Recent Innovations in AI Privacy: Protecting Data in the Age of Machine Learning

Authors

  • Siddhant Sonkar University of California, Irvine, USA Author

DOI:

https://doi.org/10.32628/CSEIT25112390

Keywords:

Privacy-Preserving AI, Federated Learning, Differential Privacy, Homomorphic Encryption, Zero-Knowledge Proofs

Abstract

This comprehensive article explores recent advancements in privacy-preserving technologies within artificial intelligence systems, focusing on five key approaches: federated learning, differential privacy, homomorphic encryption, privacy-preserving machine learning (PPML), and zero-knowledge proofs. The article examines how these technologies address critical privacy challenges in machine learning environments while maintaining model performance and utility. The article highlights the implementation of these approaches across various domains, particularly in healthcare and financial services, demonstrating their effectiveness in protecting sensitive data throughout the machine learning lifecycle. The article reveals how these technologies complement each other to create robust privacy protection frameworks while enabling organizations to leverage the power of AI without compromising data confidentiality.

Downloads

Download data is not yet available.

References

Thanh Tam Nguyen et al., "Privacy-preserving explainable AI: a survey," Science China Information Sciences, Volume 68, article number 111101, (2025), 07 November 2024. Available: https://link.springer.com/article/10.1007/s11432-024-4123-4

Soumia Zohra El Mestari et al., "Preserving data privacy in machine learning systems," Computers & Security, Volume 137, February 2024, 103605. Available: https://www.sciencedirect.com/science/article/pii/S0167404823005151

Huiming Chen et al., "Advancements in Federated Learning: Models, Methods, and Privacy," arXiv:2302.11466 [cs.AI], 5 Mar 2023. Available: https://arxiv.org/abs/2302.11466

Pallavi Dhade and Prajakta Shirke, "Federated Learning for Healthcare: A Comprehensive Review," Eng. Proc. 2023, 59(1), 230, 9 February 2024. Available: https://www.mdpi.com/2673-4591/59/1/230

Lea Demelius et al., "Recent Advances of Differential Privacy in Centralized Deep Learning: A Systematic Survey," ACM Computing Surveys, Volume 57, Issue 6, Article No.: 158, Pages 1 - 28, 10 February 2025. Available: https://dl.acm.org/doi/10.1145/3712000

Kareem Amin et al., "Practical Considerations for Differential Privacy," ResearchGate, August 2024. Available: https://www.researchgate.net/publication/383119384_Practical_Considerations_for_Differential_Privacy

Jiangjun Yuan et al., "Approximate homomorphic encryption based privacy-preserving machine learning: a survey," Volume 58, article number 82, (2025), 06 January 2025. Available: https://link.springer.com/article/10.1007/s10462-024-11076-8

Michael Brenner et al., "Practical Applications of Homomorphic Encryption," Proceedings of the International Conference on Security and Cryptography (SECRYPT-2012), pages 5-14, 2012. Available: https://www.scitepress.org/papers/2012/39694/39694.pdf

Harry Chandra Tanuwidjaja et al., "Privacy-Preserving Deep Learning on Machine Learning as a Service—a Comprehensive Survey," ResearchGate, January 2020. Available: https://www.researchgate.net/publication/346288300_Privacy-Preserving_Deep_Learning_on_Machine_Learning_as_a_Service-a_Comprehensive_Survey

Julius Atetedaye, "Privacy-Preserving Machine Learning: Securing Data in AI Systems," ResearchGate, May 2024. Available: https://www.researchgate.net/publication/380711820_Privacy-Preserving_Machine_Learning_Securing_Data_in_AI_Systems

Zhibo Xing et al., "Zero-knowledge Proof Meets Machine Learning in Verifiability: A Survey," arXiv:2310.14848v1 [cs.LG], 23 Oct 2023. Available: https://arxiv.org/pdf/2310.14848

Ryan Lavin et al., "A Survey on the Applications of Zero-Knowledge Proofs," arXiv preprint arXiv:2408.00243, 2024. https://arxiv.org/abs/2408.00243

Downloads

Published

05-03-2025

Issue

Section

Research Articles