The 240 pages of illustrations depicting dozens of naked women (and some men) in interconnected water bodies, various plants — of which many are completely unknown — astronomical diagrams, only some of which match the views of the sky from Earth, and accompanying text in a script that is more unidentifiable than rongorongo, are sufficient reason to count the Voynich manuscript among the most mysterious books ever written.

The book is thought to have been written sometime in the early 15th century, based on carbon-dating of the vellum (animal skin especially prepared as a material to write on) it was composed on, and is named after a Polish book dealer, Wilfrid Voynich, who purchased it in 1912. Since its rediscovery over 100 years ago, there have been several attempts to decipher its contents, and the manuscript itself has been a challenge for cryptographers ever since.

The latest attempt to crack its code was made by two researchers from the University of Alberta, Canada, who decided to bring in the power of artificial intelligence to help them. Greg Kondrak, a professor of computing science at the university, and Bradley Hauer, a graduate student under him, ran an initial set of algorithms and concluded the script was likely a cipher whose base language was Hebrew.

To identify the language, they used samples of 380 different languages in use today, from the Universal Declaration of Human Rights.

Kondrak, who had initially thought the language would be Arabic, said in a statement on Jan. 25: “That was surprising. And just saying ‘this is Hebrew’ is the first step. The next step is how do we decipher it.”

Assuming that the manuscript’s language was coded using alphagrams — where one phrase is used to define another — they tried to decipher the text and found over 80 percent of the words were present in a Hebrew dictionary. Somehow, the researchers couldn’t find Hebrew scholars to confirm what they found, so they used Google Translate.

And that gave them what they think is the opening line of the book: “She made recommendations to the priest, man of the house and me and people.”

“It came up with a sentence that is grammatical, and you can interpret it. … It’s a kind of strange sentence to start a manuscript but it definitely makes sense,” Kondrak said.

But the decision to use Google’s translation tool, which a simple search would show anyone is riddled with mistakes, did not sit well with many scholars who have spent time studying the Voynich manuscript.

Elonka Dunin, a cryptographer and video game developer, told Mental Floss her take on the new paper: “I have very little faith in it. Hebrew and dozens of other languages have been identified before. Everyone sees what they want to see. … They scrambled the texts using their own system, then they used their own software to de-scramble those. Then they used it on the manuscript and said, ‘Oh look, it’s Hebrew!’ So it’s a big, big leap.”

The Alberta researchers published their findings in the journal Transactions of the Association for Computational Linguistics, in a paper titled “Decoding Anagrammed Texts Written in an Unknown Language and Script.”

The only copy of the manuscript is in the Beinecke collection at Yale University, but a Spanish publisher was given the permission to reprint it in 2016. Only 868 replicas were to be published.

The book may have had more pages in the past, and contains many unusual foldouts. There are also gaps in the original page numbering in the book, and the original order they were placed in could well be different from the way it is compiled today. Based on the illustrations, the Voynich manuscript is divided divided into six parts by scholars today, for the sake of convenience. These sections are herbal, astronomical, biological, cosmological, pharmaceutical and recipes.

Given the nature of the drawings in the book, it is thought by some to be a treatise on women’s health. Others believe the entire text is just an elaborate hoax or the work of an insane person. The truth of the matter is perhaps yet to be discovered.