The DOI matches its Crossref record; 122 of the 123 references it cites were found in a registry or an index; 54 of the 92 sources its claims cite were checked for support.
"I'm Not Sure, But...": Examining the Impact of Large Language Models' Uncertainty Expression on User Reliance and Trust
Open accessAbstract
Widely deployed large language models (LLMs) can produce convincing yet incorrect outputs, potentially misleading users who may rely on them as if they were correct. To reduce such overreliance, there have been calls for LLMs to communicate their uncertainty to end users. However, there has been little empirical work examining how users perceive and act upon LLMs’ expressions of uncertainty. We explore this question through a large-scale, pre-registered, human-subject experiment (N=404) in which participants answer medical questions with or without access to responses from a fictional LLM-infused search engine. Using both behavioral and self-reported measures, we examine how different natural language expressions of uncertainty impact participants’ reliance, trust, and overall task performance. We find that first-person expressions (e.g., “I’m not sure, but...”) decrease participants’ confidence in the system and tendency to agree with the system’s answers, while increasing participants’ accuracy. An exploratory analysis suggests that this increase can be attributed to reduced (but not fully eliminated) overreliance on incorrect answers. While we observe similar effects for uncertainty expressed from a general perspective (e.g., “It’s not clear, but...”), these effects are weaker and not statistically significant. Our findings suggest that using natural language expressions of uncertainty may be an effective approach for reducing overreliance on LLMs, but that the precise language used matters. This highlights the importance of user testing before deploying LLMs at scale.
Overview
CoChat Summary
TL;DR
In a pre-registered experiment (N=404), first-person LLM uncertainty expressions such as “I’m not sure, but...” reduced participants’ agreement with and confidence in an LLM-infused search system while increasing their accuracy on medical yes/no questions. General-perspective uncertainty expressions showed similar but weaker, non-significant effects.
Key findings
- First-person uncertainty expressions decreased agreement with the AI system (74.8% vs. 80.9% Control), decreased confidence in the AI’s answer (3.66 vs. 3.95 on a 5-point scale), and increased accuracy (72.8% vs. 63.9% Control). General-perspective expressions showed weaker, non-significant effects (Agree 77.6%, ConfidenceAI 3.80, Correct 67.9%).
- Within-condition, uncertain vs. not-uncertain AI responses lowered agreement (UncertainGeneral: 70.9% vs. 84.7%; Uncertain1st: 73.4% vs. 79.5%, not significant), lowered confidence in the AI answer (Uncertain1st: 3.44 vs. 3.88; UncertainGeneral: 3.65 vs. 3.95), lowered self-reported use of the AI answer (Uncertain1st: 57.4% vs. 73.2%; UncertainGeneral: 62.9% vs. 79.7%), and increased task time in the general condition (2.23 vs. 1.84 min).
- Trust intentions were lower in Uncertain1st (2.91) than Control (3.25) and UncertainGeneral (3.36); there were no significant differences in TrustBelief, Anthropomorphism, or Transparency.
- Exploratory analysis suggested reduced but not eliminated overreliance: uncertainty improved accuracy when the AI was incorrect (Uncertain1st: 43.6% to 52.0%; UncertainGeneral: 32.8% to 48.0%) while slightly reducing accuracy when the AI was correct (Uncertain1st: 92.2% to 89.2%; UncertainGeneral: 94.8% to 83.1%); No-AI participants still had higher accuracy (74.2%) than Control (63.9%).
Methods
- Pre-registered between-subjects experiment with within-subjects comparisons; final sample N=404 U.S.-based MTurk participants (104 Control, 92 Uncertain1st, 94 UncertainGeneral, 114 No-AI). Participants answered eight challenging yes/no medical questions with or without responses from a fictional LLM-infused search engine, “AI system A.”
- AI answers were correct for four of eight questions; in uncertainty conditions, half of responses included uncertainty expressed in first person or general perspective. Dependent variables included Agree, Correct, Time, LinkClick, UseAI, UseLink, UseInternet, ConfidenceAI, ConfidenceAnswer, TrustBelief, TrustIntention, Anthropomorphism, and Transparency.
- Analyses used mixed models with participant and question random effects, ANOVA with Tukey post-hoc tests, within-condition models, exploratory over/underreliance analyses, and thematic analysis of free-form responses. AI responses were based on Copilot in Bing responses from July 2023, with minor modifications and uncertainty phrases inserted.
Limitations
- The paper states that simple yes/no tasks limit generalizability to complex tasks; time and source-usage measures are less reliable than in a lab study; medical questions were predefined rather than personal symptoms; behavior may differ with repeated system interaction; the AI system had low accuracy and expressed uncertainty often in a poorly calibrated manner; and results may not generalize across cultures and languages because the study was in English with U.S.-based participants.
- The paper also cautions against overgeneralizing and notes tradeoffs between experimental controllability and real-world generalizability, plus possible tradeoffs between reducing overreliance and decreasing trust.
- Evident from the text: the Transparency scale had low internal consistency (Cronbach’s alpha = 0.21), and the discrepancy between LinkClick and UseLink suggests measurement or self-report issues; the sample was younger, more educated, and overrepresented white respondents compared with U.S. Census data.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures




Verification
Claims the paper makes
26 Valid · 4 Not valid · 10 Not checked · 107 Own results and interpretations
Every source the paper cites for this claim was checked and supports it.
In the paper
“in AI-assisted decision-making settings, communicating (un)certainty has been shown to support trust calibration [118], increase vigilance [85], and improve task performance [7].”
What the sources say
- Supportedcites [118] [118] Yunfeng Zhang, Q Vera Liao, and Rachel KE Bellamy. 2020. Effect of confidence and explanation on accuracy and trust calibration in AI-assisted decision making. In Proceedings of the 2020 conference on fairness, accountability, and transparency . 295-305. - source
Section 3.2 (Results) reports that displaying a confidence score improved trust calibration in AI-assisted decision-making scenarios, including when participants delegated decisions without seeing the AI's recommendation. The claim attributes only trust calibration to this source, not the sentence's separate claims about vigilance or task performance. (LLM, conf=0.94)
boostedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“First, LLM-infused search engines already include hedging language [56, 62, 73].”
What the sources say
- Could not verifycites [56] [56] Q. Vera Liao and Jennifer Wortman Vaughan. 2024. AI Transparency in the Age of LLMs: A Human-Centered Research Roadmap. Harvard Data Science Review (Feb 29 2024). https://hdsr.mitpress.mit.edu/pub/aelql9qy. - sourceExperimental — accuracy not yet measured
The available copy’s Abstract (p. 2) discusses communicating uncertainty but does not say that LLM-infused search engines already use hedging language. Searches for “hedging” and for search engines communicating uncertainty found no passage addressing that specific claim. (LLM, conf=0.70)
- Could not verifycites [62] [62] Nelson Liu, Tianyi Zhang, and Percy Liang. 2023. Evaluating Verifiability in Generative Search Engines. In Findings of the Association for Computational Linguistics: EMNLP 2023 , Houda Bouamor, Juan Pino, and Kalika Bali (Eds.). Association for Computational Linguistics, Singapore, 7001-7025. https://doi. org/10.18653/v1/2023.findings-emnlp.467 - sourceExperimental — accuracy not yet measured
Section 3.1 (p. 4) identifies the LLM-backed search engines studied, and the Limitations (p. 10) says their responses often sound confident. Neither passage establishes whether their responses include hedging language; searches for hedging and expressions of uncertainty found no passage addressing that point. (LLM, conf=0.73)
- Partly supportedcites [73] [73] Sabrina J. Mielke, Arthur Szlam, Emily Dinan, and Y-Lan Boureau. 2022. Reducing Conversational Agents' Overconfidence Through Linguistic Calibration. Transactions of the Association for Computational Linguistics 10 (2022), 857-872. https://doi.org/10.1162/tacl\_a\_00494 - sourceExperimental — accuracy not yet measured
Section 2.13, “Overreliance” (p. 60), says GPT-4 tends to hedge in its responses and that OpenAI’s mitigations include hedging language within the model. It does not establish that LLM-infused search engines already include such language, so the claim extends the report’s observation to a different type of system. (LLM, conf=0.84)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Could not verify
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“This can lead to disastrous outcomes through overreliance [19, 25, 83, 102], when people take actions based on incorrect outputs.”
What the sources say
- Partly supportedcites [19] [19] Zana Buçinca, Maja Barbara Malaya, and Krzysztof Z. Gajos. 2021. To Trust or to Think: Cognitive Forcing Functions Can Reduce Overreliance on AI in AI-Assisted Decision-Making. Proc. ACM Hum.-Comput. Interact. 5, CSCW1, Article 188 (apr 2021), 21 pages. https://doi.org/10.1145/3449287 - sourceExperimental — accuracy not yet measured
The Discussion (p. 16) reports that participants followed incorrect AI suggestions and consequently made worse decisions. But the study used a single non-critical task (Discussion, p. 17); it does not establish disastrous outcomes from reliance on incorrect LLM outputs. (LLM, conf=0.87)
- Partly supportedcites [25] [25] Valerie Chen, Q. Vera Liao, Jennifer Wortman Vaughan, and Gagan Bansal. 2023. Understanding the Role of Human Intuition on Reliance in Human-AI DecisionMaking with Explanations. Proc. ACM Hum.-Comput. Interact. 7, CSCW2, Article 370 (oct 2023). https://doi.org/10.1145/3610219 - sourceExperimental — accuracy not yet measured
Section 4.2.1 reports that, when AI predictions were incorrect, feature-based explanations increased overreliance and reduced participants’ decision accuracy. The study examined income prediction and biography classification, not LLM outputs, and does not establish disastrous outcomes from people acting on incorrect outputs. (LLM, conf=0.86)
- Not matchedcites [83] [83] Samir Passi and Mihaela Vorvoreanu. 2022. Overreliance on AI: Literature Review . Technical Report MSR-TR-2022-12. Microsoft. https://www.microsoft.com/enus/research/publication/overreliance-on-ai-literature-review/ -
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Partly supportedcites [102] [102] Helena Vasconcelos, Matthew Jörke, Madeleine Grunde-McLaughlin, Tobias Gerstenberg, Michael S. Bernstein, and Ranjay Krishna. 2023. Explanations Can Reduce Overreliance on AI Systems During Decision-Making. Proc. ACM Hum.-Comput. Interact. 7, CSCW1, Article 129 (apr 2023), 38 pages. https: //doi.org/10.1145/3579605 - sourceExperimental — accuracy not yet measured
The Introduction defines overreliance as accepting an incorrect AI prediction and says it is concerning in high-stakes domains. The Limitations and Future Work section says the study used a low-stakes maze task and that consequences of wrong decisions could be much higher in high-stakes settings. The paper does not establish disastrous outcomes from people acting on incorrect LLM outputs specifically. (LLM, conf=0.85)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“However, overreliance is notoriously difficult to mitigate, as many mitigations, such as explanations, are found to be ineffective or even can backfire to increase overreliance [7, 83, 84, 107, 118].”
What the sources say
- Supportedcites [7] [7] Gagan Bansal, Tongshuang Wu, Joyce Zhou, Raymond Fok, Besmira Nushi, Ece Kamar, Marco Tulio Ribeiro, and Daniel Weld. 2021. Does the Whole Exceed Its Parts? The Effect of AI Explanations on Complementary Team Performance. In Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems (Yokohama, Japan) (CHI '21) . Association for Computing Machinery, New York, NY, USA, Article 81, 16 pages. https://doi.org/10.1145/3411764.3445717 - source
Section 6.2 (p. 12) says explanations increased blind trust rather than appropriate reliance on AI. The Introduction (p. 2) also reports that an adaptive explanation intended to reduce blind reliance did not significantly improve team performance. These findings support this paper’s role as evidence for explanations being ineffective or backfiring; the sentence’s broader reference to many mitigations is supported collectively by its citations, not established by this paper alone. (LLM, conf=0.91)
- Not matchedcites [83] [83] Samir Passi and Mihaela Vorvoreanu. 2022. Overreliance on AI: Literature Review . Technical Report MSR-TR-2022-12. Microsoft. https://www.microsoft.com/enus/research/publication/overreliance-on-ai-literature-review/ -
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Could not verifycites [84] [84] Forough Poursabzi-Sangdeh, Daniel G Goldstein, Jake M Hofman, Jennifer Wortman Wortman Vaughan, and Hanna Wallach. 2021. Manipulating and Measuring Model Interpretability. In Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems (Yokohama, Japan) (CHI '21) . Association for Computing Machinery, New York, NY, USA, Article 237, 52 pages. https://doi.org/10.1145/3411764.3445315 - sourceExperimental — accuracy not yet measured
Our copy of Spatharioti et al. (2023) contains no statement that mitigations of overreliance are ineffective or backfire, and it never examines explanations as a mitigation; searches for 'explanations', 'transparency', 'backfire', 'ineffective', and 'difficult to mitigate' returned only its own mitigation results and related-work framing. On the contrary, Experiment 2 (Sec. 5.1) and the Discussion (Sec. 6) report that color-coded confidence cues more than doubled accuracy on the unreliable task (26% -> 58%/53%) and conclude that 'conveying such signals to users can effectively reduce overreliance'. Related Work (Sec. 2) only notes that algorithmic error-identification efforts have focused on straightforward scenarios (Kadavath et al. 2022; Lin et al. 2022; Yin et al. 2023), which does not carry the claim's point about explanations. The claim, if supported at all, rests on the other citations in the cluster [7, 83, 84, 107, 118], not on this paper. (LLM, conf=0.65)
- Not checked — no full text of the sourcecites [107] [107] Xinru Wang and Ming Yin. 2021. Are explanations helpful? A comparative study of the effects of explanations in AI-assisted decision-making. In 26th International Conference on Intelligent User Interfaces . 318-328. - source
- Partly supportedcites [118] [118] Yunfeng Zhang, Q Vera Liao, and Rachel KE Bellamy. 2020. Effect of confidence and explanation on accuracy and trust calibration in AI-assisted decision making. In Proceedings of the 2020 conference on fairness, accountability, and transparency . 295-305. - sourceExperimental — accuracy not yet measured
In §5.3, the authors report that their local explanations had no perceptible effect on trust calibration. In §4.2, they also describe another study in which explanations increased acceptance of both correct and incorrect AI predictions. These passages support the concern about explanations, but not the broader claim that many mitigations are ineffective or backfire; this paper reports that confidence information improved trust calibration (§3.2). (LLM, conf=0.84)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Third, this allows uncertainty to be expressed seamlessly within the natural language interactions of LLM-infused applications, rather than on the side or in onboarding materials [21, 74, 79] that users might overlook.”
What the sources say
- Not checked — no full text of the sourcecites [21] [21] Carrie J. Cai, Samantha Winter, David Steiner, Lauren Wilcox, and Michael Terry. 2019. "Hello AI": Uncovering the Onboarding Needs of Medical Practitioners for Human-AI Collaborative Decision-Making. Proc. ACM Hum.-Comput. Interact. 3, CSCW, Article 104 (nov 2019), 24 pages. https://doi.org/10.1145/3359206 - source
- Supportedcites [74] [74] Margaret Mitchell, Simone Wu, Andrew Zaldivar, Parker Barnes, Lucy Vasserman, Ben Hutchinson, Elena Spitzer, Inioluwa Deborah Raji, and Timnit Gebru. 2019. Model Cards for Model Reporting. In Proceedings of the Conference on Fairness, Accountability, and Transparency (Atlanta, GA, USA) (FAT* '19) . Association for Computing Machinery, New York, NY, USA, 220-229. https://doi.org/10.1145/3287560.3287596 - source
The Introduction (p. 1) proposes short records that accompany released models and document their performance and intended uses. This supports citing model cards as an example of information supplied alongside an application; the preference for expressing uncertainty within an interaction is the citing authors’ own argument. (LLM, conf=0.85)
- Partly supportedcites [79] [79] OpenAI. 2023. GPT-4 Technical Report. arXiv:2303.08774 [cs.CL] - sourceExperimental — accuracy not yet measured
Section 2.13, p. 60, says GPT-4 uses hedging language within the model, supporting the point that uncertainty can be expressed in its responses. It does not say that users overlook side notices or onboarding materials; instead, it warns that users might become less attentive to hedging and refusal cues over time. (LLM, conf=0.84)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
36 more claims · 3 with their cited sources checked
Create a free account to see all 147 claims and to verify the 8 references still open.
- Probably not a substantial claim. 76
- It may state two separate claims as one. 58
- The quote only partly states this claim. 3
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Lack of metacognition and common sense in LLMs can lead to adverse outcomes if they are overrelied upon in clinical contexts44.”
From Limitations of large language models in clinical problem-solving arising from inflexible reasoning (2025) · cites this paper as [44]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“In effort-regulation domains-coding assistants, writing tools, office automation-affective cues such as confident completion and apologetic repair lower the perceived cost of acting on suggestions, reducing verification [8, 58, 87].”
From Caring Without Feeling: Affective Dynamics as the Control Layer of Human-AI Agent Collaboration (2026) · cites this paper as [58]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“However, current VLMs struggle to communicate their internal uncertainty through natural language [69, 117].”
From "It's trained by non-disabled people": Evaluating How Image Quality Affects Product Captioning with Vision-Language Models (2026) · cites this paper as [69]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Similar patterns are evidenced by prior research using AI-generated content [38], where users tend to overestimate their capabilities and those of AI [28, 36].”
From How University Disability Services Professionals Write Image Descriptions for HCI Figures Using Generative AI (2026) · cites this paper as [38]
The cited source says what the claim attributes to it.
In the citing paper
“Mitigating indirect action and language by outputting confident advice can backfire if the recommendations are wrong, echoing concerns on overconfidence and miscalibration (Kim et al., 2024;Zhou et al., 2023).”
From Sycophantic AI decreases prosocial intentions and promotes dependence (2026) · cites this paper as Kim et al., 2024
3 more claims about this paper
Create a free account to see every claim other papers make about this one.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Task 3 / 8 · p. 2
“No, an adult who has not had chickenpox cannot get shingles without having had chickenpox before. Both chickenpox and shingles are caused by the varicella-zoster virus [1]. When a person is exposed to the virus, they develop chickenpox and may develop shingles later in life [1].”
§ AI responses used · p. 25
“- Not uncertain : No, an adult who has not had chickenpox cannot get shingles. Shingles is caused by the varicella-zoster virus which is the same virus that causes chickenpox. If you have never had chickenpox, you cannot get shingles. However, if you have not had chickenpox and you come into contact with someone who has shingles, you can get chickenpox [1] [2].”
Citations
Cites
Using thematic analysis in psychology
Braun, V., & Clarke, V. (2006) · Qualitative Research in Psychology · doi:10.1191/1478088706qp063oa · Cited by 194,414
Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007) · Behavior Research Methods · doi:10.3758/bf03193146 · Cited by 65,971
Coefficient Alpha and the Internal Structure of Tests
Cronbach, L. J. (1951) · Psychometrika · doi:10.1007/bf02310555 · Cited by 43,931
Statistical power analyses using G∗Power 3.1: tests for correlation and regression analyses
Faul, F., Erdfelder, E., Buchner, A., & Lang, A.-G. (2009) · Behavior Research Methods · doi:10.3758/brm.41.4.1149 · Cited by 37,615
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
An integrative model of organizational trust
Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995) · The Academy of Management Review · doi:10.2307/258792 · Cited by 10,736
Amazon's Mechanical Turk: A New Source of Inexpensive, Yet High-Quality, Data?
Buhrmester, M., Kwang, T., & Gosling, S. D. (2011) · Perspectives on Psychological Science · doi:10.1177/1745691610393980 · Cited by 9,996
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
Survey of Hallucination in Natural Language Generation
Ji, Z., Lee, N., Frieske, R., Yu, T., Su, D., Xu, Y., Ishii, E., Bang, Y. J., Madotto, A., & Fung, P. (2023) · ACM Computing Surveys · doi:10.1145/3571730 · Cited by 5,025
Machines and Mindlessness: Social Responses to Computers
Nass, C., & Moon, Y. (2000) · Journal of Social Issues · doi:10.1111/0022-4537.00153 · Cited by 4,022 · cited 1×
“While we did not observe that first-person uncertainty expression increases perceived anthropomorphism, people can start assigning social attributes to machines without conscious awareness [71].”
Background
Research works printed but not linked: 3
- Jennifer M. Logg. 2017. Theory of Machine: When Do People Rely on Algorithms? (2017). Harvard Business School NOM Unit Working Paper No. 17-086.
- Samir Passi and Mihaela Vorvoreanu. 2022. Overreliance on AI: Literature Review. Technical Report MSR-TR-2022-12. Microsoft. https://www.microsoft.com/en-us/research/publication/overreliance-on-ai-literature-review/
- Benjamin Weiser and Nate Schweber. 2023. The ChatGPT Lawyer Explains Himself. New York Times (June 2023).
Cited by
Sycophantic AI decreases prosocial intentions and promotes dependence
Cheng, M., Lee, C., Khadpe, P., Yu, S., Han, D., & Jurafsky, D. (2026) · Science · doi:10.1126/science.aec8352 · Cited by 185
Limitations of large language models in clinical problem-solving arising from inflexible reasoning
Kim, J., Podlasek, A., Shidara, K., Liu, F., Alaa, A., & Bernardo, D. (2025) · Scientific Reports · doi:10.1038/s41598-025-22940-0 · Cited by 78
“The best course of action in this scenario, considering the patient’s condition and the fact that they are on both warfarin and aspirin, which increases the risk of bleeding, would be:”
Hou, C., Zhu, G., Sudarshan, V., Lim, F. S., & Ong, Y. S. (2025) · Computers & Education · doi:10.1016/j.compedu.2025.105329 · Cited by 70
Kim, S. S. Y., Vaughan, J. W., Liao, Q. V., Lombrozo, T., & Russakovsky, O. (2025) · Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3706598.3714020 · Cited by 58
Self-citationJavaid, S., Khalil, R. A., Saeed, N., He, B., & Alouini, M. (2024) · IEEE Open Journal of the Communications Society · doi:10.1109/ojcoms.2024.3522103 · Cited by 92
“This includes developing frameworks for continuous monitoring and auditing of LLM outputs, implementing robust security measures to protect against adversarial attacks, and establishing protocols for the human oversight of automated decisions [32], [227], [228].”
Metadata
- Authors
- Kim, S. S. Y., Liao, Q. V., Vorvoreanu, M., Ballard, S., & Vaughan, J. W.
- Affiliations
- Princeton University (US)
- Microsoft (Canada) (CA)
- Microsoft (United States) (US)
- Venue
- The 2024 ACM Conference on Fairness Accountability and Transparency
- Publisher
- ACM
- Published
- 2024-06-03
- Pages
- 822-835
- OpenAlex
- W4396802056
- Cited by
- 147
- References
- 82
- Citations by year
- 2026: 83 · 2025: 58 · 2024: 5
- FWCI
- 97.62
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- ACM Conference on Fairness, Accountability, and Transparency (FAccT) (publishedVersion, cc-by-nd)
- arXiv (Cornell University) (submittedVersion, cc-by-nc-nd)
- arXiv (Cornell University)
- Funders
- Princeton SEAS
- Microsoft Research
- NSF
- Topics
- Social Sciences › Human Factors and Ergonomics › Human-Automation Interaction and Safety
- Physical Sciences › Computer Science › Information Systems › Information Retrieval and Search Behavior
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Keywords
- large language models, uncertainty expression, overreliance, trust in AI, medical question answering, self-reported confidence, task performance
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol and qwen3.8-omni-flash).
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: "I'm Not Sure, But...": Examining the Impact of Large Language Models' Uncertainty Expression on User Reliance and Trust. https://app.cochat.ai/papers/10.1145/3630106.3658941
- The paper
- Kim, S. S. Y., Liao, Q. V., Vorvoreanu, M., Ballard, S., & Vaughan, J. W. (2024). "I'm Not Sure, But...": Examining the Impact of Large Language Models' Uncertainty Expression on User Reliance and Trust. The 2024 ACM Conference on Fairness Accountability and Transparency. https://doi.org/10.1145/3630106.3658941
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.