A Survey on the Applications of Artificial Intelligence in Cryptanalysis and Cryptographic Design

Shuangjin Wu; Wenbo Wang

doi:10.54691/akyt1k78

Authors

Shuangjin Wu
Wenbo Wang

DOI:

https://doi.org/10.54691/akyt1k78

Keywords:

Cryptography, Cryptanalysis, Side-Channel Analysis (SCA), Differential Fault Analysis (DFA), Neural Differential Cryptanalysis, Generative Adversarial Networks (GANs), Lightweight Cryptography.

Abstract

Artificial Intelligence (AI) is profoundly transforming cryptography by significantly enhancing cryptanalysis techniques and informing innovative cryptographic design approaches. This survey reviews recent advancements in applying deep learning methods to side-channel and differential fault analyses, demonstrating substantial improvements over traditional methods in attack efficiency, accuracy, and resilience. Additionally, it highlights breakthroughs such as neural differential cryptanalysis, which expand classical cryptanalytic boundaries. In cryptographic design, Generative Adversarial Networks (GANs) have successfully automated the creation of high-quality cryptographic primitives, particularly S-boxes. Furthermore, AI shows promise in post-quantum cryptography (PQC) by uncovering potential vulnerabilities and optimizing cryptographic parameters. Despite these advancements, challenges persist regarding data dependency, model generalization, and interpretability. Future research directions emphasize enhancing AI model explainability, creating standardized benchmarks, and integrating AI with emerging technologies such as quantum computing and zero-knowledge proofs.

Downloads

Download data is not yet available.

References

[1] R. Rivest: Machine Learning and Cryptanalysis, ASIACRYPT’91 (1991), p. 427–439.

[2] E. Cagli, C. Dumas and E. Prouff: Convolutional Neural Networks with Data Augmentation against Jitter-Based Countermeasures, CHES 2017 (2017), p. 45–68.

[3] H. Maghrebi, T. Portigliatti and E. Prouff: Breaking Cryptographic Implementations Using Deep Learning, SPACE 2016 (2016), p. 3–26.

[4] R. Benadjila, E. Prouff, R. Strullu, E. Cagli and C. Dumas: Deep Learning for Side-Channel Analysis and Introduction to ASCAD Database, Journal of Cryptographic Engineering, Vol. 10 (2020) No. 2, p. 163–188.

[5] L. Wu et al.: Ranking Loss: Maximizing the Success Rate in Deep Learning Side-Channel Analysis, IACR Transactions on Cryptographic Hardware and Embedded Systems, Vol. 2021 (2021) No. 1, p. 25–55.

[6] Y. Zotkin et al.: Deep Learning vs. Template Attacks: Experimental Study, IACR ePrint Archive, Report 2018/1213 (2018).

[7] Y. Cheng et al.: DLFA: Deep Learning-based Fault Analysis against Block Ciphers, IACR ePrint Archive, Report 2023/021 (2023).

[8] A. Heuser et al.: Side-Channel Analysis of Lightweight Ciphers: Does Lightweight Equal Easy?, RFIDSec 2016 (2016), p. 91–104.

[9] J. Kim et al.: Make Some Noise: Unleashing the Power of CNNs for Profiled Side-Channel Analysis, IACR Transactions on Cryptographic Hardware and Embedded Systems, Vol. 2019 (2019) No. 3, p. 148–179.

[10] A. Gohr: Improving Differential Cryptanalysis with Deep Learning, CRYPTO 2019 (2019), p. 3–24.

[11] D. Gerault et al.: SoK: Five Years of Neural Differential Cryptanalysis, IACR ePrint Archive, Report 2024/1300 (2024).

[12] R. Zhang et al.: A Novel S-Box Generation Methodology Based on the Optimized GAN Model, Computers, Materials & Continua, Vol. 76 (2023) No. 2, p. 1911–1927.

[13] L. Lerman et al.: A Machine Learning Approach Against a Masked AES, CARDIS 2013 (2013), p. 61–75.

[14] Y. Li et al.: Salsa Picante: A Machine Learning Attack on LWE with Binary Secrets, Proceedings of the 2023 ACM CCS (2023), p. 112–125.

[15] K. Ohno and M. Nakabayashi: A Security Verification Framework of Cryptographic Protocols Using Machine Learning, arXiv preprint arXiv:2304.13249 (2023).

[16] Y. Liu et al.: Machine Learning Assisted Differential Cryptanalysis, IEEE Access, Vol. 7 (2019), p. 76547–76556.

[17] E. Bursztein: Hacker’s Guide to Deep Learning Side-Channel Attacks (Part 1) (2021).

[18] S. Picek et al.: When Theory Meets Practice: Profiled Side-Channel Analysis with Hyperparameter Tuning, IACR Transactions on Cryptographic Hardware and Embedded Systems, Vol. 2021 (2021) No. 3, p. 677–707.

[19] J. Zhang et al.: Fault Template Attacks on AES and Their Countermeasures, IEEE Transactions on Very Large Scale Integration (VLSI) Systems, Vol. 28 (2020) No. 12, p. 2638–2650.

[20] A. Saha et al.: Improved Fault Analysis on LED Block Cipher Using DFA and Machine Learning, Security and Communication Networks, Vol. 2018 (2018), p. 8165294.