TALOS AI4SSH at the University of Crete welcomed Dr Jacob Murel, Technical Lead of the Princeton University Logion Project, for a talk on the possibilities and challenges of using Artificial Intelligence in classical textual scholarship.

The Logion Project works to preserve and study premodern Greek texts, developing software that enables researchers to use language models for tasks such as gap-filling and the correction of scribal errors, without requiring programming expertise. The presentation explored how these tools can support specific philological tasks while also raising broader questions about the role and limits of AI in humanities research.

Dr Murel discussed the suitability of current language model architectures for philological tasks, approaches to evaluating models for ancient languages, and the practical needs that should guide the development of AI tools for philologists. Particular attention was given to the extent to which text reconstruction and textual criticism can be reliably supported or automated through AI.

The event also featured contributions from Dr Maria Papadopoulou, Director of the TALOS Lab, who presented the Lab’s research, tools, and digital resources in classical philology and the digital humanities. Rachel Milio, TALOS RT1 PhD Candidate, also presented her ongoing research on the ancient Attic orators.

The event brought together perspectives on AI, classical philology, and digital humanities, highlighting the potential of emerging technologies to support research while underscoring the importance of critical evaluation and domain-specific expertise.

Date: September 29, 2026
Time: 10:30–13:00 (GMT+3:00)
Venue: Onsite (TALOS Meeting Room, Γ3, Building Γ, KEME/UCRC, Rethymno) & Online (Zoom, passcode: 761267 )

Featured guest

Dr Jacob Murel is the technical lead of the Princeton University Logion Project, where he designs local AI systems to help scholars reconstruct premodern Greek texts. He has also worked on other digital humanities projects, including the Women Writers Project, Digital Humanities Quarterly, and the Open Islamicate Texts Initiative. His research focuses on using AI and computational tools to support scholars in textual criticism, paleography, and bibliography.

Talk
Title: Talos versus Medea: Bridging the tensions between AI and classical philology
Speaker: Dr Jacob Murel (Technical Lead, Princeton University Logion Project)

Abstract:
The past few years have seen a number of papers and research efforts applying AI developments to ancient languages. Among these is the Princeton University Logion Project. A core concern of the Logion Project is developing meaningful tools to assist in the preservation of the global archive of premodern Greek texts. To this end, the project has built a no-code software for LM inference with gap-filling and scribal error correction. These tools have demonstrated localized success in correcting scribal errors of premodern Greek. But ongoing work requires addressing a number of unresolved tensions between AI and philology: How suitable are current model architectures for philological tasks? How should ancient language LMs be evaluated for philological tasks? What practical concerns guide the development of AI tools that meaningfully assist philologists? How do we negotiate the role of automated processes in text reconstruction? This talk surveys the Logion Project’s ongoing efforts to address these issues in building meaningful AI applications to assist in classical textual criticism.