This platform showcases how artificial intelligence is accelerating archaeological discovery. Locating hidden sites and digitally reconstructing history — all without ever breaking ground.
Topics are the threads that run through every entity in this knowledge graph. They organise researchers, AI tools, papers, and discoveries around shared themes — revealing how the six case studies connect, where techniques overlap, and how AI is reshaping the pace of archaeology.
The measurable compression of research timescales achieved when AI enters archaeological workflows - from 275:1 in Herculaneum to 3600:1 in Greek inscriptions, each case study quantifies how decades of human effort collapse into days or months.
The process of finding and interpreting material evidence of past human activity, increasingly augmented by remote sensing and AI pattern recognition that reveal what the human eye cannot.
The emerging discipline where artificial intelligence accelerates archaeological discovery, compressing centuries of manual fieldwork and analysis into months - the central thesis binding all six case studies in this knowledge graph.
A complete, verifiable knowledge graph of AI-accelerated archaeological discovery — six case studies, seventeen entity types, and nineteen on-chain claims, structured and published to GEO Protocol.
View Research Challenges →Digital Antiquity is structured as a decentralized knowledge graph. Every discovery is anchored to a specific domain model that ensures historical data remains interoperable, verifiable, and permanent across the GEO protocol.
A specific archaeological problem resolved through AI. Each entry documents the historical question, breakthrough year, researchers, and discoveries. As the storytelling spine of this space, it expresses the time-compression effect at the heart of the project's thesis.
Documents the collaborative execution of a discovery. It maps how Researchers at specific Labs develop or apply AI Tools to accelerate breakthroughs, establishing a verifiable chain of human authorship.
Details the algorithmic methodology behind the tools. By linking an AI Tool (e.g., Ithaca) to its underlying AI Techniques (e.g., Deep Learning), it makes the computational "how" as transparent as the historical "what".
The terminal nodes of the graph. Every discovery must link to a Claim backed by a Paper and a Safety Approach, recorded as immutable truth on-chain.
Ten claims published on-chain in the Digital Antiquity knowledge graph — spanning discoveries that rewrote archaeological consensus, and design decisions that kept AI in the role of decision support rather than decision maker.
"While Ithaca alone achieves 62% accuracy when restoring damaged texts, the use of Ithaca by historians improved their accuracy from 25% to 72%, confirming the synergistic effect of this research tool."
"While it is tempting to use modern methods of deep learning, there are several arguments to not currently use such approaches for the proposed style-based date prediction on a very small data set."
"There is always a risk of getting the solution for the wrong cause. For example, the decision of an artificial neural network may be based on spurious correlations with the texture of the parchment."
"As a result, the text appearing in the images is not the imagined output of a machine learning model, but is instead directly tied to the underlying data in the CT scan."
"We set the ratio of positive to negative training images to 10:25 for a reasonable balance between precision and recall."
"We are now proving not just to ourselves but to the entire global community that the scrolls are readable."
"The northern Maya Lowlands were inhabited by at least 7 to 11 million people during the Late Classic period."
"The greater Angkor region represents the largest low-density agrarian city yet identified anywhere in the world."
"When historians use Ithaca as a decision-support system, their performance improves from 25% to 72% at the character level."
"Our results lend statistical support to the hypothesis that the Great Isaiah Scroll was produced by two scribes."