A more-than-human approach to researching AI at work: Alternative narratives for human and AI co-workers

Abstract

Professional workers practice at the intersection of public narratives about artificial intelligence (AI), the AI industry, and regulatory frameworks. Yet, there is limited understanding of the interactions between workers, AI systems, and the publics they serve. To inform networked learning scholarship, there is a pressing need to study the knowledge that workers are developing as they learn to work with AI and the implications for networked learning within the workplace and higher education. We bring social and computing science perspectives alongside more-than-human sensitivities to explore how professional expertise, judgment, accountability, and control is being re-distributed between human workers and AI systems. By sketching the changes AI is provoking we highlight the fine-grained research and analysis necessary to ensure that AI design and deployment is critically informed by in-depth understandings of how people are actually engaging with algorithmic systems. We raise questions about what trust and confidence in new AI-infused work practices is needed (or possible). Attention is drawn to the complexities of AI-mediated work, which invites re-thinking ways to generate the evidence needed to inform networked work-learning practices. Highlighted throughout is the power of AI narratives and the importance of advancing alternative, more nuanced, narratives.