Trusting AI: does uncertainty visualization affect decision-making?
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Introduction Decision-making based on AI can be challenging, especially when considering the uncertainty associated with AI predictions. Visualizing uncertainty in AI refers to techniques that use visual cues to represent the level of confidence or uncertainty in an AI model's outputs, such as predictions or decisions. This study aims to investigate the impact of visualizing uncertainty on decision-making and trust in AI. Methods We conducted a user study with 147 participants, utilizing static classic gaming scenarios as a proxy for human-AI collaboration in decision-making. The study measured changes in decisions, trust in AI, and decision-making confidence when uncertainty was visualized in a continuous format compared to a binary output of the AI model. Results Our findings indicate that visualizing uncertainty significantly enhances trust in AI for 58% of participants with negative attitudes toward AI. Additionally, 31% of these participants found uncertainty visualization to be useful. The size of the uncertainty visualization was identified as the method that had the most impact on participants' trust in AI and their confidence in their decisions. Furthermore, we observed a strong association between participants' gaming experience and changes in decision-making when uncertainty was visualized, as well as a strong link between trust in AI and individual attitudes toward AI. Discussion These results suggest that visualizing uncertainty can improve trust in AI, particularly among individuals with negative attitudes toward AI. The findings also have important implications for the design of human-AI decision-support systems, offering insights into how uncertainty can be visualized to enhance decision-making and user confidence.
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.
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 | 0.004 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| 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 it