Cutting together-apart entangled conceptualisations of (machine) learning and ethics: assemblage of a ‘bricolage-pentimento’ artefact through an ethico-onto-epistem-ological approach

Artificial intelligence (AI) degree programmes commonly include machine learning courses covering techniques shaped by specific notions of learning. The ethics of such technologies are covered as part of many AI programmes, although are less often integrated into core machine learning courses. For this research, I first explore how students studying machine learning conceptualise ‘learning’ in general, and any correlations with machine learning techniques. Secondly, I consider potential ethical implications of these conceptualisations of learning in the context of AI education and development of machine learning technologies. Interrogating this entanglement of notions of (machine) learning, educational practice and ethics—involving human, nonhuman, material and abstract (more-than-human) entities—is challenging if one engages honestly with its complexity. Furthermore, I argue that integration of critical discussion of ethics into AI education design and practice requires an experimental combination of approaches—built upon a methodology emphasising difference and complexity over sameness and simplicity.