Enhancing the Security of AI-Driven Systems in Space Exploration and Research
Main Article Content
Abstract
AI's increasing presence in space exploration platforms has allowed mission opportunities from autonomous navigation to real-time anomaly detection. However, this progress brings with it a new category of cybersecurity problems that traditional cybersecurity models do not expect. This paper provides a comprehensive review of the adversarial attack vectors for AI systems in space environments, such as input manipulation, poisoning of training data, degradation of AI models, and exploitation of the supply chain. Based on the 2022 incident with ViaSat KA-SAT, which knocked out approximately 45,000 modems in Europe, this research highlights major shortcomings of today's space security stances. It is proposed to implement a five-layer defense-in-depth approach, based on guidance from NIST CSF 2.0 [5], NIST AI RMF 1.0 [6], MITRE ATLAS [7], and CCSDS standards [9]. Architecture is designed to address data integrity, model robustness, infrastructure hardening, encrypted communications, and autonomous operational resilience, all of which are specific to space deployment restrictions. The framework is validated as it is practically effective in the case of the KA-SAT attack chain
Article Details
Section
How to Cite
References
[1] V. R. Glick, A. H. Bruder, A. Futerman, K. Steval, H. G. Diament, and Z. Wawrzyniak, "Securing the final frontier: Cybersecurity risk, regulation, and compliance trends in space and satellite operations," Mayer Brown, Dec. 2025. [Online]. Available: https://www.mayerbrown.com/en/insights/publications/2025/12/securing-the-final-frontier-cybersecurity-risk-regulation-and-compliance-trends-in-space-and-satellite-operations
[2] N. Boschetti, N. Gordon, and G. Falco, "Space cybersecurity lessons learned from the ViaSat cyberattack," in Proc. AIAA ASCEND Conf., 2022. [Online]. Available: https://www.researchgate.net/publication/363558808
[3] A. Kazi, S. Kazi, and S. Bhosale, "Invisible battlefields: Analyzing the Viasat attack and its broader implications," Scientific Bulletin, 2025, doi: 10.2478/bsaft-2025-0007.
[4] C. Poirier, "Hacking the cosmos: Cyber operations against the space sector," Center for Security Studies, ETH Zürich, Zürich, Switzerland, Tech. Rep., 2024. [Online]. Available: https://www.research-collection.ethz.ch/handle/20.500.11850/697348
[5] National Institute of Standards and Technology, "The NIST cybersecurity framework (CSF) 2.0," NIST, Gaithersburg, MD, USA, Tech. Rep. NIST CSWP 29, Feb. 2024. [Online]. Available: https://nvlpubs.nist.gov/nistpubs/CSWP/NIST.CSWP.29.pdf
[6] National Institute of Standards and Technology, "Artificial intelligence risk management framework (AI RMF 1.0)," NIST, Gaithersburg, MD, USA, Tech. Rep. NIST AI 100-1, Jan. 2023. [Online]. Available: https://nvlpubs.nist.gov/nistpubs/ai/NIST.AI.100-1.pdf
[7] MITRE Corporation, "MITRE ATLAS: Adversarial threat landscape for artificial intelligence systems," MITRE, McLean, VA, USA, 2023. [Online]. Available: https://atlas.mitre.org/matrices/ATLAS
[8] National Aeronautics and Space Administration, "Artificial intelligence (AI) at NASA," NASA, Washington, DC, USA, 2023. [Online]. Available: https://www.nasa.gov/artificial-intelligence
[9] Consultative Committee for Space Data Systems, "Security guide for space data systems," CCSDS, Tech. Rep. CCSDS 350.1-G-3, 2022. [Online]. Available: https://public.ccsds.org/Pubs/350x1g3.pdf
[10] F. Cremer et al., "Cyber risk and cybersecurity: A systematic review of data availability," The Geneva Papers on Risk and Insurance, vol. 47, no. 3, pp. 698–736, 2022, doi: 10.1057/s41288-022-00266-6.
[11] X. Wu et al., "Threat analysis for a space information network based on network security attributes: A review," Complex Intell. Syst., vol. 8, pp. 3921–3936, 2022, doi: 10.1007/s40747-022-00679-7.
[12] Z. Yang, J. Zhang, W. Wang, and H. Li, "Invisible threats in the data: A study on data poisoning attacks in deep generative models," Appl. Sci., vol. 14, no. 19, p. 8742, Sep. 2024, doi: 10.3390/app14198742.
[13] F. Marulli, S. Marrone, and L. Verde, "Sensitivity of machine learning approaches to fake and untrusted data in the healthcare domain," J. Sens. Actuator Netw., vol. 11, no. 2, p. 21, Apr. 2022, doi: 10.3390/jsan11020021.
[14] I. Jada and T. O. Mayayise, "The impact of artificial intelligence on organizational cyber security: A systematic literature review," Data Inf. Manage., vol. 8, no. 2, p. 100063, 2023, doi: 10.1016/j.dim.2023.100063.
[15] R. Kaur, D. Gabrijelčič, and T. Klobučar, "Artificial intelligence for cybersecurity: Literature review and future research directions," Inf. Fusion, vol. 97, p. 101804, Sep. 2023, doi: 10.1016/j.inffus.2023.101804.
[16] I. A. D. Nesnas, L. M. Fesq, and R. A. Volpe, "Autonomy for space robots: Past, present, and future," Current Robot. Rep., vol. 2, no. 3, pp. 251–263, Sep. 2021, doi: 10.1007/s43154-021-00057-2.
[17] M. M. Yamin, M. Ullah, H. Ullah, and B. Katt, "Weaponized AI for cyber-attacks," J. Inf. Secur. Appl., vol. 57, p. 102722, Mar. 2021, doi: 10.1016/j.jisa.2020.102722.
[18] M. Pooyandeh, K.-J. Han, and I. Sohn, "Cybersecurity in the AI-based metaverse: A survey," Appl. Sci., vol. 12, no. 24, p. 12993, Dec. 2022, doi: 10.3390/app122412993.