Alexandra Weber (Telespazio Germany GmbH), Peter Franke (Telespazio Germany GmbH)

Space missions increasingly rely on Artificial Intelligence (AI) for a variety of tasks, ranging from planning and monitoring of mission operations, to processing and analysis of mission data, to assistant systems like, e.g., a bot that interactively supports astronauts on the International Space Station. In general, the use of AI brings about a multitude of security threats. In the space domain, initial attacks have already been demonstrated, including, e.g., the Firefly attack that manipulates automatic forest-fire detection using sensor spoofing. In this article, we provide an initial analysis of specific security risks that are critical for the use of AI in space and we discuss corresponding security controls and mitigations. We argue that rigorous risk analyses with a focus on AI-specific threats will be needed to ensure the reliability of future AI applications in the space domain.

View More Papers

GNNIC: Finding Long-Lost Sibling Functions with Abstract Similarity

Qiushi Wu (University of Minnesota), Zhongshu Gu (IBM Research), Hani Jamjoom (IBM Research), Kangjie Lu (University of Minnesota)

Read More

MPCDiff: Testing and Repairing MPC-Hardened Deep Learning Models

Qi Pang (Carnegie Mellon University), Yuanyuan Yuan (HKUST), Shuai Wang (HKUST)

Read More

WIP: Towards Practical LiDAR Spoofing Attack against Vehicles Driving...

Ryo Suzuki (Keio University), Takami Sato (University of California, Irvine), Yuki Hayakawa, Kazuma Ikeda, Ozora Sako, Rokuto Nagata (Keio University), Qi Alfred Chen (University of California, Irvine), Kentaro Yoshioka (Keio University)

Read More