The impact of the privacy paradox in artificial intelligence and digital transformation: a dynamic capability approach to managing deepfakes.

This study examines how artificial intelligence (AI) and digital transformation reshape consumer privacy concerns, focusing on the heightened risks posed by deepfake technology. Digital transformation refers to integrating digital technologies into all business areas, fundamentally changing how organizations operate and deliver customer value. It is driven by AI's ability to enable real-time decision-making, personalization, and operational efficiency, making it a cornerstone of modern business strategy. While AI enhances personalization, operational agility, and real-time decision-making, it also introduces complex privacy and security challenges. Deepfakes—AI-generated manipulations of media—intensify these risks, undermining consumer privacy protection and creating a threat to digital transformation. This research aims to understand the privacy paradox in the context of AI and deepfake exposure, explore the associated organizational challenges, and propose actionable strategies for mitigation. Through an experimental approach, we assess the impact of deepfake exposure on consumer privacy concerns. Our findings indicate that participants often cannot recognize a deepfake as fake. Moreover, genuine videos are frequently perceived as fake. Therefore, it is essential to identify AI practices to mitigate deepfake threats and safeguard businesses and consumers. We contribute to the literature by extending the concept of the privacy paradox within AI-driven environments and broadening the dynamic capabilities framework to include privacy as a core organizational competency. We propose an innovative Dynamic privacy strategy (DPS) framework to address these challenges, transforming privacy management as a strategic capability rather than a regulatory requirement. We also offer a practical model for organizations navigating the complexities of AI innovation and digital transformation.