THE IMPACT OF GENERATIVE ARTIFICIAL INTELLIGENCE ON COGNITIVE BIAS, CONSUMER TRUST AND ENGAGEMENT IN DIGITAL PLATFORM: EVIDENCE FROM THE UNITED KINGDOM

Abstract

The fast incorporation of Generative Artificial Intelligence (GenAI) into online applications has created serious implications and uncertainties for consumers regarding how GenAI affects their thoughts and actions. This study utilizes cognitive bias theory and trust-based models of consumer engagement to evaluate the impact of using GenAI on forming cognitive biases, and how cognitive biases affect consumer trust and engagement with GenAI-enabled digital platforms (GenAI-EDPs). Survey data (N=386) from GenAI users conducted through Prolific Academic were analyzed using structural equation modelling (SEM) to test the proposed model. The study's findings indicate a significant increase in cognitive biases when consumers use GenAI (β = 0.57, p < 0.001). Cognitive biases have an adverse effect on consumer confidence in GenAI (β = -0.61, p <0.001), and consumer confidence, in turn, has a positive influence on consumer engagement with GenAI (β = 0.65, p < 0.001) and serves to mediate the relationship between cognitive bias and consumer engagement (Indirect effect β = -0.40, 95% CI [ -0.48, -0.31]). In addition, consumer perceptions of the level of transparency associated with the use of GenAI systems have an impact on the relationship between cognitive bias and confidence (β = 0.18, p = 0.001). The higher the level of perceived transparency, the less negative impact cognitive bias had on consumer confidence. Based on the conclusion of this study, the results indicate that GenAI improves a user’s interaction but also supports the same cognitive biases that may negatively impact a user's trust and engagement with the platform if they are not addressed. Trust and transparency are essential components to the successful and responsible long-term use of Generative Artificial Intelligence by digital platforms.