Rogue Decryption Failures: Reconciling AE Robustness Notions

Abstract

An authenticated encryption scheme is deemed secure (AE) if ciphertexts both look like random bitstrings and are unforgeable. AE is a much stronger notion than the traditional IND–CCA. One shortcoming of AE as commonly understood is its idealized, all-or-nothing decryption: if decryption fails, it will always provide the same single error message and nothing more. Reality often turns out differently: encode-then-encipher schemes often output decrypted ciphertext before verification has taken place whereas pad-then-MAC-then-encrypt schemes are prone to distinguishable verification failures due to the subtle interaction between padding and the MAC-then-encrypt concept. Three recent papers provided what appeared independent and radically different definitions to model this type of decryption leakage.

We reconcile these three works by providing a reference model of security for authenticated encryption in the face of decryption leakage from invalid queries. Having tracked the development of AE security games, we provide a single expressive framework allowing us to compare and contrast the previous notions. We find that at their core, the notions are essentially equivalent, with their key differences stemming from definitional choices independent of the desire to capture real world behaviour.