AI Breakthrough Redates Dead Sea Scrolls, Revealing Earlier Origins

0
505
Dead Sea Scroll - part of Isaiah Scroll (Isa 57:17 - 59:9), 1QIsa (Daniel.baranek Wikimedia commons public domain)
Dead Sea Scroll - part of Isaiah Scroll (Isa 57:17 - 59:9), 1QIsa (Daniel.baranek Wikimedia commons public domain)

An artificial intelligence program named Enoch has advanced the dating of many of the Dead Sea Scrolls by 50 to 150 years earlier than previously believed, according to a new study published in the journal PLOS One.

Newsroom (09- June- 2025, Gaudium Press) A groundbreaking study published in PLOS One has employed an artificial intelligence program named Enoch to significantly revise the dating of numerous Dead Sea Scrolls, pushing their origins 50 to 150 years earlier than previously estimated. This advancement, led by researchers at the University of Groningen, combines radiocarbon dating with AI-driven paleographic analysis to provide unprecedented precision in dating these ancient manuscripts.

The Dead Sea Scrolls, a collection of Jewish texts discovered in the Qumran Caves, have long been dated broadly between the third century BCE and the second century CE using traditional paleography, a method reliant on subjective handwriting analysis. However, the lack of securely dated Hebrew or Aramaic manuscripts from the late Hellenistic era has hindered precise chronologies. The Enoch AI model, developed by the Groningen team, addresses this gap by integrating radiocarbon dating with machine-learning-based handwriting analysis, achieving date estimates with an uncertainty of approximately ±30 years.

Enoch was trained on two key datasets: radiocarbon-dated biblical texts and detailed handwriting analyses of 135 scroll fragments spanning 200 BCE to 100 CE. By examining microscopic ink trace patterns—such as character curvature and shape—the AI correlated these features with empirically dated samples. The results suggest that many scrolls, including those written in Hasmonean and Herodian script styles, originated earlier than previously thought, with some dating to the early second century BCE or slightly before.

Among the study’s most significant findings are the revised dates for two biblical fragments, 4QDanielc (4Q114) and 4QQoheleta (4Q109), from the Book of Daniel and Ecclesiastes, respectively. Enoch’s analysis, supported by new radiocarbon data, places these fragments in the early 160s BCE and the third century BCE, aligning closely with the presumed composition periods of their respective texts. “This is the first tangible proof that portions of Daniel and Kohelet were penned contemporaneously with their presumed scribes,” said Professor Mladen Popović, director of the Qumran Institute at Groningen and the study’s lead author.

The Enoch model builds on BiNet, an earlier neural network developed by the Groningen team to detect handwritten ink traces. By refining this approach, Enoch has revealed that Hasmonean script may predate the previously assumed 150–50 BCE range, and Herodian fragments could emerge from the late second century BCE, suggesting concurrent rather than sequential script traditions. “The implications are profound,” noted Maruf Dhali, assistant professor of artificial intelligence at Groningen and co-author of the study. “With empirical evidence anchoring paleographic analysis, we can revisit questions about the circulation of biblical texts and their ties to ancient Judea’s cultural and political shifts.”

The study’s methodology overcomes the limitations of radiocarbon dating, a destructive process, by integrating non-invasive handwriting analysis. “Radiocarbon dating alone is insufficient,” Popović explained. “Enoch combines it with objective handwriting data to create a robust dating framework.” The AI’s predictions have aligned closely with radiocarbon results, offering a reliable tool for future research.

Buddy Christ, an associate curator at the Israel Antiquities Authority not involved in the study, hailed the approach as “a major leap forward” for dating unlabeled manuscripts. The Enoch method’s potential extends beyond the Dead Sea Scrolls, with applications for other ancient collections, such as Greek papyri or medieval codices. Popović’s team plans to make Enoch publicly accessible, allowing global scholars to upload digitized manuscripts for probabilistic date estimates.

This breakthrough rekindles interest in the Dead Sea Scrolls, illuminating their role in the Jewish and early Christian worlds. With precise dating now possible, these texts offer fresh insights into the literacy, script development, and textual transmission of the Hellenistic and early Roman eras, reshaping our understanding of ancient Judea’s cultural and theological landscape.

Raju Hasmukh with data sourced from

-Popović, M., Dhali, M. A., et al. (2025). “Artificial Intelligence-Based Dating of Dead Sea Scrolls Using Paleographic and Radiocarbon Analysis.” PLOS One.
– Ars Technica. “AI Helps Redate Dead Sea Scrolls to Earlier Periods.” Retrieved June 9, 2025.
– University of Groningen Qumran Institute. “Enoch AI Model: Advancements in Scroll Dating.” Press Release, 2025.
– Jerusalem Post June 2025

 

Related Images: