Individual time interval preferences and the impact of recommendation sources: comparing human andAI influence

Purpose This study aims to examine whether time interval preferences are stable and whether human or artificial intelligence (AI) recommendations more strongly influence related decision-making. Design/methodology/approach An online interaction experiment with 157 participants assigned to three groups: one received AI-based recommendations, one received human-generated ones and one served as a control. Participants completed a simulated learning task, choosing time intervals for training videos across two stages. Recommendations, based on participants’ initial preferences, were identical in logic across AI and human conditions. Chi-square tests assessed whether participants maintained their preferences and how the source of information influenced their decisions. Findings Individuals consistently showed stable time interval preferences and generally chose options that matched them. When humans provided recommendations, participants were more likely to adhere to their original preferences, whereas AI recommendations led to greater deviation, suggesting that AI may disrupt the alignment of preferences. Research limitations/implications The study advances the theory on time perception and AI trust, demonstrating how framing and decision confidence influence behaviour. Limitations include cultural factors, prior exposure to AI and a possibly tech-savvy sample. Future work should explore diverse contexts and cognitive mechanisms behind AI recommendation acceptance. Practical implications AI tools should reflect personal time preferences while supporting user autonomy. Presenting AI recommendations as human-like or diminishing their authority can support individual preferences in decision-making. Originality/value This study introduces the concept of individual time interval preferences and applies cognitive consistency theory to human–AI interaction. It broadens understanding of AI’s role in personal decision-making and contributes to research on cognitive processes in preference formation.