Personalized chatbot trustworthiness ratings
Open accessAbstract
This article contains examples of offensive and abusive language as a necessary part of illustrating its research findings. In this article, we address a setting where a conversation agent, also known as a chatbot, cannot be modified and its training data cannot be accessed, and yet a neutral party wants to assess and communicate its trustworthiness to a user in a way that is tailored to the user's priorities over the various trust issues (such as bias, abusive language, information leakage, or inappropriate communication complexity). Such a rating can help users choose among alternative chatbots, developers test their systems, business leaders price their offerings, and regulators set policies. We describe a chatbot rating methodology that relies on separate rating modules for each trust issue, and on users' priority orderings among the relevant trust issues, to generate an aggregate personalized rating for the trustworthiness of a chatbot. The method is independent of the specific trust issues and is parametric to the aggregation procedure, thereby allowing for seamless generalization. We illustrate its general use, integrate it with a live chatbot, and evaluate it on four dialog datasets and representative user profiles, validated with a user survey.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
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?
After implementing fairness checks, the chatbot provides culturally sensitive and fair interactions.
“After implementing fairness checks, the chatbot provides culturally sensitive and fair interactions[302,303].”
From AI Fairness in Data Management and Analytics: A Review on Challenges, Methodologies and Applications (2023) · cites this paper as [303]
Cites
Maximizing the spread of influence through a social network
Kempe, D., Kleinberg, J. M., & Eva Tardos, D. (2003) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/956750.956769 · Cited by 7,541 · see citing works
ELIZA—a computer program for the study of natural language communication between man and machine
Weizenbaum, J. (1966) · Communications of the ACM · doi:10.1145/365153.365168 · Cited by 4,452 · see citing works
Semantics derived automatically from language corpora contain human-like biases
Caliskan, A., Bryson, J. J., & Narayanan, A. (2017) · Science · doi:10.1126/science.aal4230 · Cited by 2,887 · see citing works
ELIZA — a computer program for the study of natural language communication between man and machine
Weizenbaum, J. (1983) · Communications of the ACM · doi:10.1145/357980.357991 · Cited by 2,765 · see citing works
Automated Hate Speech Detection and the Problem of Offensive Language
Davidson, T., Warmsley, D., Macy, M., & Weber, I. (2017) · Proceedings of the International AAAI Conference on Web and Social Media · doi:10.1609/icwsm.v11i1.14955 · Cited by 2,610 · see citing works
Bellamy, R. K. E., Dey, K., Hind, M., Hoffman, S. C., Houde, S., Kannan, K., Lohia, P., Martino, J., Mehta, S., Mojsilovic, A., et al. (2018) · arXiv · doi:10.48550/arxiv.1810.01943 · Cited by 1,581 · see citing works
Vinyals, O., & Le, Q. (2015) · arXiv · doi:10.48550/arxiv.1506.05869 · Cited by 1,400 · see citing works
POMDP-Based Statistical Spoken Dialog Systems: A Review
Young, S. J., Gašić, M., Thomson, B., & Williams, J. D. (2013) · Proceedings of the IEEE · doi:10.1109/jproc.2012.2225812 · Cited by 898 · see citing works
Computer Power and Human Reason: From Judgment to Calculation
Agassi, J., & Wiezenbaum, J. (1976) · Technology and Culture · doi:10.2307/3103715 · Cited by 887 · see citing works
Computer Power and Human Reason: From Judgement to Calculation.
Crowther, J. F., & Weizenbaum, J. (1977) · Contemporary Sociology A Journal of Reviews · doi:10.2307/2062802 · Cited by 726 · see citing works
The Second Dialog State Tracking Challenge
Henderson, M., Thomson, B., & Williams, J. D. (2014) · Annual Meeting of the Special Interest Group on Discourse and Dialogue · doi:10.3115/v1/w14-4337 · Cited by 547 · see citing works
Semantically Equivalent Adversarial Rules for Debugging NLP models
Ribeiro, M. T., Singh, S., & Guestrin, C. (2018) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/p18-1079 · Cited by 435 · see citing works
Fairness testing: testing software for discrimination
Galhotra, S., Brun, Y., & Meliou, A. (2017) · Proceedings of the 2017 11th Joint Meeting on Foundations of Software Engineering · doi:10.1145/3106237.3106277 · Cited by 405 · see citing works
Applying deep learning to answer selection: A study and an open task
Feng, M.-W., Xiang, B., Glass, M. R., Wang, L., & Zhou, B. (2015) · Automatic Speech Recognition & Understanding · doi:10.1109/asru.2015.7404872 · Cited by 392 · see citing works
The Ubuntu Dialogue Corpus: A Large Dataset for Research in Unstructured Multi-Turn Dialogue Systems
Lowe, R., Pow, N., Serban, I. V., & Pineau, J. (2015) · Annual Meeting of the Special Interest Group on Discourse and Dialogue · doi:10.18653/v1/w15-4640 · Cited by 282 · see citing works
The challenge of crafting intelligible intelligence
Weld, D. S., & Bansal, G. (2019) · Communications of the ACM · doi:10.1145/3282486 · Cited by 253 · cited 1× · see citing works
“For example, trust in an AI model is often related (also) to the possibility to get an explanation for its output (see [11], [25], [34]), so explainability could be another trust issue to consider in the future.”
BackgroundAn Introduction to Vote-Counting Schemes
Levin, J., & Nalebuff, B. (1995) · The Journal of Economic Perspectives · doi:10.1257/jep.9.1.3 · Cited by 235 · see citing works
The Handbook of Computational Linguistics and Natural Language Processing
Clark, A., Fox, C., & Lappin, S. (2010) · Wiley eBooks · doi:10.1002/9781444324044 · Cited by 229 · see citing works
Wang, W., Chen, L., Thirunarayan, K., & Sheth, A. (2014) · ACM Conference on Computer Supported Cooperative Work (CSCW) · doi:10.1145/2531602.2531734 · Cited by 175 · see citing works
Why we should have seen that coming
Wolf, M. J., Miller, K., & Grodzinsky, F. S. (2017) · ACM SIGCAS Computers and Society · doi:10.1145/3144592.3144598 · Cited by 174 · see citing works
Ethical Challenges in Data-Driven Dialogue Systems
Henderson, P., Sinha, K., Gontier, N., Ke, N. R., Fried, G., Lowe, R., & Pineau, J. (2018) · Proceedings of the AAAI/ACM Conference on AI Ethics and Society · doi:10.1145/3278721.3278777 · Cited by 166 · see citing works
Automation, Algorithms, and Politics| Talking to Bots: Symbiotic Agency and the Case of Tay
Neff, G., & Nagy, P. (2016) · International journal of communication · Cited by 161 · see citing works
Explanation and trust: what to tell the user in security and AI?
Pieters, W. (2011) · Ethics and Information Technology · doi:10.1007/s10676-010-9253-3 · Cited by 153 · see citing works
(2018) · arXiv (Cornell University) · Cited by 113 · see citing works
Conversational Interfaces: Past and Present
McTear, M., Callejas, Z., & Griol, D. (2016) · Springer eBooks · doi:10.1007/978-3-319-32967-3_4 · Cited by 33 · see citing works
Computationally Detecting and Quantifying the Degree of Bias in Sentence-Level Text of News Stories
Hutto, C. J., Folds, D. J., & Appling, D. S. (2015) · Cited by 11 · see citing works
Minimizing the length of non-mixed initiative dialogs
Inouye, R. B. (2004) · doi:10.3115/1219079.1219081 · Cited by 10 · see citing works
A Measure for Dialog Complexity and its Application in Streamlining Service OperationsPreprint
Liao, Q. V., Srivastava, B., & Kapanipathi, P. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1708.04134 · Cited by 6 · see citing works
Self-citationA Train Status Assistant for Indian RailwaysPreprint
Mishra, H., Gaurav, R., & Srivastava, B. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1809.08509 · Cited by 2 · see citing works
Self-citation
Flagged first, then by citations
Cited by
Sorted by citations
AI Fairness in Data Management and Analytics: A Review on Challenges, Methodologies and Applications
Chen, P., Wu, L., & Wang, L. (2023) · Applied Sciences · doi:10.3390/app131810258 · Cited by 155 · see citing works
- Authors
- Srivastava, B., Rossi, F., Usmani, S., & Bernagozzi, M.
- Affiliations
- University of South Carolina (US)
- IBM (United States) (US)
- Venue
- IEEE Transactions on Technology and Society
- Published
- 2020-12
- Volume
- 1
- Issue
- 4
- Pages
- 184-192
- ISSN
- 2637-6415 (ISSN-L)
- OpenAlex
- W3114928178
- Cited by
- 40
- References
- 29
- Citations by year
- 2026: 2 · 2025: 10 · 2024: 6 · 2023: 10 · 2022: 6 · 2021: 6
- FWCI
- 7.50
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE Transactions on Technology and Society (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Social Sciences › Safety Research › Ethics and Social Impacts of AI
- Physical Sciences › Computer Science › Artificial Intelligence › Speech and dialogue systems
- Physical Sciences › Computer Science › Artificial Intelligence › AI in Service Interactions
- Keywords
- trustworthiness assessment, offensive language, abusive language, bias, information leakage, user priority, rating aggregation, conversation agents, user profiles
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
