Fable 5.1 Solves the Cyphral Distich, a 370-year-old cipher

AI models can now reliably solve complex, long-standing historical ciphers, democratizing access to insights previously requiring specialized human expertise.

Claude Fable 5.1, an AI model, has successfully deciphered the Cyphral Distich, a 370-year-old cipher attributed to Sir Thomas Urquhart. The AI solved the puzzle, which consists of two lines of 64 numbers, within a day. This cipher had eluded human cryptanalysts for centuries, appearing in historical cryptography literature and listed among the top unsolved encrypted messages. Fable 5.1's breakthrough involved recognizing that the cipher's key was embedded within the text itself, specifically referencing the preceding "32 Proquiritations" and the concept of "wishes" mentioned in the accompanying poem. By using the number of each Proquiritation as an index to select the first word and then its first letter, the AI revealed the plaintext: "O GOD UPHOLD KING CHARLS THE SECOND AND MAKE HIM THE SUPREME RULER OF THIS LAND." The AI also deciphered a second, larger cryptogram by Urquhart, the Cyphral Octastich, though with minor lacunae and requiring verification against a physical copy of the text. This success highlights the AI's capability in solving historical and complex textual puzzles that have stumped human experts.

The successful decipherment of these historical ciphers by Fable 5.1 signals a significant advancement in AI's ability to tackle not only mathematical and logical problems but also nuanced, context-dependent textual analysis. Historically, such puzzles were bottlenecked by human attention, requiring extensive manual research and persistent trial-and-error. AI models, with their capacity for processing vast amounts of data and tireless exploration of possibilities, are poised to remove this human bottleneck, potentially unlocking many other historical mysteries and archival puzzles that were previously considered intractable.

This development suggests a broader trend of AI models moving beyond pattern recognition to deeper semantic understanding and creative problem-solving within specialized domains. The AI did not employ extraordinary cryptanalytic techniques but rather demonstrated persistence and an ability to identify simple, elegant solutions embedded within textual context. This capability implies that AI could become an invaluable tool for historians, linguists, and cryptographers, accelerating research and uncovering hidden information across various fields that rely on deciphering complex or obscure texts.