Modern Cryptography
The canonical five primitives that founded the modern era — plus the wider landscape they unlocked
Every cipher in the previous ten halls eventually fell. The five primitives in this hall — DES, Diffie-Hellman, RSA, AES, SHA-256 — were designed in the public eye, scrutinised for decades, and now anchor essentially every secure system on Earth. They are not the only modern primitives — ChaCha20, ECDH (X25519), Ed25519, BLAKE2/3, and the post-quantum NIST finalists all sit alongside them — but they are the canonical foundation from which the rest of the field is built. Only one of them, AES, is technically a cipher in the classical sense. The other four solve problems that classical cryptography could not: distributing keys without meeting in person, signing documents at a distance, and verifying integrity without a shared secret.
For the broader landscape — ChaCha20, ECDH, post-quantum standardisation, and the cipher-vs-protocol taxonomy — see the Modern Cryptography wing. For the design rule that underpins every card in this hall, see Kerckhoffs's Principle.
The first public, government-standardised cipher. A 16-round Feistel network with a 56-bit key — broken by EFF's Deep Crack in 1998 (56 hours). Triple-DES extended its life until 2023. The cipher that birthed academic cryptanalysis: differential and linear cryptanalysis were both invented to attack it.
Two strangers create a shared secret over an open wire. Not a cipher — a key-agreement protocol. The 1976 paper that overturned cryptography's three-thousand-year-old ‘you must meet first' assumption. Every TLS 1.3, SSH, Signal, and WireGuard handshake descends from it.
The first practical public-key cryptosystem. One key encrypts, a different key decrypts — and you can publish the encrypting key. Solved key distribution, authentication, and digital signatures in one stroke. Every TLS certificate, code-signing chain, and PGP keypair is descended from this 1977 paper.
The cipher that encrypts the modern world. Chosen by open international competition in 2001; designed by Joan Daemen and Vincent Rijmen of Belgium. Substitution-permutation network, 128-bit blocks, 128/192/256-bit keys, 10–14 rounds. After two decades of intense scrutiny, no practical attack exists.
A 256-bit fingerprint for any input. One-way, collision-resistant, and the verification engine of Git, Bitcoin, TLS certificates, and software integrity. Not a cipher — there is no key and no decryption. Irreversibility is the entire point.
The museum's closing lesson: Classical cryptography failed because it relied on obscurity, physical key distribution, and small key spaces. Modern cryptography replaces all three with mathematical hardness, public-key mathematics, and key sizes large enough that brute force is physically impossible — and expanded the field beyond ciphers into protocols, signatures, and proofs.