Towards Human-Centered Artificial Intelligence: Harnessing AI to Improve Human Decision-Making

Language plays a central role in human activities, including policy applications. For instance, if AI systems can understand millions of medical notes produced every day, they can enhance medical diagnosis and health care in general. To enhance human decision-making with AI, human decision makers need to appropriately understand and use recommendations from AI. Chenhao Tan will present recent efforts in developing best practices for incorporating AI into human decision-making, using deceptive review detection as a testbed. The results demonstrate a tradeoff between human performance and human agency and show that explanations of machine predictions can moderate this tradeoff.

Presenter: Chenhao Tan, assistant professor, Department of Computer Science, and affiliate, Harris School of Public Policy, University of Chicago

Papers:

• On Human Predictions with Explanations and Predictions of Machine Learning Models: A Case Study on Deception Detection 
• “Why is ‘Chicago’ deceptive?” Towards Building Model-Driven Tutorials for Humans
• Understanding the Effect of Out-of-distribution Examples and Interactive Explanations on Human-AI Decision Making