The Mediating Role of Academic Stress in ChatGPT Dependency and Executive Dysfunction with Special Reference to University Students

Abstract Background: The fast amalgamation of artificial intelligence (AI) tools like ChatGPT to our educational ecosystems has created a need to understand its cognitive implications on students. Aim and Objectives: The study examines the contribution of problematic ChatGPT use in global executive functioning deficits and whether academic stress mediates this relationship. Materials and Methods: A sample of 122 students (aged 18–27 years) from Indian universities was selected using the Problematic ChatGPT Use Scale, the Student Academic Stress Scale, and the Behavior Rating Inventory of Executive Function–Adult Version. Data analysis was conducted through SPSS version 16 and Hayes’ PROCESS macro (Model 4) for mediation analysis. Results: Regression analysis indicated that academic stress significantly predicted executive dysfunction (β = 0.706, P < 0.001), while ChatGPT use alone did not contribute significantly when stress was accounted for β = 0.069, P = 0.328. Further, mediation analysis confirmed that academic stress fully mediates the relationship between ChatGPT use and executive functioning. Increased ChatGPT dependency leads to higher academic stress (β = 0.406, P < 0.001), which in turn is strongly associated with reduced executive capabilities (β = 0.705, P < 0.001). Conclusions: The use of AI may appear to compensate for cognitive deficits; it potentially disrupts the natural development of core executive functions. The study warns of a cyclical relationship where academic stress worsens executive dysfunction, driving further AI dependence and potentially undermining cognitive growth.