Scholarly Editing and AI: Machine Predicted Text and Herculaneum Papyri

Groundbreaking advancements in digital restoration have enabled the virtual unwrapping of ancient carbonized parchment scrolls, revealing previously inaccessible texts. A notable achievement was the 2019 application of machine learning to decode Greek characters from a Herculaneum papyrus fragment using carbon ink tomography. This study proposes a theoretical model for critical editions of 'true-born virtual' texts, emphasizing the need to document AI's role and ensure transparency in data science processes. The research advocates for a new virtual edition model that integrates humanities and scientific rigor, addressing the challenges posed by AI-dependent text editing.

Brusuelas, J. H. (2021). Scholarly Editing and AI: Machine Predicted Text and Herculaneum Papyri. Magazén, (1). https://doi.org/10.30687/mag/2724-3923/2021/03/002