Using AI and Behavioral Finance to Cope with Limited Attention and Reduce Overdraft Fees
Bibliographic record
Abstract
We test how effective a human–algorithm interaction is at stopping users from overdrawing their bank accounts. We use a randomized field experiment and draw our sample from users of a large personal financial management platform operating in the United States and Canada. We find that sending as-needed reminders is effective in and of itself, and the impact is intensified by the human response to the structure of the message. More simple messages are more effective, and the framing of the simplified message makes a difference. Users with medium to high annual incomes and users with fair to good credit scores are most likely to respond positively. We find that the investigated artificial intelligence solution reduces information-gathering costs and has a positive effect but is not sufficient in all cases. Those with challenging financial situations may find it harder to act upon the warning. For our analysis, we employ parametric identifications and time-to-event semiparametric analysis. Our work contributes to the literature on financial technology as advisors, human–computer interaction, limited attention, behavioral finance, and experimental finance. This paper was accepted by Jean-Edouard Colliard, Special Issue on the Human-Algorithm Connection. Funding: O. Sade acknowledges financial support for this research from the Krueger Center and the Albertson-Waltuch Chair in Business Administration. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2022.00304 .
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".