Cognitive Risks of AI: Literacy, Trust, and Critical Thinking
Abhinandan Kulal
Abstract
This study investigates how the use of artificial intelligence (AI) tools affects critical thinking and examines the moderating roles of trust calibration and AI literacy in a quantitative survey of 625 postgraduate students, teachers, and research scholars. Drawing on Cognitive Offloading Theory, we hypothesized that greater AI usage would undermine critical thinking, and that trust in AI would exacerbate this effect. Consistent with our predictions, results from moderated moderation analysis indicate that AI usage negatively predicts critical thinking, trust in AI amplifies this negative relationship, and high AI literacy significantly buffers the detrimental impact of trust. These findings demonstrate that fostering AI literacy is a key strategic approach for calibrating users’ trust and preserving critical engagement with AI-generated information. The study underscores the necessity of integrating AI literacy initiatives into educational curricula to mitigate cognitive risks posed by widespread AI adoption.