Binge or Bounce? Understanding User Satisfaction and Trust in AI-driven Streaming Recommendations

The rapid expansion of AI in media platforms has transformed how audiences engage with content, particularly through personalized recommendation systems. This study examines the factors thar influence users perceptions of AI-driven steaming recommendations, extending the Technology Acceptance Model by incorporating User Satisfaction and User Trust as key post-adoption constructs. Using a quantitative research design approach, a structured questionnaire was distributed to 291 Portuguese streaming service users. The results of this survey were analyzed through Partial Least Squares Structural Equation Modeling (PLS-SEM). The results show that Perceived Ease of Use and Perceived Usefulness strongly predict user attitudes and intentions, while Actual System Use significantly enhances both satisfaction and trust. These findings highlight that successful recommendation systems depend not only on technical performance but also on users' emotional responses, particularly their trust in AI and their satisfaction with content suggestions. These findings reinforce the applicability of TAM in the context of AI-driven personalization and extend the model by integrating satisfaction and trust as crucial post-use constructs. The research contributes to both academic theory and managerial practice by offering a comprehensive understanding of user acceptance of AI-driven recommendation systems in Portugal.