Privacy in Quantum Computing: A Comprehensive Review

Abstract

Quantum technologies are beginning to reshape how large volumes of data are processed, yet they also introduce a new class of privacy challenges with no classical counterparts. These concerns stem from the availability of quantum side information, the encoding of data in non-classical states, and the fragility of conventional cryptographic systems in post-quantum environments. This work reviews the main privacy-preserving strategies proposed in peer-reviewed literature from 2020 to 2025, covering both quantum and post-quantum settings. The survey examines several key directions: (i) the integration of differential privacy into quantum machine learning pipelines, (ii) privacy guarantees derived from gentle-measurement principles, (iii) secure aggregation for quantum federated learning, and (iv) privacy amplification methods designed to remain robust under noisy quantum-channel conditions. Representative examples include the mechanisms introduced by Watkins et al. for embedding differential privacy into variational circuits, the theoretical limits established by Aaronson and Rothblum for privacy under gentle measurements, the distributed-learning protections developed by Liu and Zhang, and the noise-aware key-extraction analysis presented by Singh and colleagues. System-level considerations for future quantum infrastructures are discussed in the work of, while previous literature proposes hybrid quantum classical designs to strengthen privacy in post-quantum frameworks. Taken together, these studies reveal open challenges related to scalability, composability, and adversarial modelling, and they motivate a roadmap toward developing practical and verifiable privacy guarantees for emerging quantum platforms.