The University of Turku team of the DIGHT-net project organised an internal on-site workshop titled “AI Visualisations of Lotman Data” at Tallinn University on 3 December 2025. The aim of the workshop was to familiarise humanities scholars involved in the digitisation of Lotman data with the generation of data visualisations from large-scale metadata using generative Large Language Models. The hands-on component was intended to support joint experimentation and the collective exploration of the opportunities and challenges associated with using AI for visualisation.

At the outset, the workshop addressed the ethical and ecological implications of employing generative LLMs, as well as the types of data that may or may not be fed into such models. Simple, rapidly produced data visualisations can be useful during the digitisation process, illustrating the progress of the work and the effects of the descriptive categories applied to the materials. The experiment also provided project members with insights into the visualisation potential of the dataset, which will inform the planning of the future interface and prototype.

Mila Oiva and Merit Maran