In a historic first, an international research team led by scientists from Maastricht University and Leiden University in the Netherlands has used modern AI tools to decode the rules of an ancient, mysterious Roman-era board game that has eluded researchers since the enigmatic object was found.
With input from researchers at Flinders University in South Australia, the Université Catholique de Louvain in Belgium, and the Roman Museum and restoration studio Restaura in Heerlen, the successful study offers researchers a potentially powerful new method for investigating unexplained historical mysteries using modern tools.
Ancient Mysterious Board Game Covered in Intersecting Lines
According to a statement announcing the game’s AI decoding, when researchers first spotted the unusual-looking piece of stone in what is now Heerlen, the Netherlands, they immediately noted its distinctive surface features, including a clear pattern of intersecting lines that hinted at a repeated, yet unexplained, use.
“Given its shape, size, and the type of stone, it is unlikely to have been a brick,” they write.
Instead, they note, the pattern of intersecting lines suggested that the stone block was “purposely made” to display the lines on its upper surface. Although the stone’s exact purpose remains a mystery, some have suggested that the ancient Romans, known to play board games, may have fashioned this object into a game surface.
“The stone shows a geometric pattern and visible wear that are consistent with sliding game pieces across the surface, which point strongly to repeated play rather than another purpose,” explained lead author, Dr Walter Crist, an archaeologist at Leiden University who specialises in ancient games.
Curious whether the markings constituted a previously unknown board game, and given that most games from this era were made from materials like wood that have not survived the intervening centuries, the research team said that his limestone piece “offered a rare opportunity to investigate ancient gameplay.”
Competitive Simulations ‘Point Strongly’ to a Classic ‘Trap Game’
Instead of trying to decode the ancient board game mystery by hand, the international team used AI. Of course, Dr. Crist cautioned, the pattern of uneven wear along the object’s carved lines “raises a key question about whether AI‑driven simulated play could reproduce that same pattern,”
After providing the software with high-definition images of the object’s surface, the team asked the AI to simulate hundreds of possible rule sets that could account for the repeated, intersecting patterns. Instead of creating the tools needed from scratch, they employed the LUDII AI-driven system.
During simulated play of a mysterious ancient board game, two AI agents were tasked with playing against each other, using the object as the game board. The agents were instructed with rule sets from previously documented ancient board games from Europe. These include the ancient game haretavl from Scandinavia and gioco dell’orso from Italy.
“We ran the simulations repeatedly, adjusting rules each time to see which movements would cause the same concentrated friction seen on the original stone,” says Dr Stephenson, from Flinders’ College of Science and Engineering.
After examining the results, Dr. Stephenson said the simulated gameplay “pointed strongly” to a type of classic strategy game known as a blocking game. The researcher said that in basic blocking games, participants attempt to trap their opponent’s pieces, thereby “preventing movement rather than capturing them.”
“This is the first time that AI-driven simulated play has been used together with archaeological methods to identify a board game,” Dr. Crist said.
“A Promising New Tool for Understanding Ancient Games”
The study findings are part of an ongoing study at Maastricht University, part of the Digital Ludeme Project in Europe. According to the study authors, the international effort used AI to produce “more reliable reconstructions of ancient games” that are historically and mathematically plausible.
According to Flinders University computer scientist Dr Matthew Stephenson, modern AI techniques like those used by his team can “bridge the gap between” the historical and computational studies of ancient board games.
“It shows how AI can contribute to our understanding of materials that would otherwise be difficult to interpret,” the researcher explained.
When discussing the implications of their findings, the researchers noted that the results suggest that the games played during the Roman era “may have a deeper history than previously documented.” The team also notes that the study “demonstrates the transformative potential of AI for archaeology.”
“The success of this approach suggests that many other mysterious artefacts may hold hidden stories waiting to be uncovered with the help of modern technology,” Dr Stephenson said. “It offers archaeologists a promising new tool for understanding ancient games that don’t resemble those known from surviving texts or artworks.”
The study “Ludus Coriovalli: using artificial intelligence-driven simulations to identify rules for an ancient board game” was published in the journal Antiquity.
Christopher Plain is a Science Fiction and Fantasy novelist and Head Science Writer at The Debrief. Follow and connect with him on X, learn about his books at plainfiction.com, or email him directly at christopher@thedebrief.org.