NYU Researchers Receive $10 Million Grant for AI Archaeology

The funding supports a new project using generative AI to reconstruct ancient artifacts.

Updated on Oct. 27, 2026 in Artificial Intelligence

Bold flat-color editorial illustration showing terracotta pottery shards on a stone surface, evoking the precision of archaeological research. — NYU researchers received a $10 million grant from Schmidt Sciences to develop generative AI tools for automating archaeological artifact reconstruction. AI Illustration. Upload story photo >

Live Poll

Should artificial intelligence be used to reconstruct physical artifacts from archaeological sites?

New York University researchers were awarded a $10 million grant from Schmidt Sciences in October 2026 to develop artificial intelligence for archaeological reconstruction. The project aims to automate the complex process of reassembling broken historical objects.

Why it matters

The initiative seeks to reduce the manual labor required for archaeological refitting while uncovering new insights into ancient tool manufacturing. It represents a significant step in applying modern generative models to preserve cultural history.

The project uses a generative AI method known as flow matching to estimate the spatial positioning of artifact fragments. This system integrates human feedback loops to iteratively refine and suggest plausible reconstructions of missing portions.

The players

New York University

This private research university serves as the institutional home for the interdisciplinary team of engineers and archaeologists working on the project.

Schmidt Sciences

This organization provided the $10 million funding as part of its Humanities and AI Virtual Institute grant program.

The details

Titled Fragmented Legacy: Turning Broken 3D Materials into Archaeological Interpretations, the project combines the expertise of NYU engineers and archaeologists. The research team intends to release their developed models and datasets under open-source licenses for public use.

The $10 million grant award was publicly announced in October 2026.

The Tech Race

This project aligns with the broader push to modernize archaeological workflows using generative AI models. It marks a shift away from traditional manual assembly methods by automating the identification of structural connections in ancient materials.

The project intends to make its AI models and research datasets available under open-source licenses for the public and academic community. This accessibility allows other researchers to utilize the reconstruction tools for their own historical projects.

The takeaway

Using generative AI to solve physical puzzles can significantly accelerate the speed at which historians analyze ancient artifacts. Researchers looking to integrate new tools should watch for the planned public release of these open-source datasets.