A Qualitative Study on Cultural Hegemony and the Impacts of AI

Abstract

Understanding the future consequences of artificial intelligence requires a holistic consideration of its cultural dimensions, on par with its technological intricacies and potential applications. Individuals and institutions working closely with AI, and with considerable resources, have significant influence on how impact is considered, particularly with regard to how much attention is paid to epistemic concerns (including issues of bias in datasets or potential misinterpretations of data, for example) versus normative concerns (such as societal and ecological effects of AI in the medium- and long-term). In this paper we review qualitative studies conducted with AI researchers and developers to understand how they position themselves relative to each of these two dimensions of impact, and how geographies and conditions of work influence their positions. Our findings underscore the need to gather more perspectives from low- and middle-income countries, whose notions of impact extend beyond the immediate technical concerns or impacts in the short- to medium-term. Rather, they encapsulate a broader spectrum of impact considerations, including the deleterious effects perpetrated by global corporate entities, the unwarranted influence of wealthy nations, the encroachment of philanthrocapitalism, and the adverse consequences of excluding communities affected by these phenomena from active participation in discussions surrounding impact.