What is it about?
The secrets of ancient Mesopotamia are vanishing beneath modern development, but we have found a new way to uncover them. By training a deep learning model with decades-old spy satellite photos, known as CORONA, we have given our AI a glimpse into a world completely transformed over the last fifty years. The results are astonishing. Our model's accuracy in identifying archaeological sites soared to 90%, and it successfully located four brand-new sites previously invisible to human experts. This is a game-changing breakthrough for the study of landscapes with vanishing archaeological evidence, proving that AI can help us find history before or even after it's gone for good.
Featured Image
Photo by NASA Hubble Space Telescope on Unsplash
Why is it important?
The true novelty is that for the first time, an AI, in collaboration with human experts and using old data (CORONA imagery), discovered four new archaeological sites in Mesopotamia
Perspectives
A crucial aspect to emerge from this scientific experience is the importance of cross-disciplinary collaboration. This is a dialogue between different fields of knowledge, not just a sum of competencies.
Marco Roccetti
University of Bologna
Read the Original
This page is a summary of: AI-ming backwards: Vanishing archaeological landscapes in Mesopotamia and automatic detection of sites on CORONA imagery, PLOS One, August 2025, PLOS,
DOI: 10.1371/journal.pone.0330419.
You can read the full text:
Contributors
The following have contributed to this page
