The Polyglottal Cipher

The Polyglottal Cipher is a post-quantum steganographic encryption system that hides encrypted data inside Unicode glyphs across 202 languages and scripts.

Unlike conventional encryption — which marks itself as encrypted, making it a target for interception, legal compulsion, and brute-force attack — the Polyglottal Cipher produces output that looks like natural multilingual text. Even with the data in hand, an attacker doesn't know they're holding anything.

The core encryption layer uses ChaCha20-Poly1305. The steganographic layer maps encrypted bytes into the vast unused space of Unicode — 133,387 characters across 202 language traditions, most of which are never used in normal text. The result is encrypted data that inhabits language rather than hiding behind it.

Encryption: ChaCha20-Poly1305 (IETF standard, post-quantum resistant design). Entropy-seeded stochastic selection via PoemRotor — 117 poem forms across 40+ language traditions generate the entropy that drives glyph selection.

Steganography: Encrypted bytes are mapped into Unicode glyph space using the ProvenanceRotor — 12 heterogeneous database backends that ensure no single point of pattern analysis. The output passes statistical analysis for natural language distribution.

Infrastructure: MongoDB wire protocol proxy enables existing applications to encrypt transparently with a single connection string change. No code modification required.

The Polyglottal Cipher is under formal review by the COSIC group at KU Leuven — one of the most respected cryptography research centers in the world. Tim Beyne was assigned by Professor Vincent Rijmen, co-inventor of the AES encryption standard that protects every bank transaction on Earth.

European Patent application EP26025007.1 has been filed with the European Patent Office.