Answers are grounded in this paper and cite their sources.
Building on prior studies that have documented the dehumanization of the users of artificial intelligence (AI) technology, that is, the tendency for observers to attribute less humanness to individuals who use AI than to those who do not, the present research examines whether and how different AI use purposes (i.e., accuracy- vs. efficiency-driven use) influence this effect. We conducted four preregistered studies (total N = 1,270) spanning diverse professional domains. We observed that accuracy-driven AI use negatively predicted the dehumanization of AI users, whereas efficiency-driven AI use positively predicted such dehumanization (Study 1). Furthermore, accuracy-driven AI users elicited less dehumanization than their efficiency-driven counterparts did, a pattern that proved robust across both intuitive assessments of dehumanization (Study 2) and deliberative ratings of humanness traits (Studies 3 and 4). However, both accuracy- and efficiency-driven AI users were significantly dehumanized compared with non-AI users (Study 4), demonstrating that while purpose matters, AI user dehumanization persists. These findings extend the literature on the social cost of AI and offer novel insights into how human–AI interaction shapes interpersonal dynamics .
No takes yet. Share an insight, caveat, or question.
Zhang et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: