The Effect of Interpretable Artificial Intelligence on Repeated Managerial Decision-Making Under Uncertainty
Bibliographic record
Abstract
Many business decisions, such as managing investment instruments, selecting medical care, and managing various supply chains, are being made repetitively in uncertain environments. At times, decisions under uncertainty could be perceived as wrong ex-post even though they are really ex-ante optimal. Therefore, decision-makers tend to intuitively resist taking algorithmic advice under high uncertainty. While recent studies claim that interpretability potentially increases the adoption of and trust in sophisticated AI and ML applications, these claims have not been tested with humans who must make repeated managerial decisions under uncertainty. Our study fills this gap by empirically investigating the impact of various common interpretability formats on AI adoption and trust under uncertainty. Interestingly, we show that explaining the inner workings of a model, in fact, decreases AI adoption. This is probably due to what we call 'computational overconfidence' on the part of the users. Moreover, contrary to other findings in the literature, we present clear evidence that AI adoption is significantly higher in more-uncertain business environments. Surprisingly, although AI adoption was greater in such environments, the level of trust in AI was considerably lower for all the interpretability types tested in this study. On the other hand, frequently sharing the cumulative performance over time of both the users and the AI improves users’ trust in the model. On the other hand, it had no negative impact on the AI adoption rate.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.019 | 0.239 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.003 |
| Scholarly communication | 0.004 | 0.004 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.003 | 0.004 |
| Insufficient payload (model declined to judge) | 0.007 | 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 source (direct Gemma or distilled Codex), 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".