Skip to content
arXiv· arXiv:2610.14038· Tuo Feng, Yue Zhang, Shiji Zhou, Zhaoxin Fan, Wenguan Wang, Xiaowei Chi, Tianyu Shen, Xiangtai Li, Xuesong Bai, Zheng Zhu, Sirui Han, Boyu Zhou, Zhaosong Wang, Shizhe Chen, Zilong Zheng, Chenjia Bai, Jiangmiao Pang, Yaodong Yang, Hao Zhao, Di Hu, Li Yi, Yan Huang, Qi Wu, Hanwang Zhang, Fei Gao, Shanghang Zhang, Hongsheng Li, Jianye Hao, Yuan Zhang, Yingjie Xia, Jian Shen, Cewu Lu, Hui Zhang, Kai Chen, Wenwu Yu, Badong Chen, Jungong Han, Yike Guo·Indexed by this site · Originally published

From Bits to Atoms: A Survey of Cross-Layer Safety for Embodied Intelligence

From bits to atoms: A survey of cross-layer safety in Embodied AI

AI summary

A survey proposes a four-layer reference architecture consisting of physical interaction, system and middleware, algorithm and decision-making, and cloud-edge and data, to connect vulnerabilities, propagation paths, and protection measures across layers of embodied intelligence. The survey examines how faults and attacks cross architectural boundaries to cause physical consequences, and organizes composite attack chains, multi-layer defense strategies, and representative safety benchmarks. The authors advocate unifying software alignment with strict hardware constraints to build trustworthy systems that can operate safely in human-centered environments.

Read originalAI Plus 2026