From Sandbox to Laboratory: Refractive and Ethical Human / AI Knowledge Co-Production Through Game-ful Research Practices

Abstract

The use of artificial intelligence in knowledge production processes and in academia has been illuminated from different disciplinary perspectives. The introduction of digital tools and research software in non-computational disciplines, the emergence of fields like the digital humanities, and the introduction of AI-driven tools for general application in the research process impact the types of engagement and interaction between researcher and tool. Starting from the notion of playing software, I introduce three metaphors for engagement with AI-driven research software: 1) software as laboratory, 2) research software as sandbox game, and 3) the scientific method as game rules. Based on a conceptual analysis of game studies and knowledge production literature, this theory adaptation of research as a game-ful process aims at creating awareness for the agencies within socio-technical epistemic systems by examining one's own practices in relation to (AI-driven) technology and increasing transparency by recognising intra-actions to account for the decisions made in knowledge production processes. By breaking down the components of games they can be mapped onto research processes and the ways digital tools are integrated. Approaching software-supported research as a game-ful process allows for engaging with AI-driven software as a research tool as a more entangled and co-constructive process.