On October 1, 2026, historian Benjamin Breen used AI to discover a new eyewitness account of the extinct dodo bird in a 1615 ship’s journal from the Dutch East India Company archives. Inspired by Breen’s work, software engineer Jesse Waites applied AI to millions of historical records, uncovering a forgotten meteorite report, three lost rhinos, and unrecorded volcanic eruptions across 400 years of documents, according to jessewaites.com.
Breen’s discovery came from the GLOBALISE project, which digitized millions of handwritten pages from the Dutch East India Company, active from 1602 to 1799. By searching these transcriptions with AI, Breen found detailed records of dodos and tortoises on Mauritius. Waites then expanded this approach by creating AI workflows that searched multiple historical mysteries in parallel, using chained AI models to filter and automate the extraction of significant findings from vast archives.
This application of AI to historical archives highlights the technology’s potential to reveal overlooked or lost information in large datasets. The Dutch East India Company’s extensive records, previously difficult to analyze at scale, have yielded new insights into extinct species and natural events. Such AI-driven archival research complements traditional historical methods and opens avenues for discoveries in other fields reliant on large, complex data collections.
Waites’ project, detailed on his website, demonstrates how AI agents can efficiently process centuries of handwritten documents, surfacing valuable historical data. The GLOBALISE project’s digitization effort and the AI tools applied mark a significant step in historical research methodologies, with the potential to uncover further unknown events and species documented in archival materials.