Cryptographic randomness is unpredictable data generated for security-sensitive purposes, such as creating encryption keys, session tokens, nonces, and initialization values. Unlike ordinary random numbers used in simulations, it must be difficult for an attacker to predict or reproduce.
Weak or repeated random values can let attackers guess secrets, impersonate users, or undermine encryption. Secure software uses cryptographically secure random number generators, typically seeded with unpredictable system entropy, and avoids homemade generators or predictable seeds. In practice, runtime fixes involving random-number handling matter because applications may rely on those facilities without generating randomness directly. Keep runtimes and cryptographic libraries patched, and use their documented secure random APIs.