AUTHOR=Robayo-Pinzon Oscar , Rojas-Berrio Sandra , Camargo Jorge E. , Foxall Gordon R. 

TITLE=Generative artificial intelligence (GenAI) use and dependence: an approach from behavioral economics

JOURNAL=Frontiers in Public Health

VOLUME=Volume 13 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/public-health/articles/10.3389/fpubh.2025.1634121

DOI=10.3389/fpubh.2025.1634121

ISSN=2296-2565

ABSTRACT=ObjectiveThis study aims to explore the perceived dependence on Generative Artificial Intelligence (GenAI) tools among young adults and examine the relative reinforcing value of AI chatbots use compared to monetary rewards, applying a behavioral economics approach.Participants/methodsA total of 420 university students from Bogotá, Colombia, participated in an online survey. The study employed a Multiple Choice Procedure (MCP) to assess the relative reinforcement between different durations of GenAI use (1, 2, and 4 weeks) and monetary rewards, which varied in amount and delay. Additionally, an adapted AI Dependence Scale evaluated levels of dependence on AI tools. Data analysis included repeated measures ANOVA to examine the effects of reward magnitude and delay on choices, and correlations to assess the relationship between perceived dependence and reinforcement values.ResultsParticipants reported low average dependence on AI tools (mean AI Dependence Scale score = 65.6), with no significant gender differences. MCP findings indicated significant differences in crossover points across varying durations or delays for AI chatbots use, suggesting a higher relative value of use for the option to use AI chatbots immediately. The average reinforcement value for AI use versus monetary rewards did significantly vary with reward magnitude. On the other hand, significant differences were found in the levels of perceived dependence on AI, according to the average daily time of AI tool use.ConclusionThe results suggest that young adults exhibit low perceived dependence on GenAI tools but show differential reinforcement values based on usage duration or delay conditions. This behavioral economics approach provides novel insights into decision-making patterns related to AI chatbots use, emphasizing the need for further research to understand the psychological and social factors influencing dependence on AI technologies.