AI has paved the way to an unprecedented pace of change in work processes across several industries. As individuals and organizations increasingly employ AI to improve performance, investigating the link between AI-enabled systems characteristics, AI reliance, and resulting performance benefits remains largely overlooked in IS research so far. The failure of individuals and organizations to leverage the actual benefits of AI points us toward the importance of studying this linkage. Although the body of human-AI collaboration research is growing, no generally accepted theory of task-technology fit, a concept that leads to enhanced utilization and performance, has emerged in the context of AI-enabled systems. Drawing on the widely tested task-technology-fit theory (TTF), we develop a task-AI fit framework for AI-specific tasks (knowledge exploitation and knowledge validation) and present our research methodology to empirically test the model in this ERF paper.
