Security of AI Models using Paillier cryptosystem

Brook Hilemriam; Asamene  Kelelom; Beer Singh

doi:10.70454/JRICST.2026.30104

Authors

Brook Hilemriam College of Engineering and Technology Samara University, Ethiopia. Author
Asamene Kelelom College of Engineering and Technology Samara University, Ethiopia. Author
Beer Singh Seth Vishambhar Nath Institute of Engineering & Technology, Barabanki, UP, INDIA Author

DOI:

https://doi.org/10.70454/JRICST.2026.30104

Keywords:

AI Models , Paillier cryptosystem, Cryptography, data security

Abstract

This paper demonstrates the practical operation of the Paillier cryptosystem in securing direct retrogression conclusion while conserving data sequestration. A customer- garçon armature is used, where sensitive input data is translated on the customer side using Paillier’s cumulative homomorphic encryption and reused on the garçon without revealing the raw values. The translated data is subordinated to a direct retrogression model (with a predefined weight and bias), using homomorphic operations to cipher translated prognostications directly. After decryption, the works corresponded exactly to the expected direct results, thus confirming the correctness of the calculations in the translated space. A performance test carried out on a dataset of 30 numerical values showed an average of the encryption and decryption times of 0.31 seconds and 0.28 seconds respectively. Nevertheless, the performance is quite effective for small to medium-scale datasets. Such a system yields a highly secure sequestration- conserving output for sensitive operations like healthcare analytics or fiscal modelling. The findings indicate that the model keeps computational consistency even when the data is encrypted, thus enabling precise and reliable predictions that do not compromise data security. Also, the simplicity and availability of the perpetration using Python and the PHE library make it a practical choice for real- world deployments. While presently limited to direct models due to the cumulative nature of the Paillier scheme, the approach presents a strong case for extending sequestration- conserving ways to broader classes of machine literacy algorithms. This exploration highlights the eventuality of homomorphic encryption in enabling secure and secure AI systems, especially in scripts where confidentiality is consummate.

Downloads

Download data is not yet available.

References

[1] T. Li, A. K. Sahu, A. Talwalkar, and V. Smith, “Federated learning: Challenges, methods, and future directions,” IEEE Signal Process. Mag., vol. 37, no. 3, pp. 50–60, 2018.

[2] K. Bonawitz et al., “Towards federated learning at scale: System design,” in Proc. 2nd SysML Conf., 2019, pp. 1–10.

[3] M. Shah, W. Zhang, H. Hu, and N. Yu, “Paillier cryptosystem based mean value computation for encrypted domain image processing operations,” ACM Trans. Multimedia Comput. Commun. Appl., vol. 15, no. 3, pp. 1–21, 2019.

[4] A. K. Sari and F. M. W. Prasetya, “Linear support vector regression in cloud computing on data encrypted using Paillier cryptosystem,” in Proc. Int. Seminar Res. Inf. Technol. Intell. Syst. (ISRITI), Dec. 2019, pp. 434–438.

[5] M. M. S. Altaee and M. Alanezi, “Enhancing cloud computing security by Paillier homomorphic encryption,” Int. J. Electr. Comput. Eng., vol. 11, no. 2, pp. 1771–1779, 2021.

[6] R. Palle and A. Punitha, “Privacy-preserving homomorphic encryption schemes for machine learning in the cloud,” ESP J. Eng. Technol. Advancements, 2021.

[7] H. J. Kiratsata and M. Panchal, “A comparative analysis of machine learning models developed from homomorphic encryption based RSA and Paillier algorithm,” in Proc. 5th Int. Conf. Intell. Comput. Control Syst. (ICICCS), May 2021, pp. 1458–1465.

[8] L. Su, H. Geng, S. Guo, and S. He, “A secure two-party Euclidean distance computation scheme through a covert adversarial model based on Paillier encryption,” IEEE Access, vol. 11, pp. 80986–80996, 2023.

[9] M. M. Hasan et al., “Privacy-preserving quantum key distribution ensemble Paillier cryptosystem for securing IoT based smart metering system,” in Proc. IEEE Int. Conf. Artif. Intell. Eng. Technol. (IICAIET), Aug. 2024, pp. 603–608.

[10] B. Gong et al., “Efficient zero-knowledge arguments for Paillier cryptosystem,” in Proc. IEEE Symp. Security Privacy (SP), May 2024, pp. 1813–1831.

[11] P. Paillier, “Public-key cryptosystems based on composite degree residuosity classes,” in Proc. EUROCRYPT, 1999.

[12] A. Acar, H. Aksu, A. S. Uluagac, and M. Conti, “A survey on homomorphic encryption schemes: Theory and implementation,” ACM Comput. Surv., 2018.

[13] C. Jost, H. Lam, A. Maximov, and B. Smeets, “Encryption performance improvements of the Paillier cryptosystem,” Cryptology ePrint Archive, 2015.

[14] V. O. Odunfa, T. B. Fateye, and A. O. Adewusi, “Application of artificial intelligence approach to African real estate market analysis opportunities and challenges,” Corrosion Manag., vol. 35, no. 1, pp. 10–18, 2025.

[15] V. Saxena and P. Kumar, “Secure transaction of digital currency through fuzzy based cryptography,” Indian J. Sci. Technol., vol. 16, no. 37, pp. 3148–3158, 2023.

[16] P. Kumar and V. Saxena, “Nested levels of hybrid cryptographical technique for secure information exchange,” J. Comput. Commun., vol. 12, no. 2, pp. 201–210, 2024.

[17] P. Kumar, V. Saxena, and K. V. Singh, “Analysis of hybrid cryptography for secure exchange of information,” Int. J. Comput. Appl., vol. 185, no. 4, pp. 37–42, 2023.

[18] P. Kumar and V. Saxena, “Hybrid cryptography for security key exchange through AES and Paillier,” Eur. Chem. Bull., vol. 12, no. 10, pp. 3913–3921, 2023.

[19] S. Kumar et al., “Securing cloud-based systems: DDoS attack mitigation using hypervisor-intrusion detection approach,” Procedia Comput. Sci., vol. 259, pp. 1366–1375, 2025.