ciphertextonly

Ciphertext-only refers to a cryptanalytic model in which an attacker has access only to ciphertexts produced by an encryption scheme, with no access to the corresponding plaintexts or the secret keys. The attacker may know which algorithm was used and may observe many ciphertext samples, but they do not possess plaintext-ciphertext pairs. The objective is to deduce the plaintexts, recover the key, or gain any usable information about the underlying message, from the ciphertexts alone.

This model represents a relatively weak information-leakage assumption for assessing confidentiality. If a scheme remains secure under ciphertext-only conditions, it is considered resistant to adversaries who do not obtain any plaintext-ciphertext pairs. In practice, many classical ciphers could be broken with ciphertext-only analysis by exploiting statistical regularities, frequency patterns, or other structural weaknesses. Modern encryption schemes, when implemented correctly and used with appropriate modes, are designed to be secure against ciphertext-only attacks, making it infeasible to recover plaintexts from ciphertexts alone.

Security considerations for ciphertext-only attacks emphasize the strength of the underlying algorithm and the mode of operation. Modern cryptographic theory often relates ciphertext-only security to stronger notions such as semantic security or indistinguishability, for example IND-CPA, which imply that ciphertexts do not reveal informative distinctions about the plaintext even under more powerful attack models. In practice, even ciphertext-only observations can leak auxiliary information, such as message lengths, timing, or traffic patterns, which may be exploited in side-channel or traffic-analysis contexts.

See also: ciphertext-only attack, known-plaintext attack, chosen-plaintext attack, chosen-ciphertext attack, cryptanalysis.