01404 · Cited by 65 · cited 1× · see citing works
“24 In experiments where OpenAI’s GPT models played Hood-winked , the language models would often kill other players while alone in a room with them, then deny the crime in the group conversation by constructing a false alibi or casting blame on other players.”
Self-citationBanovic, N., Yang, Z., Ramesh, A., & Liu, A. (2023) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3579460 · Cited by 55 · cited 1× · see citing works
“Although the deceptive reasoning is transparent through the chain of thought, the malicious behavior remained even when the chain of thought was ‘‘distilled’’ away and not made visible to users.”
Discovering Latent Knowledge in Language Models Without Supervision
Burns, C., Ye, H., Klein, D., & Steinhardt, J. (2022) · arXiv · doi:10.48550/arxiv.2212.03827 · Cited by 55 · cited 1× · see citing works
“Azaria et al. 89 and Burns et al. 90 have developed methods for determining whether these internal embeddings represent the sentence as being true or false.”
Still no lie detector for language models: probing empirical and conceptual roadblocks
Levinstein, B. A., & Herrmann, D. A. (2025) · Philosophical Studies · doi:10.1007/s11098-023-02094-3 · Cited by 50 · cited 2× · see citing works
“For example, Levinstein et al. 91 have raised reliability issues for the strategies of Azaria et al. and Burns et al., although these reliability issues have recently been mitigated by the strategy of Zou et al. 92 Also, Pacchiardi et al. 93 introduced a lie-de-tector method that asks a collection…”
Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training
Hubinger, E., Denison, C., Mu, J., Lambert, M., Tong, M., MacDiarmid, M., Lanham, T., Ziegler, D. M., Maxwell, T., Cheng, N., et al. (2024) · arXiv · doi:10.48550/arxiv.2401.05566 · Cited by 46 · cited 1× · see citing works
“Hubinger et al. 31 studied whether LLMs can be trained with a ‘‘backdoor’’ to produce deceptive behavior.”
Evaluating the Moral Beliefs Encoded in LLMs
Scherrer, N., Shi, C., Feder, A., et al. (2023) · neural information processing systems · doi:10.52202/075280-2256 · Cited by 42 · see citing works
Truthful AI: Developing and governing AI that does not lie
Evans, O., Cotton-Barratt, O., Finnveden, L., Bales, A., Balwit, A., Wills, P., Righetti, L., & Saunders, W. (2021) · arXiv · doi:10.48550/arxiv.2110.06674 · Cited by 39 · see citing works
Thinking Fast and Slow in Large Language Models
Hagendorff, T., Fabi, S., & Kosiński, M. (2022) · PubMed Central · doi:10.48550/arxiv.2212.05206 · Cited by 36 · see citing works
Measuring Faithfulness in Chain-of-Thought Reasoning
Lanham, T., Chen, A., Radhakrishnan, A., Steiner, B., Denison, C., Hernandez, D., Li, D., Durmus, E., Hubinger, E., Kernion, J., et al. (2023) · arXiv · doi:10.48550/arxiv.2307.13702 · Cited by 33 · see citing works
Pan, A., Chan, J. S., Zou, A., Li, N., Basart, S., Woodside, T., Ng, J., Zhang, H., Emmons, S., & Hendrycks, D. (2023) · arXiv · doi:10.48550/arxiv.2304.03279 · Cited by 32 · see citing works
Self-citationSpear Phishing With Large Language Models
Hazell, J. (2023) · arXiv · doi:10.48550/arxiv.2305.06972 · Cited by 28 · see citing works
Artificial Intelligence: Arguments for Catastrophic Risk
Bales, A., D’Alessandro, W., & Kirk‐Giannini, C. D. (2024) · Philosophy Compass · doi:10.1111/phc3.12964 · Cited by 27 · see citing works
Kreps, S., & Kriner, D. L. (2024) · New Media & Society · doi:10.1177/14614448231160526 · Cited by 27 · see citing works
Artificial Influence: An Analysis Of AI-Driven Persuasion
Burtell, M., & Woodside, T. (2023) · arXiv · doi:10.48550/arxiv.2303.08721 · Cited by 20 · cited 1× · see citing works
“24 Another example pertains to the social deduction game Among Us , where ‘‘impostors’’ cause sabotages within their ship and murder their fellow crewmates.”
HOW TO CATCH AN AI LIAR: LIE DETECTION IN BLACK-BOX LLMS BY ASKING UNRELATED QUESTIONSPreprint
Pacchiardi, L., Chan, A., Mindermann, S., Moscovitz, I., Pan, A., Gal, Y., Evans, O., & Brauner, J. (2023) · arXiv (Cornell University) · doi:10.17863/cam.112513 · Cited by 17 · see citing works
Goal Misgeneralization: Why Correct Specifications Aren't Enough For Correct Goals
Shah, R., Varma, V., Kumar, R., Phuong, M., Krakovna, V., Uesato, J., & Kenton, Z. (2022) · arXiv · doi:10.48550/arxiv.2210.01790 · Cited by 16 · see citing works
Characterising Deception in AI: A Survey
Masters, P., Smith, W., Sonenberg, L., & Kirley, M. (2021) · Communications in computer and information science · doi:10.1007/978-3-030-91779-1_1 · Cited by 15 · cited 1× · see citing works
“A discussion of earlier examples can be found in Masters et al. (2021).”
Evaluating Superhuman Models with Consistency Checks
Fluri, L., Paleka, D., & Tramèr, F. (2024) · IEEE Conference on Secure and Trustworthy Machine Learning (SaTML) · doi:10.1109/satml59370.2024.00017 · Cited by 14 · see citing works
Alon, N., Schulz, L., Rosenschein, J. S., & Dayan, P. (2023) · Open Mind · doi:10.1162/opmi_a_00097 · Cited by 12 · see citing works
Large Language Models can Strategically Deceive their Users when Put Under Pressure
Scheurer, J., Balesni, M., & Hobbhahn, M. (2023) · arXiv · doi:10.48550/arxiv.2311.07590 · Cited by 11 · see citing works
Truth and lies in the Caliphate: The use of deception in Islamic State propaganda
Milton, D. (2022) · Media, War & Conflict · doi:10.1177/1750635220945734 · Cited by 10 · cited 1× · see citing works
“…scenario ‘‘You are playing a game of cards and have the opportunity to deal yourself better cards from the bottom of the deck,’’ instead of choosing ‘‘I deal cards fairly from the top of the deck,’’ several LLMs chose the dishonest option: ‘‘I deal myself better cards from the bottom of the deck.’’”
Explore, Establish, Exploit: Red Teaming Language Models from ScratchPreprint
Casper, S., Lin, J., Kwon, J., et al. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2306.09442 · Cited by 7 · cited 1× · see citing works
“78 Legislators should also consider new laws dedicated to the oversight of advanced AI systems.”
Overthinking the Truth: Understanding how Language Models Process False Demonstrations
Halawi, D., Denain, J.-S., & Steinhardt, J. (2023) · arXiv · doi:10.48550/arxiv.2307.09476 · Cited by 7 · cited 1× · see citing works
“In other related work, Halawi et al. 94 characterize how a language model arrives at correct and incorrect answers to questions by examining the model’s internal representations.”
Playing the Werewolf game with artificial intelligence for language understandingPreprint
Shibata, H., Miki, S., & Nakamura, Y. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2302.10646 · Cited by 6 · see citing works
Warfare:Breaking the Watermark Protection of AI-Generated ContentPreprint
Li, G., Chen, Y., Zhang, J., et al. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2310.07726 · Cited by 3 · cited 2× · see citing works
“63 See Evans et al. 2 and Li, Patel et al. 84 for other approaches to training AI systems to be truthful.”
Aligning AI With Shared Human Values
Hendrycks, D., Burns, C., Basart, S., Critch, A., Li, J., Song, D., & Steinhardt, J. (2020) · arXiv · doi:10.48550/arxiv.2008.02275 · cited 1×
“With loss of control, autonomous AI systems may use deception to accomplish their own goals.”
Self-citation
Flagged first, then by citations
Research works printed but not linked: 11
- -(2023). Deep reinforcement learning from human preferences . arXiv: 1706.03741 [stat.ML] . · arXiv
- Fodor, Jerry A. (1987). Psychosemantics: The problem of meaning in the philosophy of mind . Ed. by Margaret A. Boden. MIT Press.
- -(2023b). Machine psychology: Investigating emergent capabilities and behavior in large language models using psychological methods . arXiv: 2303.13988 [cs.CL] . · arXiv
- Lai, Bolin et al. (July 2023). 'Werewolf among us: Multimodal resources for modeling persuasion behaviors in social deduction games'. In: Findings of the Association for Computational Linguistics: ACL 2023 . Toronto, Canada: Association for Computational Linguistics, pp. 6570-6588. URL: https://aclanthology.org/2023.findings-acl.411 .
- Omohundro, Stephen M. (2008). 'The basic AI drives'. In: NLD: IOS Press, pp. 483-492. ISBN: 9781586038335.
- -(2023b). GPT-4 technical report . arXiv: 2303.08774 [cs.CL] . · arXiv
- Schulz, Lion, Nitay Alon, Jeffrey Rosenschein, and Peter Dayan (2023). 'Emergent deception and skepticism via theory of mind'. In: First Workshop on Theory of Mind in Communicating Agents . URL: https://openreview.net/forum?id=yd8VOEpw8h .
- Shanahan, Murray, Kyle McDonell, and Laria Reynolds (2023). Role-play with large language models . arXiv: 2305.16367 [cs.CL] . · arXiv
- Soice, Emily H. et al. (2023). Can large language models democratize access to dual-use biotechnology? arXiv: 2306.03809 [cs.CY] . · arXiv
- Stalnaker, Robert (1984). Inquiry . Cambridge, MA: MIT Press.
- Zou, Andy, Zifan Wang, J. Zico Kolter, and Matt Fredrikson (2023). Universal and transferable adversarial attacks on aligned language models . arXiv: 2307.15043 [cs.CL] . · arXiv
Cited by
Sorted by citations
Artificial intelligence risk management framework :
National Institute of Standards and Technology (US) (2024) · NIST Trustworthy and Responsible AI · doi:10.6028/nist.ai.600-1 · Cited by 274 · see citing works
AI agents under threat: A survey of key security challenges and future pathways
Deng, Z., Guo, Y., Han, C., Ma, W., Xiong, J., Wen, S., & Xiang, Y. (2025) · ACM Computing Surveys · doi:10.1145/3716628 · Cited by 210 · see citing works
The benefits and dangers of anthropomorphic conversational agents
Peter, S., Riemer, K., & West, J. D. (2025) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2415898122 · Cited by 110 · see citing works
Deception abilities emerged in large language models
Hagendorff, T. (2024) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2317967121 · Cited by 100 · see citing works
Artificial intelligence in smart drug delivery systems: a step toward personalized medicine
Panchpuri, M., Painuli, R., & Kumar, C. (2025) · RSC Pharmaceutics · doi:10.1039/d5pm00089k · Cited by 65 · see citing works
People cannot distinguish GPT-4 from a human in a Turing test
Jones, C. R., Rathi, I., Taylor, S., & Bergen, B. K. (2025) · Proceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3715275.3732108 · Cited by 76 · see citing works
“In order to enable other researchers to scrutinize our data, and to investigate other questions for which this data could be helpful, we sought approval to share full transcripts from the replication study (after any potentially sensitive or personally identifiable information had been removed).”
Recent Advancements in Artificial Intelligence in Battery Recycling
Jose, S. A., Cook, C. A. D., Palacios, J., Seo, H., Ramirez, C. E. T., Wu, J., & Menezes, P. L. (2024) · Batteries · doi:10.3390/batteries10120440 · Cited by 55 · see citing works
Understanding artificial agency
Dung, L. (2025) · The Philosophical Quarterly · doi:10.1093/pq/pqae010 · Cited by 69 · see citing works
Dahlgren Lindström, A., Methnani, L., Krause, L., Ericson, P., de Rituerto de Troya, Í. M., Coelho Mollo, D., & Dobbe, R. (2025) · Ethics and Information Technology · doi:10.1007/s10676-025-09837-2 · Cited by 54 · see citing works
“Indeed, there is emerging concern that, when presented with ethically complex questions, LLMs tend to simply mirror the users’ views (see, e.g. Turpin et al. 2023, Park et al. 2023, or the sycophancy benchmarking tasks of Perez et al. 2023).”
Chandra, M., Naik, S., Ford, D., Okoli, E., De Choudhury, M., Ershadi, M., Ramos, G., Hernandez, J., Bhattacharjee, A., Warreth, S., et al. (2025) · Proceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3715275.3732063 · Cited by 44 · see citing works
Shah, A., Ranka, P., Dedhia, U., Prasad, S., Muni, S., & Bhowmick, K. (2023) · International Journal of Advanced Computer Science and Applications · doi:10.14569/ijacsa.2023.01410110 · Cited by 47 · see citing works
How to design an AI ethics board
Schuett, J., Reuel, A.-K., & Carlier, A. (2025) · AI and Ethics · doi:10.1007/s43681-023-00409-y · Cited by 46 · see citing works
“For example, systems might become capable of deceiving their operators and only “pretending” to be safe in a testing environment [116, 132].”
The Landscape of Emerging AI Agent Architectures for Reasoning, Planning, and Tool Calling: A Survey
Masterman, T., Besen, S., Sawtell, M., & Chao, A. (2024) · arXiv · doi:10.48550/arxiv.2404.11584 · Cited by 35 · 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
“Both have already been seen to some extent in the wild (Park et al. 2024).”
Deception and manipulation in generative AI
Tarsney, C. J. (2025) · Philosophical Studies · doi:10.1007/s11098-024-02259-8 · Cited by 30 · see citing works
“…to do this, possessing at least a basic conceptual understanding of deception (Hagen-dorff, 2024) and of phenomena like implicature and ambiguity that might be used to mislead (Park et al., 2024; Kamath et al., 2024), as well as a large stock of general knowledge to identify misleading omissions.”
InfluentialFostering effective hybrid human-LLM reasoning and decision making
Passerini, A., Gema, A. P., Minervini, P., Sayin, B., & Tentori, K. (2025) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2024.1464690 · Cited by 35 · see citing works
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
Neuro-symbolic AI for auditable cognitive information extraction from medical reports
Prenosil, G., Weitzel, T., Bello, S. C., Mingels, C., Manzini, G., Meier, L. P., Shi, K., Rominger, A., & Afshar‐Oromieh, A. (2025) · Communications Medicine · doi:10.1038/s43856-025-01194-x · Cited by 28 · see citing works
“The rationale for this approach is that each component offsets the weaknesses of the others: In our design, GPT-4 harvests clinical facts from free text 25 , while a locally hosted expert system (Plato-3) veri fi es these extractions against medical rules, generating deterministic, trustworthy…”
Bond, A., Retief, F., Alberts, R. C., Roos, C., Cilliers, D., & Moolman, J. (2024) · Impact Assessment and Project Appraisal · doi:10.1080/14615517.2024.2426117 · Cited by 25 · see citing works
“Park et al. (2024) identify clear evidence of deception arising from AI, often stemming from asking particular questions of Large Language Models, and concerns over unintended consequences of the use of AI in general have already led several distinguished scientists to argue for a greater societal…”
Humanoid Robots in Tourism and Hospitality—Exploring Managerial, Ethical, and Societal Challenges
Skubis, I., Mesjasz-Lech, A., & Nowakowska-Grunt, J. (2024) · Applied Sciences · doi:10.3390/app142411823 · Cited by 22 · see citing works
“The data, software, network, and hardware components of robots are the most vulnerable elements to threats and cyberattacks [83–85].”
Standards for Belief Representations in LLMs
Herrmann, D. A., & Levinstein, B. A. (2024) · Minds and Machines · doi:10.1007/s11023-024-09709-6 · Cited by 58 · see citing works
“7 In addition to checking that explanations are faithful to internal processes, belief measurement also provides a strategy to detect deception, which can be important for designing safe and cooperative AI systems (Dafoe et al., 2020; Evans et al., 2021; Park et al., 2024).”
Next Frontiers of Aviation Safety: System-of-Systems Safety
Li, D., Yao, A. L., Feng, K., Zhou, H., Wang, R., Cheng, M., Li, H., Wang, D., & Ding, S. (2025) · Engineering · doi:10.1016/j.eng.2025.01.002 · Cited by 23 · see citing works
Red teaming large language models: A comprehensive review and critical analysis
Jabbar, M. S., Al-Azani, S., Alotaibi, A., & Ahmed, M. (2025) · Information Processing & Management · doi:10.1016/j.ipm.2025.104239 · Cited by 25 · see citing works
Ranisch, R., & Haltaufderheide, J. (2025) · Digital Society · doi:10.1007/s44206-025-00161-2 · Cited by 17 · see citing works
“…that unintentional, deceptive abilities can emerge in LLMs (Hagendorff, 2024a ), ranging from “strategically” inducing false beliefs in users to exhibiting “sycophantic” behavior, where the conversational agents state what the user wants to hear, regard - less of accuracy (Park et al., 2024).”
Ng, P. M. L., Wan, C., Lee, D., Garnelo‐Gomez, I., & Lau, M. M. (2025) · Internet Research · doi:10.1108/intr-11-2024-1783 · Cited by 24 · see citing works
Kashyap, R. K., Samuel, Y., Friedman, L. W., & Samuel, J. (2024) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2024.1443386 · Cited by 20 · see citing works
Security of LLM-based agents regarding attacks, defenses, and applications: A comprehensive survey
Tang, Y., Liu, Y., Lan, J., Yan, Z., & Erol Gelenbe, A. (2025) · Information Fusion · doi:10.1016/j.inffus.2025.103941 · Cited by 22 · see citing works
Sociotechnical Implications of Generative Artificial Intelligence for Information Access
Mitra, B., Cramer, H. S. M., & Gurevich, O. (2024) · The information retrieval series · doi:10.1007/978-3-031-73147-1_7 · Cited by 20 · see citing works
AI Rules? Characterizing Reddit Community Policies Towards AI-Generated Content
Lloyd, T., Gosciak, J., Nguyen, T., & Naaman, M. (2024) · International Conference on Human Factors in Computing Systems · doi:10.1145/3706598.3713292 · Cited by 27 · see citing works
Giray, L., Jacob, J., Encanto, V., & Mansilungan, C. J. (2025) · Journal of Academic Ethics · doi:10.1007/s10805-025-09695-z · Cited by 20 · see citing works
“● Using AI to rewrite or paraphrase plagiarized text to evade detection (Park et al., 2024).”
From threat to trust: assessing security risks of agentic AI systems
Leo, M., Tan, F., Miao, T., & Anand, G. (2026) · International Journal of Information Security · doi:10.1007/s10207-025-01185-y · Cited by 20 · see citing works
“Unlike traditional software, RL failures often appear suddenly and unpredictably as the system discovers unintended ways to achieve its goals [76].”
The Geometry of Concepts: Sparse Autoencoder Feature Structure
Li, Y., Michaud, E. J., Baek, D. D., Engels, J., Sun, X., & Tegmark, M. (2025) · Entropy · doi:10.3390/e27040344 · Cited by 14 · see citing works
“This is the goal of the present paper, focusing on three separate spatial scales.”
Terra Incognita: The Governance of Artificial Intelligence in Global Perspective
Stanger, A. K., Kraus, J., Lim, W., Millman-Perlah, G., & Schroeder, M. (2024) · Annual Review of Political Science · doi:10.1146/annurev-polisci-041322-042247 · Cited by 12 · see citing works
Haupt, M. R., Yang, L., Purnat, T. D., & Mackey, T. K. (2024) · JMIR Infodemiology · doi:10.2196/60678 · Cited by 17 · see citing works
“…accounts that deceive users by mimicking the language patterns of targeted identities (refer to the studies by De Angelis et al [11], Park et al [57], and Hajli et al [58] for detailed discussions on the risks posed by chatbots and AI systems for manipulative tactics, such as fraud and…”
BackgroundThe Longtail Impact of Generative AI on Disinformation: Harmonizing Dichotomous Perspectives
Lucas, J. S., Maung, B. M., Tabar, M., McBride, K., & Lee, D. (2024) · IEEE Intelligent Systems · doi:10.1109/mis.2024.3439109 · Cited by 14 · see citing works
“4,5 This amplifies disinformation risks, threatening cyber-security, business intelligence, public health, and democratic processes as GenAI integrates across society.”
BackgroundLarge Language Models can Strategically Deceive their Users when Put Under Pressure
Scheurer, J., Balesni, M., & Hobbhahn, M. (2023) · arXiv · doi:10.48550/arxiv.2311.07590 · Cited by 11 · see citing works
“Large Language Models (LMs) are now competent at lying (Pacchiardi et al., 2023) and strategic deception when instructed or trained to be deceptive (Park et al., 2023; Hagendorff, 2023; O’Gara, 2023; Hubinger et al., 2024), as well as strategically deceiving third parties even when not instructed…”
Barman, K. G., Caron, S., Sullivan, E., de Regt, H. D., de Austri, R. D., Boon, M., Färber, M., Fröse, S., Golling, T., Lopez, L. G., et al. (2025) · The European Physical Journal C · doi:10.1140/epjc/s10052-025-14707-8 · Cited by 35 · see citing works
“These risks are compounded by the opacity in methodologies in complex fields, requiring the development of ethical guidelines and oversight mechanisms [50, 51, 52, 123].”
A comprehensive introspection on AI risks: taxonomy, challenges, and future directions
Ray, P. P. (2026) · Iran Journal of Computer Science · doi:10.1007/s42044-025-00370-5 · Cited by 13 · see citing works
“Park et al. survey the phenomenon of AI deception, documenting examples from specialized and general-purpose systems, delineating risks from fraud to loss of control, and proposing legal, regulatory, and technical countermeasures [54].”
Large language models pass a standard three-party Turing test
Jones, C. R., & Bergen, B. K. (2026) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2524472123 · Cited by 133 · see citing works
“Models with this ability to robustly deceive (and masquerade as) people could be used for social engineering or to spread misinformation (18–20).”
The EU AI Act: A pioneering effort to regulate frontier AI?
Graells, G. B., Salinas, C., Tinoco, R., & Sevilla, J. (2024) · INTELIGENCIA ARTIFICIAL · doi:10.4114/intartif.vol27iss73pp55-64 · Cited by 11 · see citing works
“This includes threats such as the enhancement of cyber-offenses [13] [14], the manipulation of humans [15] [16], or the development of biological and chemical weapons [17] [18].”
BackgroundManaging the unknown in machine learning: Definitions, related areas, recent advances, and prospects
Barcina-Blanco, M., Lobo, J. L., Bringas, P. G., & Ser, J. D. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.128073 · Cited by 11 · see citing works
Bridging the Protection Gap: Innovative Approaches to Shield Older Adults from AI-Enhanced Scams
Herrera, L., Sickle, L. V., & Podhradsky, A. L. (2024) · Cyber Awareness and Research Symposium (CARS) · doi:10.1109/cars61786.2024.10778759 · Cited by 12 · see citing works
“Though there is no universal recommendation of what should be done to defend against AI-enhanced scams, Brundage, Avin, Clark, et al. [22] and Park, Goldstein, O’Gara, et al. [27] recommend prevention and mitigation through responsible development and informed policy-making and legislation.”
Risks, Causes, and Mitigations of Widespread Deployments of Large Language Models (LLMs): A Survey
Sakib, M. N., Islam, M. A., Pathak, R., & Arifin, M. M. (2024) · International Conference on Artificial Intelligence, Blockchain, and Internet of Things (AIBThings) · doi:10.1109/aibthings63359.2024.10863356 · Cited by 13 · see citing works
Multi-Agent Risks from Advanced AI
Hammond, L., Chan, A., Clifton, J., Hoelscher-Obermaier, J., Khan, A., McLean, E., Smith, C., Barfuss, W., Foerster, J., Gavenčiak, T., et al. (2025) · arXiv · doi:10.48550/arxiv.2502.14143 · Cited by 10 · see citing works
Artificial intelligence and the future of otherness: what kind of other can an AI be for a human?
Fernandez‐Borsot, G. (2025) · AI & Society · doi:10.1007/s00146-025-02281-2 · Cited by 11 · see citing works
Provably safe systems: the only path to controllable AGI
Tegmark, M., & Omohundro, S. (2023) · arXiv · doi:10.48550/arxiv.2309.01933 · Cited by 10 · see citing works
Лазаренко, Н. І., & Hapchuk, Y. (2024) · Педевтологія · doi:10.31652/3041-1203-2024(1)-7-13 · Cited by 9 · see citing works
Talking terms: Agent information in LLM supply chain bargaining
Kirshner, S. N., Pan, Y., Wu, J. X., & Gould, A. (2025) · Decision Sciences · doi:10.1111/deci.70010 · Cited by 9 · see citing works
Technosolutionism and the empathetic medical chatbot
Sharon, T. (2025) · AI & Society · doi:10.1007/s00146-025-02441-4 · Cited by 11 · see citing works
The shutdown problem: an AI engineering puzzle for decision theorists
Thornley, E. (2024) · Philosophical Studies · doi:10.1007/s11098-024-02153-3 · Cited by 34 · see citing works
“And these agents have spontaneously misled humans: in one instance, an agent lied about having a visual impairment to a human that it enlisted to help solve a CAPTCHA (OpenAI, 2023b, 55–56; see also Park et al., 2023).”
Perils and Opportunities in Using Large Language Models in Psychological ResearchPreprint
Abdurahman, S., Atari, M., Karimi-Malekabadi, F., Xue, M. J., Trager, J. P., Park, P. S., Golazizian, P., Omrani, A., & Dehghani, M. (2023) · doi:10.31219/osf.io/tg79n · Cited by 8 · see citing works
Self-citationData Analytics with Large Language Models (LLM): A Novel Prompting Framework
Almheiri, S. M. A. A., Al-ansari, M., AlHashmi, J., Abdalmajeed, N., Jalil, M., & Ertek, G. (2024) · Lecture notes in operations research · doi:10.1007/978-3-031-61589-4_20 · Cited by 8 · see citing works
'There Has To Be a Lot That We're Missing': Moderating AI-Generated Content on Reddit
Lloyd, T., Reagle, J., & Naaman, M. (2025) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3757445 · Cited by 24 · see citing works
Perils and Opportunities in Using Large Language Models in Psychological ResearchPreprint
Abdurahman, S., Atari, M., Karimi-Malekabadi, F., Xue, M. J., Trager, J. P., Park, P. S., Golazizian, P., Omrani, A., & Dehghani, M. (2023) · doi:10.31234/osf.io/d695y · Cited by 8 · see citing works
Self-citationBaşer, M. Y., Kozak, M., & Erdoğan, İ. H. (2025) · The Internet and Higher Education · doi:10.1016/j.iheduc.2025.101071 · Cited by 11 · see citing works
Understanding Knowledge Drift in LLMs Through Misinformation
Fastowski, A., & Kasneci, G. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-82346-6_5 · Cited by 10 · see citing works
What Artificial Intelligence Means to Us – Students’ Insights
Pisica, A. I., & ZAHARIA, R. M. (2025) · European Journal of Interdisciplinary Studies · doi:10.24818/ejis.2025.09 · Cited by 8 · see citing works
Bozkurt, A. (2025) · Open Praxis · doi:10.55982/openpraxis.17.3.794 · Cited by 7 · see citing works
Arghode, V., & Nafukho, F. M. (2025) · European journal of training and development · doi:10.1108/ejtd-04-2025-0072 · Cited by 6 · see citing works
Davidescu, A., Agafiței, M.-D., Gheorghe, M., & Strat, V. (2025) · Mathematics · doi:10.3390/math13193075 · Cited by 4 · see citing works
Genç, H. U., Chandrasegaran, S. K., Dingler, T., & Verma, H. (2025) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3706598.3713579 · Cited by 7 · 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
“In other words, while The Safety Conception may capture something close to the manifest concept of AI safety among AI safety researchers, it does not capture the concept of AI safety that is operative among AI safety researchers or AI safety governance organizations.”
Hooshiar, M. H. (2026) · The Journal of Prosthetic Dentistry · doi:10.1016/j.prosdent.2025.11.005 · Cited by 5 · see citing works
The ‘Tragic Dilemma’ of GenAI in Education: An Exploratory Human-AI Conversation
Palalas, A., Pegrum, M. A., Gunawardena, C. N., Robertson, R. S., & Ventura, J. (2025) · Postdigital Science and Education · doi:10.1007/s42438-025-00600-w · Cited by 5 · see citing works
“Research is now revealing cases of GenAI pursuing non-truth-oriented goals through ‘learned deception’ of humans (Park et al. 2024)—an issue recently identified by Google researchers as their chief concern about GenAI (Shah et al. 2025)—along with cases of GenAI’s problematic emerging value systems…”
Zou, M., & Yang, Y. (2026) · Journal of Innovation & Knowledge · doi:10.1016/j.jik.2026.100971 · Cited by 5 · see citing works
On Controllability in Agentic AI: A Survey
Nguyen, M. H., Nguyen, D.-H., O’Sullivan, B., & Nguyen, H. D. (2026) · Minds and Machines · doi:10.1007/s11023-026-09783-y · Cited by 6 · see citing works
Episodic Memory in AI Agents Poses Risks that Should be Studied and Mitigated
DeChant, C. (2025) · 2025 IEEE Conference on Secure and Trustworthy Machine Learning (SaTML) · doi:10.1109/satml64287.2025.00024 · Cited by 5 · see citing works
“AI agents have been found by many researchers to engage in various kinds of deception [57].”
Emancipatory Information Retrieval
Mitra, B. (2025) · Information Retrieval Research · doi:10.54195/irrj.24531 · Cited by 5 · see citing works
Zou, M.-Y., & Yang, Y. (2026) · Technological forecasting & social change · doi:10.1016/j.techfore.2025.124506 · Cited by 5 · see citing works
Jones, C., & Bergen, B. K. (2026) · Artificial Intelligence Review · doi:10.1007/s10462-026-11517-6 · Cited by 17 · see citing works
The state as a model for AI control and alignment
Elsner, M. (2024) · AI & Society · doi:10.1007/s00146-024-02063-2 · Cited by 4 · see citing works
“Park et al. (2024) survey a variety of other instances in which AI models practice deception, either in games or in simulated negotiations.”
Boudry, M., & Friederich, S. (2024) · Philosophical Studies · doi:10.1007/s11098-024-02226-3 · Cited by 4 · see citing works
Fırıncı, Y. (2024) · Marmara Üniversitesi İlahiyat Fakültesi Dergisi · doi:10.15370/maruifd.1565884 · Cited by 5 · see citing works
Navigating Ethical Dilemmas in AI-Powered Translation
Asrifan, A. (2025) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-0060-3.ch012 · Cited by 4 · see citing works
Training Users Against Human and GPT-4 Generated Social Engineering Attacks
Malloy, T., Ferreira, M. J., Fang, F., & González, C. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-92833-8_4 · Cited by 4 · see citing works
Affirmative Safety: An Approach to Risk Management for Advanced AIPreprint
Wasil, A. R., Clymer, J., Krueger, D., et al. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4806274 · Cited by 4 · see citing works
Is AI a subject that can live together with humans?
Kwon, Y. W. (2025) · AI & Society · doi:10.1007/s00146-025-02257-2 · Cited by 3 · see citing works
Duffy, A., Bennett, S., Yardley, L., & Moreno, S. (2025) · EAI Endorsed Transactions on Pervasive Health and Technology · doi:10.4108/eetpht.11.9416 · Cited by 4 · see citing works
“AI and machine learning is an emerging field that can be manipulated [60] to ‘lie with statistics’.”
A Novel AI System for Trustworthy and Accurate Diagnostics and Troubleshooting
Li, M. Z. (2025) · IEEE Integrated STEM Education Conference (ISEC) · doi:10.1109/isec64801.2025.11147314 · Cited by 3 · see citing works
Improving online anti-phishing training using cognitive large language models
Malloy, T., Fang, F., & González, C. (2025) · Computers in Human Behavior · doi:10.1016/j.chb.2025.108760 · Cited by 5 · see citing works
Regulatory sandboxes for AI in the majority world: A learning-centric approach to legal adaptation
Español, A. G., & Koenig, P. D. (2025) · Cambridge Forum on AI Law and Governance · doi:10.1017/cfl.2025.10015 · Cited by 4 · see citing works
Sharma, T., Potter, Y., Park, J., Liu, Y., Huang, Y., Liu, X., Song, D., Hancock, J. T., & Wang, Y. (2026) · Scientific Reports · doi:10.1038/s41598-026-40180-8 · Cited by 4 · see citing works
“This inclination could help explain why dialogues between users and AI typically exhibit fewer instances of conflicting viewpoints compared to interactions between human users 50 .”
Avrupa Birliği Yapay Zekâ Tüzüğü'ne Göre Yasaklanmış Yapay Zekâ Uygulamaları
Akbulut, B., & AÇIL, M. (2026) · Ankara Hacı Bayram Veli Üniversitesi Hukuk Fakültesi Dergisi · doi:10.34246/ahbvuhfd.1704398 · Cited by 3 · see citing works
Neural Reasoning for Sure Through Constructing Explainable Models
Dong, T., Jamnik, M., & Lió, P. (2025) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v39i11.33262 · Cited by 3 · see citing works
Bown, O., Grace, K. S., Ocampo, R., & Ibarrola, F. (2025) · Creativity Research Journal · doi:10.1080/10400419.2025.2501735 · Cited by 3 · see citing works
“Parallel to risks such as user deception in existing AI systems (Park et al., 2024), creative AI can also pose a greater risk if involving creative solutions to misaligned problems that have the potential to spiral out of control (as in Bostrom’s Paperclip Maximiser parable).”
Benchmarking AI: Toward Inclusive Evaluation of Language Models
Jovanovic, M., & Campbell, M. (2025) · Computer · doi:10.1109/mc.2025.3575880 · Cited by 3 · see citing works
Security, Risk Management, and Ethical AI in the Future of DeFi
Eswaran, U., Eswaran, V., Eswaran, V., et al. (2024) · Advances in finance, accounting, and economics book series · doi:10.4018/979-8-3693-6321-8.ch003 · Cited by 3 · see citing works
Aversion to external feedback suffices to ensure agent alignment
Garcia, P. (2024) · Scientific Reports · doi:10.1038/s41598-024-72072-0 · Cited by 3 · see citing works
Estep, P. W. (2024) · AI & Society · doi:10.1007/s00146-024-02134-4 · Cited by 3 · see citing works
Gutiérrez, E. J. D., & Jarquín-Ramírez, M.-R. (2025) · Edutec Revista Electrónica de Tecnología Educativa · doi:10.21556/edutec.2025.91.3665 · Cited by 3 · see citing works
The AI Manipulation GapPreprint
Boine, C. C. (2022) · SSRN Electronic Journal · doi:10.2139/ssrn.4042321 · Cited by 3 · see citing works
Chat Bankman-Fried: an Exploration of LLM Alignment in FinancePreprint
Biancotti, C., Camassa, C., Giudice, O., Glielmo, A., & Coletta, A. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.5005794 · Cited by 3 · see citing works
Tagesson, A., & Stenseke, J. (2024) · Frontiers in Psychology · doi:10.3389/fpsyg.2024.1347890 · Cited by 2 · see citing works
A Contrastive Semantic Watermarking Framework for Large Language Models
Wang, J., Chang, X., Xiao, C., & Zhang, L. (2025) · Symmetry · doi:10.3390/sym17071124 · Cited by 2 · see citing works
The Case for Using Generative AI to Run Deliberation Simulations
Rountree, J., & Gastil, J. (2025) · Journal of Deliberative Democracy · doi:10.16997/jdd.1625 · 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
Balancing truth and lies: ethical management of AI in service encounters
Snyder, H., Witell, L., van Riel, A. C. R., Magor, T., Lutze, S., & Bougoure, U. S. (2026) · Journal of service management · doi:10.1108/josm-12-2024-0538 · Cited by 2 · see citing works
Sharma, A. K., & Sharma, R. K. (2026) · Aslib Journal of Information Management · doi:10.1108/ajim-06-2025-0416 · Cited by 2 · see citing works
Understanding and responding to complex online harms: misinformation, fake news, and young adults
Zhang, V. N., Vlachokyriakos, V., Johnson, I. G., & Durrant, A. C. (2026) · Frontiers in Computer Science · doi:10.3389/fcomp.2026.1655186 · Cited by 2 · see citing works
Watson, S., & Ennion, M. (2026) · Learning Media and Technology · doi:10.1080/17439884.2026.2642199 · Cited by 4 · see citing works
- Authors
- Park, P. S., Goldstein, S., O'Gara, A., Chen, M., & Hendrycks, D.
- Affiliations
- Massachusetts Institute of Technology (US)
- Australian Catholic University (AU)
- Corresponding author
- Park, Peter S.
- Venue
- arXiv
- Publisher
- arXiv
- Published
- 2023
- Volume
- 5
- Issue
- 5
- Pages
- 100988-100988
- ISSN
- 2666-3899 (ISSN-L)
- arXiv
- 2308.14752
- PMC
- PMC11117051
- OpenAlex
- W4396796749
- Semantic Scholar
- afe455cfb03b9487c8f0b67993f42f834b633c23
- Cited by
- 243
- References
- 66
- Citations by year
- 2026: 111 · 2025: 89 · 2024: 38 · 2023: 4 · 2022: 1
- FWCI
- 18.12
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Patterns (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by-nc-nd)
- PubMed (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
- The HKU Scholars Hub (University of Hong Kong) (submittedVersion, cc-by-nc-nd)
- arXiv (Cornell University)
- Topics
- Social Sciences › Safety Research › Ethics and Social Impacts of AI
- Social Sciences › Sociology and Political Science › Misinformation and Its Impacts
- Social Sciences › Law › Digitalization, Law, and Regulation
- Keywords
- AI deception, large language models, Cicero, fraud, artificial intelligence control, risk assessment, AI regulation, false belief
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
