By analysing 1.13 million complaint narratives submitted to the US Consumer Financial Protection Bureau (CFPB), we show that large language model (LLM)-assisted complaints surged after the release of ChatGPT and increased the chances of consumers of receiving relief. Furthermore, consumers with unobserved disadvantages in self-advocacy were found to be more likely to adopt LLMs, which highlights a potential for LLM tools to act as an equalizer.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$32.99 / 30 days
cancel any time
Subscribe to this journal
Receive 12 digital issues and online access to articles
$119.00 per year
only $9.92 per issue
Buy this article
- Purchase on SpringerLink
- Instant access to the full article PDF.
USD 39.95
Prices may be subject to local taxes which are calculated during checkout
References
Petty, R. E. & Cacioppo, J. T. The elaboration likelihood model of persuasion. Adv. Exp. Soc. Psychol. 19, 123–205 (1986). This paper suggests that persuasion operates via central argument quality or peripheral cues such as style and credibility.
Lusardi, A. & Mitchell, O. S. The economic importance of financial literacy: theory and evidence. J. Econ. Lit. 52, 5–44 (2014). This paper documents how low financial literacy is widespread and economically consequential for household decisions.
Hermann, E., Williams, G. Y. & Puntoni, S. Deploying artificial intelligence in services to aid vulnerable consumers. J. Acad. Mark. Sci. 52, 1431–1451 (2024). This paper proposes a framework for using AI in services to detect, better serve and empower vulnerable consumers.
Dang, J. & Liu, L. Extended artificial intelligence aversion: people deny humanness to artificial intelligence users. J. Pers. Soc. Psychol. https://doi.org/10.1037/pspi0000480 (2024). This paper reports that people might dehumanize AI users, provide them with reduced support, and be less willing to help, a phenomenon that was termed ‘extended AI aversion’.
Salvi, F., Horta Ribeiro, M., Gallotti, R. & West, R. On the conversational persuasiveness of GPT-4. Nat. Hum. Behav. 9, 1645–1653 (2025). This paper demonstrates that GPT-4 can be highly persuasive in multiround debates in a laboratory experiment setting.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This is a summary of: Shin, M. et al. The adoption and efficacy of large language models in US consumer financial complaints. Nat. Hum. Behav. https://doi.org/10.1038/s41562-026-02409-4 (2026).
Rights and permissions
About this article
Cite this article
Large language models have the potential to level the playing field in consumer financial complaints. Nat Hum Behav (2026). https://doi.org/10.1038/s41562-026-02411-w
Published:
Version of record:
DOI: https://doi.org/10.1038/s41562-026-02411-w
