Improving Human-AI Collaboration With Descriptions of AI Behavior
This addresses the problem of suboptimal human-AI collaboration for users in decision-making tasks, but it is incremental as it builds on existing mental model concepts.
The paper tackled the problem of people under- or over-relying on AI predictions in decision-making by proposing behavior descriptions to improve reliance, and found through user studies with 225 participants across three domains that this approach increased human-AI accuracy by helping identify AI failures and increasing reliance when the AI was more accurate.
People work with AI systems to improve their decision making, but often under- or over-rely on AI predictions and perform worse than they would have unassisted. To help people appropriately rely on AI aids, we propose showing them behavior descriptions, details of how AI systems perform on subgroups of instances. We tested the efficacy of behavior descriptions through user studies with 225 participants in three distinct domains: fake review detection, satellite image classification, and bird classification. We found that behavior descriptions can increase human-AI accuracy through two mechanisms: helping people identify AI failures and increasing people's reliance on the AI when it is more accurate. These findings highlight the importance of people's mental models in human-AI collaboration and show that informing people of high-level AI behaviors can significantly improve AI-assisted decision making.