EU - KEYLESS - Keyless Cryptography for Efficiency and Security

Cryptographic primitives are the foundation of security in any secure information system. Permutations and similar keyless cryptographic primitives have risen to great popularity in recent years thanks to their flexibility and performance. They power new lightweight cryptography standards or serve as core building blocks of post-quantum cryptography and advanced privacy-preserving protocols. However, the security analysis of these designs still follows the traditional cryptanalytic methodology based on decades of research in block ciphers, largely ignoring the substantial differences between the two design paradigms. In KEYLESS, we propose new methodologies to achieve an accurate, transparent security evaluation of keyless primitives. This has the potential to enable drastic performance improvements as well as prevent security vulnerabilities arising from hidden dependencies. We will establish new, finegrained models of keyless primitives to obtain tighter proofs and lightweight designs. This allows to simultaneously improve both security and efficiency. We will tackle the challenges of keyless settings with novel cryptanalytic techniques and develop formal methods to prove optimality of attacks. In particular, we will systematically take dependencies between rounds or primitive calls into account and thus achieve complete models of complex attacks. Finally, we will explore the full potential of keyless primitives to not only provide efficient security, but also practical robustness and resilience under suboptimal conditions, including misuse and side-channel attacks. Unlike previous work, we will study robustness properties in conjunction to exploit synergies, and obtain new designs that achieve full robustness while maintaining efficiency.