AI @C²DH: Artificial Intelligence in the Service of History — and Under Its Critical Gaze - C²DH

At the C²DH, AI is not merely a tool: it is both a research instrument and an object of study. As a pioneering centre in digital history, the C²DH explores how artificial intelligence technologies are transforming historical practice while critically examining their epistemological and ethical implications through the lenses of digital hermeneutics.

‟ We already use AI in all kinds of ways”

Most researchers at the C²DH already use generative AI platforms in their work. From transcribing handwritten documents to analysing vast textual corpora, coding, creative writing, and even generating traditional folk embroidery patterns — the uses are as diverse as they are inventive.

This reality led to the creation of a dedicated AI Working Group and the development of an AI Manifesto that lays the groundwork for collective reflection on the major challenges AI poses to the practice of history.

C²DH AI Manifesto. The major challenges posed by AI to the practice of history

Read the AI Manifesto

Two Complementary Approaches: researching with AI, researching on AI

Researching with AI

The C²DH develops projects that harness AI as a methodological lever to renew access to sources and historical analysis. Among our initiatives:

LUXSCRIBENS: Using Handwritten Text Recognition (HTR) via Transkribus to transcribe Luxembourgish correspondence from the 19th and 20th centuries, including the development of a custom-trained model specifically for Luxembourgish — a language that lacked standardised spelling until 1984.
LETTERBOX — Leveraging Natural Language Processing (NLP) and digital history methods to make shell company networks visible and renew financial historiography.
AISTER — Developing human-centred solutions combining AI and citizen participation for the safeguarding of Ukrainian cultural heritage in emergency situations.
Voices from Auschwitz – Unlocking Collective Memory with the Multimodal Analysis of Survivor Testimonies.

Researching on AI

AI is not a neutral tool. It embeds worldviews, produces biases, and transforms our relationship with sources. The C²DH conducts critical research on:

AI as a producer of primary sources — How interactions with chatbots and AI-generated content are themselves becoming materials for understanding contemporary collective memory.
Epistemological implications — What transformations does source criticism face when dealing with probabilistic systems? How should AI use be documented to ensure transparency and reproducibility?
Ethical challenges — Attribution of creation, risks of writing standardisation, representation of minority languages, “instant history” and the public’s relationship with the complexity of historical narrative.