Researchers using artificial intelligence have uncovered surprising new information regarding the age of some of the world’s oldest biblical manuscripts.

According to CBN, A new study suggests that certain Dead Sea Scrolls are up to 150 years older than previously believed, placing them around 2,300 years old.

The Dead Sea Scrolls, discovered between 1946 and 1947 in the Qumran caves of the Judaean Desert by Bedouin shepherds, contain some of the earliest known versions of biblical texts, including Genesis, Exodus, and Isaiah. These scrolls, written on parchment, have long been studied for the insights they offer into ancient religious practice and scripture development.

Ancient scrolls meet modern algorithms

Up to now, scholars have relied on methods such as paleography—the analysis of ancient handwriting—and radiocarbon dating to determine the origin and age of these artifacts. These methods placed the scrolls’ creation between the third century B.C. and the first century A.D.

A recent breakthrough, however, came through the application of artificial intelligence. Researchers led by Mladen Popović, dean of the Faculty of Religion, Culture, and Society at the University of Groningen in the Netherlands, introduced a new AI-based tool known as Enoch.

Enoch was used to analyze handwriting samples from 27 fragments of parchment, comparing letter formations, script styles, and other features. When combined with existing radiocarbon data and knowledge of regional script trends, this analysis led to revised estimates of the scrolls' ages.

Scrolls may date back to the author’s era

According to the study published in the academic journal PLOS One, some fragments now appear to be about 2,300 years old, dating them to the third century B.C., significantly earlier than previous estimates. This includes texts from the book of Daniel.

This new timeline sets the Daniel scroll fragments in the same historical period as the book’s assumed author. That alignment could change scholarly assumptions about the relationship between the text and its historical context.

Popović described the AI tool as a profound development for historical studies. “With the Enoch tool we have opened a new door into the ancient world,” he said, adding that it allows researchers to investigate how biblical texts were physically produced and copied.

More than religious texts revealed

The Dead Sea Scrolls contain more than religious writings. While approximately 200 of the roughly 1,000 manuscripts discovered are biblical Old Testament texts, many include legal materials and community rules that shaped the societies from which they came.

Popović emphasized their broader historical value, noting that they serve as a “time machine” offering rare access to ancient reading, writing, and education practices. “They are physical, tangible evidence of a period of history that is crucial—whether you're Christian, Jewish, or don't believe at all,” he said.

He underscored that the scrolls remain essential for understanding the historical evolution of the Bible, which he called one of humanity’s most influential works.

AI gains support—but not total reliance

Christopher Rollston, a professor at George Washington University specializing in ancient scripts, praised the research as insightful but stopped short of calling it groundbreaking. “The results of this study are very interesting and presumably important, but not Earth-shattering,” he said.

Rollston pointed out that many conclusions drawn from the AI work support what paleographers such as Frank Moore Cross had theorized decades ago. This alignment suggests AI may be validating, rather than rewriting, long-held scholarly views.

Still, Rollston cautioned against depending solely on artificial intelligence for such research. “Enoch could and should never be the only tool in the toolbox,” he explained, citing the uniquely human variation of handwritten texts.

Blending technology and tradition

To Rollston, the best outcomes arise when human expertise and machine precision collaborate. “Machines can help isolate features of a script,” he said, “but the presence of a gifted palaeographer is at least as valuable as a machine-learning tool.”

Popović agrees that human judgment plays a critical role, even with advanced computational tools. His team’s fusion of machine learning with traditional methods may serve as a model for future studies in archaeology and manuscript conservation.

As the field continues to evolve, the combination of innovation and tradition could yield richer, more accurate narratives of the past, establishing a new standard for studying mankind’s earliest recorded texts.