LLMs for Financial Advisement: A Fairness and Efficacy Study in Personal Decision Making
Open accessAbstract
As Large Language Model (LLM) based chatbots are becoming more accessible, users are relying on these chatbots for reliable and personalized recommendations in diverse domains, ranging from code generation to financial advisement. In this context, we set out to investigate how such systems perform in the personal finance domain, where financial inclusion has been an overarching stated aim of banks for decades. We test widely used LLM-based chatbots, ChatGPT and Bard, and compare their performance against SafeFinance, a rule-based chatbot built using the Rasa platform. The comparison is across two critical tasks: product discovery and multi-product interaction, where product refers to banking products like Credit Cards, Certificate of Deposits, and Checking Accounts. With this study, we provide interesting insights into the chatbots’ efficacy in financial advisement and their ability to provide fair treatment across different user groups. We find that both Bard and ChatGPT can make errors in retrieving basic online information, the responses they generate are inconsistent across different user groups, and they cannot be relied on for reasoning involving banking products. On the other hand, despite their limited generalization capabilities, rule-based chatbots like SafeFinance provide safe and reliable answers to users that can be traced back to their original source. Overall, although the outputs of the LLM-based chatbots are fluent and plausible, there are still critical gaps in providing consistent and reliable financial information.
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.
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
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 · cited 1× · see citing works
“In [3], the authors argue that while these models have pushed the boundaries of what is possible through architectural innovations and sheer size, there are potential risks associated with their use.”
BackgroundA Survey of Large Language Models
Zhao, W. X., Zhou, K., Li, J., Tang, T., Wang, X., Hou, Y., Min, Y., Zhang, B., Zhang, J., Dong, Z., et al. (2023) · arXiv · doi:10.48550/arxiv.2303.18223 · Cited by 2,090
Sparks of Artificial General Intelligence: Early experiments with GPT-4
Bubeck, S., Chandrasekaran, V., Eldan, R., Gehrke, J., Horvitz, E., Kamar, E., Lee, P., Lee, Y. T., Li, Y., Lundberg, S., et al. (2023) · arXiv · doi:10.48550/arxiv.2303.12712 · Cited by 1,526 · see citing works
Recent Advances in Natural Language Processing via Large Pre-Trained Language Models: A Survey
Min, B., Ross, H., Sulem, E., Veyseh, A. P. B., Nguyen, T. H., Sainz, O., Agirre, E., Heinz, I., & Roth, D. (2021) · arXiv · doi:10.48550/arxiv.2111.01243 · Cited by 1,387 · see citing works
Recent Advances in Natural Language Processing via Large Pre-trained Language Models: A Survey
Min, B., Ross, H., Sulem, E., Veyseh, A. P. B., Nguyen, T. H., Sainz, O., Agirre, E., Heintz, I., & Roth, D. (2024) · ACM Computing Surveys · doi:10.1145/3605943 · Cited by 1,387 · see citing works
Torous, J., Bucci, S., Bell, I. H., Kessing, L. V., Faurholt‐Jepsen, M., Whelan, P., Carvalho, A. F., Keshavan, M., Linardon, J., & Firth, J. (2021) · World Psychiatry · doi:10.1002/wps.20883 · Cited by 1,199 · cited 1× · see citing works
“Lastly, [24] provides a comprehensive update on the overall field of digital psychiatry, including using chatbots.”
BackgroundBlut, M., Wang, C., Wünderlich, N. V., & Brock, C. (2021) · Journal of the Academy of Marketing Science · doi:10.1007/s11747-020-00762-y · Cited by 1,194 · see citing works
Topic Aware Neural Response Generation
Chen, X., Wu, W., Wu, Y., Liu, J., Huang, Y., Zhou, M., & Ma, W. (2017) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v31i1.10981 · Cited by 448 · see citing works
Rasa: Open Source Language Understanding and Dialogue ManagementPreprint
Bocklisch, T., Faulkner, J., Pawlowski, N., & Nichol, A. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1712.05181 · Cited by 442 · see citing works
Mathematical Capabilities of ChatGPT
Frieder, S., Pinchetti, L., Chevalier, A., Griffiths, R.-R., Salvatori, T., Lukasiewicz, T., Petersen, P. C., & Berner, J. (2023) · arXiv · doi:10.48550/arxiv.2301.13867 · Cited by 385 · see citing works
Examining Gender and Race Bias in Two Hundred Sentiment Analysis Systems
Kiritchenko, S., & Mohammad, S. (2018) · Proceedings of the Seventh Joint Conference on Lexical and Computational Semantics · doi:10.18653/v1/s18-2005 · Cited by 384 · see citing works
BloombergGPT: A Large Language Model for Finance
Wu, S., Irsoy, O., Lu, S., Dabravolski, V., Dredze, M., Gehrmann, S., Kambadur, P., Rosenberg, D., & Mann, G. (2023) · arXiv · Cited by 357 · see citing works
Bang, Y., Cahyawijaya, S., Lee, N., Dai, W., Su, D., Wilie, B., Lovenia, H., Ji, Z., Yu, T., Chung, W., et al. (2023) · arXiv · doi:10.48550/arxiv.2302.04023 · Cited by 337 · see citing works
Lucia, C. D., Pazienza, P., & Bartlett, M. (2020) · Sustainability · doi:10.3390/su12135317 · Cited by 314 · see citing works
Language models for financial news recommendation
Lavrenko, V. P., Schmill, M., Lawrie, D., Ogilvie, P., Jensen, D. D., & Allan, J. (2000) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/354756.354845 · Cited by 141 · see citing works
Is ChatGPT the Ultimate Programming Assistant -- How far is it?
Tian, H., Lu, W., Li, T. O., Tang, X., Cheung, S.-C., Klein, J., & Bissyandé, T. F. (2023) · arXiv · doi:10.48550/arxiv.2304.11938 · Cited by 114 · see citing works
Powering an AI Chatbot with Expert Sourcing to Support Credible Health Information Access
Xiao, Z., Liao, Q. V., Zhou, M., Grandison, T., & Li, Y. (2023) · Proceedings of the 28th International Conference on Intelligent User Interfaces · doi:10.1145/3581641.3584031 · Cited by 59 · see citing works
A Systematic Investigation of Commonsense Knowledge in Large Language Models
Li, X. L., Kuncoro, A., Hoffmann, J., d’Autume, C. D. M., Blunsom, P., & Nematzadeh, A. (2022) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2022.emnlp-main.812 · Cited by 59 · see citing works
Protein Language Models and Structure Prediction: Connection and ProgressionPreprint
Hu, B.-Z., Xia, J.-X., Zheng, J.-B., et al. (2022) · arXiv.org · doi:10.48550/arxiv.2211.16742 · Cited by 50 · see citing works
Plansformer: Generating Symbolic Plans using TransformersPreprint
Pallagani, V., Muppasani, B. C., Murugesan, K., Rossi, F., Lior Horesh, C., Srivastava, B., Fabiano, F., & Loreggia, A. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2212.08681 · Cited by 11 · see citing works
Self-citationLakkaraju, K., Vuruma, S. K. R., Pallagani, V., et al. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2307.07422 · Cited by 5 · see citing works
Self-citationOn Safe and Usable Chatbots for Promoting Voter ParticipationPreprint
Muppasani, B. C., Pallagani, V., Lakkaraju, K., et al. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2212.11219 · Cited by 1 · see citing works
Self-citationOpenAI, Achiam, J., Adler, S., Agarwal, S., Ahmad, L., Akkaya, I., Aleman, F. L., Almeida, D., Altenschmidt, J., Altman, S., et al. (2023) · arXiv · doi:10.48550/arxiv.2303.08774
Flagged first, then by citations
Research works printed but not linked: 3
- Daniel Borrajo Alberto Pozanco, Kassiani Papasotiriou. 2022. PFPT: a Personal Finance Planning Tool by means of Heuristic Search and Automated Planning. https: //icaps22.icaps-conference.org/workshops/FinPlan/FinPlan22_paper_2.pdf
- Zhuang Liu, Degen Huang, Kaiyu Huang, Zhuang Li, and Jun Zhao. 2021. Finbert: A pre-trained financial language representation model for financial text mining. In Proceedings of the twenty-ninth international conference on international joint conferences on artificial intelligence. 4513-4519.
- Thomas Yue and Chi Chung Au. 2023. GPTQuant's Conversational AI: Simplify- ing Investment Research for All. Available at SSRN 4380516 (2023).
Cited by
Sorted by citations
A closer look at ChatGPT's role in concept map generation for education
Schicchi, D., Limongelli, C., Monteleone, V., & Taibi, D. (2026) · Interactive Learning Environments · doi:10.1080/10494820.2025.2497110 · Cited by 11 · see citing works
Large Language Model Agent in Financial Trading: A Survey
Ding, H., Li, Y., Wang, J., Chen, H., Guo, D., & Zhang, Y. (2026) · Proceedings of the 2026 9th International Conference on Computers in Management and Business · doi:10.1145/3802463.3802475 · Cited by 10 · see citing works
- Authors
- Lakkaraju, K., Jones, S. E., Vuruma, S. K. R., Pallagani, V., Muppasani, B. C., & Srivastava, B.
- Affiliations
- Artificial Intelligence Institute at UofSC, USA
- University of South Carolina (US)
- ORCID
- Venue
- 4th ACM International Conference on AI in Finance
- Publisher
- ACM
- Published
- 2023-11-25
- Pages
- 100-107
- OpenAlex
- W4388994240
- Cited by
- 42
- References
- 23
- Citations by year
- 2026: 9 · 2025: 22 · 2024: 10 · 2023: 1
- FWCI
- 7.46
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- ACM International Conference on AI in Finance (ICAIF) (publishedVersion, cc-by-nc)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › AI in Service Interactions
- Social Sciences › Safety Research › Ethics and Social Impacts of AI
- Physical Sciences › Computer Science › Artificial Intelligence › Speech and dialogue systems
- Keywords
- large language models, financial advice, chatbots, algorithmic fairness, personal finance, banking products, rule-based chatbots, ChatGPT, Google Bard, rasa, financial inclusion, product discovery, personal decision-making
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
