AI Helps Decode Ancient Scrolls from Vesuvius Eruption
Researchers have successfully decoded text from ancient scrolls damaged by the eruption of Mount Vesuvius in 79 AD, using AI and machine learning techniques to penetrate layers of carbonized papyrus.
The scrolls from Herculaneum were buried under volcanic ash and heat, turning them into fragile, charred cylinders that had proven nearly impossible to unroll without destroying them. Traditional methods of physically unrolling these artifacts risked irreversible damage.
Instead, researchers employed advanced imaging techniques, including X-ray phase-contrast tomography, to scan the scrolls digitally. Machine learning algorithms were then used to distinguish ink from papyrus fibers and reconstruct text that had been hidden for nearly two millennia.
The techniques have already revealed passages of ancient Greek philosophical texts, offering potential new insights into Epicurean thought and other works that were lost when the eruption destroyed the library of Herculaneum.
The project represents a significant advancement in how technology can aid archaeological research, allowing scholars to access texts that would otherwise remain permanently illegible.