AI at Work: A Hybrid Study of Artificial Intelligence and Machine Learning Research in Practice

The aim of the proposed studentship is to provide a bridge between, on the one hand, work in Artificial Intelligence (AI) and the Machine Learning (ML) techniques at its cutting edge and, on the other, the social sciences. AI and ML are at the forefront of computational innovations designed to respond to an increasingly digital world (and the digital data that represent it). The ubiquity and range of usages of digital data have ensured that AI and ML techniques have filtered into a multitude of sociologically-relevant domains including industry, healthcare, government, policymaking, and more. Though the social sciences have produced studies of topics relevant to AI and computational innovation, however, these typically focus on their 'high level', general or abstract aspects - for instance, Burrell (2016) on "opacity" in machine learning algorithms, Seaver (2017) on algorithms as a cultural form or Mittelstadt et al (2016) on the ethics of algorithms. In contrast, little attention has yet been paid to the "shop work" of producing and working with AI and ML algorithms (and the practical reasoning that informs those practices). As a consequence, the content of algorithms and algorithmic work remains underexplored. AI and ML thus present an important challenge to the social sciences, as a means of making sense of an increasingly commonplace though often poorly-understood phenomenon.