Mapping AI’s role in NSW governance: a socio-technical analysis of GenAI integration

This paper examines the integration of Generative Artificial Intelligence (GenAI) in New South Wales (NSW) Government processes through a socio-technical lens using Actor-Network Theory. Rather than viewing GenAI as a passive tool, the study conceptualizes these systems as active actants that reshape governance networks, redistribute authority, and reconfigure democratic accountability mechanisms. Through document analysis, actor-network mapping, and scenario analysis of potential breakdowns, the research reveals that while NSW has established comprehensive AI governance frameworks, significant gaps remain in addressing the unique challenges posed by GenAI systems. Historical analysis of algorithmic failures in Australian public administration, including Revenue NSW’s automated debt recovery system and the federal Robodebt scheme, demonstrates the consequences when technical systems undermine democratic principles. The paper proposes a regulatory sandbox approach to balance innovation with democratic safeguards, highlighting the need for governance frameworks that recognize GenAI’s role in reshaping political relationships. The findings contribute to scholarly debates by demonstrating the insufficiency of purely technical or ethical frameworks that do not address the political dimensions of AI integration in public governance.