Understanding Human Trust in AI Systems - Evaluation and Analysis

Learn about the complex dynamics of human-AI trust relationships in this 79-minute lecture that explores application-grounded evaluations and task selection in AI systems. Delve into the fundamental concepts of interpersonal trust and how they translate to human-AI interactions. Examine the critical aspects of AI trustworthiness, including what specific tasks and decisions humans are entrusting to AI systems. Explore the distinction between general trust and warranted trust in AI, while understanding the motivations behind building trustworthy AI systems. Gain insights into how to evaluate and measure trust in AI applications through practical examples and theoretical frameworks.