Monday, November 24, 2025

AI is transcribing the whole Cairo Geniza

ALGORITHM WATCH: National Library initiative aims to make all Cairo Genizah texts searchable worldwide. A new initiative using the National Library of Israel’s digital Hebrew manuscript database will enable automatic transcription of the entire Cairo Genizah, making the world’s largest trove of medieval Jewish texts searchable and accessible worldwide (Yogev Israeli, Ynet News).
Dr. Tzafra Siew, the National Library’s project manager for digital humanities, said MiDRASH is transforming the study of medieval manuscripts. By combining machine learning with the library’s digitized collections, she said, tasks that once required years of painstaking work can be done quickly and at scale. Researchers will be able to identify individual scribes, track how texts traveled between regions and ask new kinds of questions about the past. In practical terms, she said, hidden links between documents will come to light and many manuscripts that have never been deciphered will gain new meaning.
I have posted on the Friedberg Genizah Project, which has been around for some time, here and links, here, here, and here. As you can see, it has associations with Princeton and Tel Aviv Universities.

There is also a crowdsourcing Cairo Geniza digitization project that is associated with the Princeton project and with with the University of Pennsylvania.

For background on the Cairo Geniza and more PaleoJudaica posts see here and links. And for many PaleoJudaica posts noting Cairo Geniza Fragments of the Month in the Cambridge University Library's Taylor-Schechter Genizah Research Unit, start here and follow the links.

Visit PaleoJudaica daily for the latest news on ancient Judaism and the biblical world.