A historical cipher meets persistent AI search
Vals AI says Claude Fable 5.1 produced a plaintext for Sir Thomas Urquhart’s Cyphral Distich, a two-line 64-number cryptogram printed in a 1653 work. The result is a claim, not an independently reviewed academic finding, but it is concrete enough to examine: a reproducible indexing rule, a proposed plaintext and a reason the method may have been missed.What Vals AI says Fable 5.1 solved
Vals AI reports that it gave Claude Fable 5.1 an open task: solve Sir Thomas Urquhart’s Cyphral Distich. According to the post, the model reached a proposed solution after 44 minutes, 176,000 tokens and no interjections from the human operator.The Cyphral Distich is described as two lines of 32 numbers each, printed at the end of Urquhart’s Logopandecteision. Vals AI says the cipher had remained seemingly unsolved for centuries, had been posed in Notes and Queries in 1899, appeared in later cryptography literature and was listed by historical-cipher researcher Klaus Schmeh among his Top 50 unsolved encrypted messages. The implication is narrow but notable: if the proposed method checks out against the cited edition, an old archival puzzle may have yielded to a search process that combined persistence with close reading rather than advanced cryptanalysis.
The proposed key is inside the book
The claimed breakthrough is that the key was not an external alphabet or frequency pattern, but the surrounding text. Vals AI says earlier attempts tended to explore substitution, frequency analysis or homophonic substitution, while Fable 5.1 focused on the cipher’s placement immediately after Urquhart’s 32 Proquiritations.The rule described by Vals AI is simple: for the i-th number in a cipher line, go to the i-th Proquiritation, use that number as a word index, and take the first letter of that word. Applying that procedure allegedly yields:
The result is presented as self-checking because each cipher line contains 32 letters, the two lines end with and and land, and the output forms a rhyming two-line distich. Vals AI also argues that the content fits Urquhart’s Royalist politics, making a hidden prayer for Charles II historically plausible rather than arbitrary.
Why the caveats matter
The finding should be read as a reported solution, not as a settled scholarly consensus. The source packet provides Vals AI’s account and bibliographic pointers, but not an independent peer review, a full external reproduction or a separate primary-source verification of every coordinate.There is also a small counting caveat inside the Vals AI post itself: one coordinate depends on treating the Latin expression hinc inde as a single unit; otherwise that position shifts by one. That does not necessarily undermine the proposed plaintext, but it matters for reproducibility. Historical ciphers can look convincing when partial patterns align, so the practical next step is straightforward: researchers can test the indexing rule line by line against the cited 1834 Maitland Club edition of Urquhart’s works and decide whether the reading is robust.
A second Urquhart cipher broadens the claim
Vals AI says Urquhart left another, much larger cryptogram in the same style: the Cyphral Octastich in The Jewel from 1652. The post describes it as 285 numbers rather than 64 and says it was also unsolved before Fable 5.1 deciphered it.The source packet does not provide the full Octastich plaintext in the excerpt, so that part of the claim is weaker for publication than the Distich account. Still, its inclusion is relevant because it shows how Vals AI frames the experiment: not as a one-off guess, but as a model identifying a hidden indexing mechanism and then applying it to a related Urquhart puzzle. If later documentation supplies the larger solve in reproducible form, the Octastich would become a stronger test of whether the method generalizes.
The AI lesson is persistence, not magic
Vals AI’s own interpretation is restrained in one useful respect: it says the answer was simple in hindsight and that the model did not perform an extraordinary feat of cryptanalysis. The reported advantage was sustained search across obscure material, noticing that this particular problem was unusually tractable, and continuing long enough to connect the 32 Proquiritations with the two 32-number cipher lines.That matters for AI research because many historical or archival problems are bottlenecked by attention. A human specialist may need hours or days to read old editions, test dead ends and follow faint references. A language model that can persist through those loops may not replace expert validation, but it can generate candidate solutions that scholars can check. The distinction is essential: the model can accelerate discovery, while verification still depends on transparent methods and source-grounded review.
Conclusion
The Cyphral Distich claim is newsworthy because it offers a specific, testable proposed solution to a long-discussed historical cipher. The strongest part of the case is the internal alignment: 32 Proquiritations, two 32-number lines, a clear word-indexing rule and a plaintext that fits both form and political context.The weakest part is the current evidence boundary. The account comes from Vals AI, and the broader claim needs independent reproduction before it can be treated as final. For now, the most accurate reading is that Fable 5.1 has produced a plausible and unusually concrete candidate solution, with enough detail for historians and cryptography researchers to verify or challenge it.
Sources
Editorial Team - CoinBotLab
