Secure Multi-Party Computation for AI-Driven Financial Risk Analytics
DOI:
https://doi.org/10.63282/3050-922X.IJERET-V4I2P108Keywords:
Secure Multi-Party Computation, AI-Driven Risk Analytics, Financial Privacy, Cryptographic Security, Fraud Detection, Federated Learning, Privacy-Preserving AI, Anti-Money Laundering (AML)Abstract
There has been a rise in the use of AI in finance risk solutions, and today, it is easier to detect fraud credit risk and engage in anti-money laundering. However, privacy issues and regulations do not allow financial firms to share their data, reducing AI models' ability to assess risk. Secure Multi-Party Computation (SMPC) offers a viable solution by allowing multiple institutions to engage in joint computation activities on encrypted data information. In this paper, we consider the integration of SMPC with AI-based financial risk analysis; its function is to protect data and share information. In this work, we describe and analyse the simplest methods on which SMPC relies, which include secret sharing, oblivious transfer, and garbled circuits in relation to their suitability for the financial system. This paper presents an information security architecture based on federated learning, differential privacy, and encrypted model inference for risk analysis while being compliant with the regulations. Hence, the outcomes of the numerical experiments reveal that the use of SMPC in developing the AI models allows for detecting fraud and assessing risk with high accuracy if a small trade-off of privacy is acceptable. However, there are some disadvantages due to the use of the algorithm, such as high computational overhead and scalability. As for future work, research is needed on improving cryptographic protocols, applying combined CP and DP methods, and improving AI’s performance in secure settings. The insights also reveal the opportunity to utilize SMPC-driven AI solutions to enhance financial risk mitigation since it fosters interoperability between organisations and datasets, with particular attention to data privacy standards, including GDPR and PSD2
References
[1] Alghamdi, W., Salama, R., Sirija, M., Abbas, A. R., & Dilnoza, K. (2023). Secure multi-party computation for collaborative data analysis. In E3S Web of Conferences (Vol. 399, p. 04034). EDP Sciences.
[2] Ivanovic, M., Autexier, S., Kokkonidis, M., & Rust, J. (2023). Quality medical data management within an open AI architecture–cancer patients case. Connection Science, 35(1), 2194581.
[3] Archer, D. W., Bogdanov, D., Lindell, Y., Kamm, L., Nielsen, K., Pagter, J. I., ... & Wright, R. N. (2018). From keys to databases real-world applications of secure multi-party computation. The Computer Journal, 61(12), 1749-1771.
[4] AI, Machine Learning, and the Future of Credit Risk Management, Birlasoft, 2020. online. https://www.birlasoft.com/articles/ai-machine-learning-and-future-credit-risk-management
[5] Nookala, G. (2023). Secure Multiparty Computation (SMC) for Privacy-Preserving Data Analysis. Journal of Big Data and Smart Systems, 4(1).
[6] Chakraborty, S. (2016). Mobile Commerce: Secure Multi-party Computation & Financial Cryptography. Cryptology ePrint Archive.
[7] Hu, W., Xia, X., Ding, X., Zhang, X., Zhong, K., & Zhang, H. F. (2022). SMPC-ranking: A privacy-preserving method for identifying influential nodes in multiple private networks. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 53(5), 2971-2982.
[8] Boinapalli, N. R. (2023). AI-Driven Predictive Analytics for Risk Management in Financial Markets. Silicon Valley Tech Review, 2(1), 41-53.
[9] Pillai, V. (2023). Integrating AI-Driven Techniques in Big Data Analytics: Enhancing Decision-Making in Financial Markets. International Journal of Engineering and Computer Science, 12(07), 10-18535.
[10] Malhotra, Y. (2018). AI, machine learning & deep learning risk management & controls: beyond deep learning and generative adversarial networks: model risk management in AI, machine learning & deep learning. Machine Learning & Deep Learning (May 16, 2018). Paper Accepted for Presentation at the.
[11] Roberts, T., & Tonna, S. J. (2022). Risk modeling: practical applications of artificial intelligence, machine learning, and deep learning. John Wiley & Sons.
[12] Fritz-Morgenthal, S., Hein, B., & Papenbrock, J. (2022). Financial risk management and explainable, trustworthy, responsible AI. Frontiers in artificial intelligence, 5, 779799.
[13] Deebak, B. D., & Fadi, A. T. (2021). Privacy-preserving in smart contracts using blockchain and artificial intelligence for cyber risk measurements. Journal of Information Security and Applications, 58, 102749.
[14] Cuervo, R. (2023). Predictive AI for SME and Large Enterprise Financial Performance Management. arXiv preprint arXiv:2311.05840.
[15] Chong, H. Y., & Diamantopoulos, A. (2020). Integrating advanced technologies to uphold the security of payment: Data flow diagram. Automation in construction, 114, 103158.
[16] Becker, T., Mencer, O., Weston, S., & Gaydadjiev, G. (2015). Maxeler data flow in computational finance. FPGA Based Accelerators for Financial Applications, 243-266.
[17] Sousa, M. R., Gama, J., & Brandão, E. (2016). A new dynamic modeling framework for credit risk assessment. Expert Systems with Applications, 45, 341-351.
[18] Liu, Z. (2023). MPC-enabled privacy-preserving machine learning.
[19] Zhao, C., Zhao, S., Zhao, M., Chen, Z., Gao, C. Z., Li, H., & Tan, Y. A. (2019). Secure multi-party computation: theory, practice, and applications. Information Sciences, 476, 357-372.
[20] Dispan, J. (2023). Confidential Computing via Multiparty Computation and Trusted Computing.
