A Systematic Review of Privacy-Enhancing Technologies (PETs) for Securing Personally Identifiable Information in Public Cloud Architectures

The rapid adoption of public cloud services has significantly increased the storage and processing of Personally Identifiable Information (PII), raising critical concerns about data privacy and regulatory compliance. Privacy-Enhancing Technologies (PETs) have emerged as a crucial set of methods and tools designed to protect sensitive data while maintaining functional utility for cloud applications. This systematic review examines the current landscape of PETs deployed for securing PII in public cloud architectures, including homomorphic encryption, differential privacy, federated learning, and confidential computing. A structured literature search was conducted across major scientific databases from 2021 to 2026, following PRISMA guidelines, resulting in the inclusion of studies that evaluate PET performance, scalability, security guarantees, and integration challenges. Thematic synthesis highlights key trends, such as the growing adoption of federated learning for cross-silo data sharing, the application of homomorphic encryption in real-time cloud environments, and the trade-offs between privacy preservation and computational efficiency. Additionally, operational, technical, and regulatory challenges are identified, including computational overhead, standardization barriers, and compliance with global data protection regulations. This review underscores the critical role of PETs in enhancing trust and security in public cloud ecosystems and provides insights for researchers and practitioners seeking to design and implement privacy-aware cloud architectures. Future research directions are discussed, emphasizing the need for optimized PET frameworks that balance security, efficiency, and compliance in increasingly complex cloud environments.