Artificial Intelligence is Used in Historical Research and Deciphering Ancient Texts

Researchers and institutions are beginning to use machine learning and artificial intelligence tools to examine, date, and restore damaged inscriptions, cuneiform tablets, and lost languages.

The Impact of Artificial Intelligence on Historical Science

Reconstructing the past is a very difficult process due to the extinction of languages and the decay of records. While this process was carried out solely with human expertise for decades, today researchers are turning to artificial intelligence to understand ancient tablets and documents.

Restoring Damaged Inscriptions

Experts state that completing and dating the missing parts of texts that have been damaged for thousands of years is critical for understanding daily life. Models such as Ithaca and Aeneas routinely determine the geographical location and date of Greek and Latin inscriptions.

Cuneiform and Digital Libraries

The Electronic Babylonian Library at LMU Munich in Germany is digitizing and bringing together cuneiform fragments in museum collections. This platform enables the matching of long-lost texts.

Newly Developed Models

Models such as Pythia, Ithaca, and Aeneas have been developed through the collaboration of Google DeepMind and academic institutions. Additionally, the DeepHadad model developed at KAUST simulates stone degradation to reconstruct damaged inscription surfaces.