Multimedia Privacy and Security Landscape in the Wake of AI/ML

Privacy and security of multimedia content came under scrutiny with the wider application of Artificial Intelligence (AI) and Machine Learning (ML)-based technologies. The issues range from AI/ML-based profiling, creating faceswaps, deepfakes, verifying the authenticity of multimedia content to secure delivery of the content and data-driven network and service management of multimedia applications in next-generation networks. Moreover, advanced AI/ML-based techniques pose a challenge to multimedia content publishers and providers, who are concerned with the unauthorized distribution of their content over the Internet. This chapter provides a comprehensive review of AI/ML-inspired threat landscape and open challenges for multimedia content, service delivery and management. Furthermore, AI/ML-inspired countermeasures for these threats and future directions for embedding AI/ML-based data-driven management approaches for multimedia service delivery are also discussed in this chapter. The main application areas covered in this chapter are deepfakes, digital rights management, AI/ML-driven multimedia network and service management and network-based attacks and defences for next-generation multimedia.