The Soul of Work: Evaluation of Job Meaningfulness and Accountability in Human-AI Collaboration

Work is an important part of our lives - not only as a way to earn a living but as a crucial source for experiencing meaningfulness. The introduction of autonomous systems (or in the widest sense "artificial intelligence", AI) will fundamentally impact work practices. However, while most existing models of human-AI collaboration focus on performance goals, less is known about their potential influence on job satisfaction. In this paper, we present an online experiment in which we compared the perception of job meaningfulness and accountability in a human-AI collaboration across three interaction paradigms: Supervisory, Advisory, and Interactive. Our results showed that, unlike the common notion of supervisory control, people find their job more satisfying when they directly interact with the AI and are involved and remain accountable for action and decision-making. Introducing AI as a teammate in the interactive paradigm was associated with the highest job meaningfulness.