For the first fourth dimension , researchers have identified the writer and championship of a papers that ’s been locked inside a charred roll for nearly 2,000 long time — without undress back a unmarried layer .
The scroll , PHerc . 172 , was recovered from the ruins of Herculaneum , the ancient Roman townsfolk buried by the ash and rubble of Mount Vesuvius in 79 CE . The coil is one of three Herculaneum scrolls that now reside at Oxford ’s Bodleian Libraries .
Thanks to high - resolution scans and some seriously clever car learning , scholars were capable to almost “ unwrap ” the papyrus and study the name inwardly : On Vices , by the Epicurean philosopher Philodemus .

A Herculaneum papyrus fragment at the National Library of Naples.Photo: Antonio Masiello/Getty Images
The treatise — its full name beingOn Vices and Their Opposite Virtues and In Whom They Are and About What , according toFine Books Magazine , is basically ancient ego - help , exploring how to live a pure living by avoiding frailty . Philodemus spell the work in the first century BCE and it is now being read for the first meter since it was buried in the devastating volcanic eruption nearly 2,000 yr ago .
The discovery — confirmed by multiple inquiry team — bring in the project ’s collaborators the $ 60,000 First Title Prize from the Vesuvius Challenge , an open - science rivalry that ’s been ca-ca ancient texts readable using AI .
In recent years , stilted news has been implemental in deciphering the ancient , carbonized scrolls from Herculaneum , a popish town buried by the extravasation of Mount Vesuvius in 79 . These scroll , first discovered in the 18th century in what is now known as the Villa of the Papyri , comprise one of the only surviving libraries from the Hellenic world .

Top to bottom: a reference photograph, a texture image, a network-generated prediction image, and a network-generated photorealistic rendering. Image: Parker et al., PLOS One 2019
Due to their fragile , charred condition , traditional ( read : manual ) methods of unrolling the curlicue often destroyed them . Now , investigator are using advanced imaging and machine learning to study these texts without ever give them .
How AI Is Deciphering lose Scrolls From the Roman Empire
The turning point came in 2015 , when scientists used X - ray imaging to read a different ancient scroll from En - Gedi , creating a 3D scan that could be near “ unwrapped . ” build on this , researchers at the University of Kentucky develop the Volume Cartographer , a syllabus that uses micro - CT imaging to observe the faint trace of carbon - based ink on the scroll .

Because the ink contains no alloy , unlike many ancient authorship materials , a neural connection had to be trained to recognise insidious patterns indicating ink on the carbonized papyrus . In 2019 , researchers successfullydemonstratedthis proficiency , set the stage for across-the-board applications .
These discovery culminate in theVesuvius Challenge , launched in 2023 to crowdsource the decoding of unopened scrolls . Participants use AI tools — in particular convolutional neural web and transformer models — to key and rebuild text within the scrolls . In October 2023 , the first word ( “ purple ” ) was take from an unopened scroll , earning a $ 40,000 pillage . The challenge go on , with prizes offered for decipher extra text and ameliorate the engineering science .
Brent Seales , a data processor scientist at the University of Kentucky and Centennial State - laminitis of the Vesuvius Challenge , toldThe Guardianthat the team ’s current chokepoint is cleaning , organizing , and enhancing the scan data so that researchers can in reality translate the carbonized ink as text .

Importantly , the digital unwrapping mental process is guided by human expertness . AI spotlight probable areas of ink on the ancient documents , but scholar interpret the convention to learn if they form coherent words or phrase . The finish is not only to recover lose philosophic texts , many of which are possibly by Epicurus or his follower , but also to establish a scalable system for digitalize and decoding ancient textbook — translate our understanding of the classical world .
AIAncient romecomputer visionPompeiiVolcanic eruptions
Daily Newsletter
Get the best technical school , skill , and culture news in your inbox daily .
intelligence from the future , render to your present .
You May Also Like













