CfP: AI through History, History through AI

June 15, 2026 – June 16, 2026 | University of Luxembourg

Eighth Conference on Digital Humanities and Digital History hosted by the Luxembourg Centre for Contemporary and Digital History (C²DH) at the University of Luxembourg | Co-Conveners: German Historical Institute Washington (GHI), Chair for Digital History at Humboldt-Universität zu Berlin, and Herder Institute for Historical Research on East Central Europe.

The Eighth Conference on Digital Humanities and Digital History will revolve around Artificial Intelligence in the historical disciplines. Generative AI has emerged as a transformative tool in historical research, serving as a method to answer historical questions, a means to streamline historians’ workflows, or even a subject of methodological and epistemological reflection itself. Even though its roots stretch back decades, generative AI only recently passed a critical threshold, bursting into a widespread applicability just a few years ago. Since the mid-20th century, the history of AI has been marked by cycles of advancement and stagnation. These fluctuations have often stemmed from tensions between symbolic (rule-based) and statistical (data-driven) approaches – two paradigms that, though historically antagonistic, are increasingly synthesized in contemporary AI systems. Since 2017 and the emergence of transformer-based systems, generative AI, although constantly “under construction,” has become an integral part of research practices. It brings profound opportunities and challenges. New capabilities come with fragility and new layers of vulnerabilities, accessibility with opacity, and widespread adoption with ethical, legal and power-related questions.  Different tendencies, cultures, and attitudes have shaped how historians respond to artificial intelligence in their work. While many historians quietly integrate AI into their workflows, others engage in methodological development and innovation or critical reflection on its capabilities and limitations, and still others voice profound concerns, particularly about its impact on interpretative and reflective dimensions of qualitative research. 

Topics

Our conference welcomes contributions across all these areas and particularly encourages work that fundamentally re-thinks or re-invents analytical or explorative approaches and workflows using AI. We equally welcome critical reflections on how AI quietly becomes woven into established research practices (often without explicit recognition). We provide a forum to examine these questions from three perspectives:

(1) Developing methods for analyzing historical sources

While implementations of AI algorithms for accessing and analyzing historical sources – including information retrieval and extraction – have so far often relied on “off-the-shelf” models, growing concerns about privacy, environmental impact, and appropriateness for the historical field have started to prompt a shift toward specialized models for historical research. Initiatives such as “Small Models for GLAM” at Hugging Face are taking important steps in facilitating the use of responsible AI by providing small models fine-tuned on high quality and open cultural heritage datasets. In this regard we want to ask:

Methodology: What specialized models would benefit the field, and what benchmark datasets or evaluation criteria should we develop to assess them? What ethical guidelines and best practices should govern AI method development? Where have AI methods based on current models failed in historical research? What can we learn from those failures? 

New possibilities: What new research questions become addressable through AI? What novel analytical methods emerge?

Collaboration and participation: How can historians, cultural heritage institutions, and computer scientists work together in the design of new methods for the historical field? How do we build sustainable, community-driven AI infrastructure?

(2) Creating everyday historical research workflows with AI

The integration of AI into tools used every day by historians – tools often made available by their institutions or imposed from outside, for example, by publishing houses – raises the question of AI as a “discreet digital practice” (Muller and Clavert, 2025). The increasing widespread but undocumented use of AI in historical research affects the very methods of history and its writing without us collectively understanding its consequences. In this sense, we welcome contributions that reflect on the integration of AI, particularly generative AI, into the “everyday” methodological repertoire of historians: 

Workflow transformation: How is AI re-shaping historical research workflows such as reading, note-taking, writing, citation management, and archival research? What new workflows are emerging, and what established practices are being displaced or altered?

Documentation and transparency: How can we make AI usage in historical research transparent and accountable? What are the legal aspects? What documentation protocols should historians adopt? How do we balance the benefits of AI tools with concerns about bias, privacy, legal aspects, environmental impact, and epistemic responsibility?

Education and training: How should we integrate critical AI literacy into education, ensuring that students understand not only AI’s capabilities and limitations, but also its material foundations and their histories (e.g. resource extraction, energy use, data infrastructures) and ethical implications (e.g. bias, surveillance, accountability)? What pedagogical approaches effectively build critical AI literacy?

(3) Conducting research on and about the history of AI

As historians, we can also contribute to major developments in AI. Research at the intersection of STEM and the humanities has become increasingly important since the 1970s, and historians have a specific contribution to make: providing historical perspectives, whether long-term (the human quest to simulate life in its own image, from the golem to the mechanical Turk) or short-term (the history of computer science itself). AI algorithms are deeply intertwined in complex socio-technical systems, where technology, institutions, and culture intersect.
We would welcome contributions that concentrate on any aspects of the history of AI, but would encourage those who focus on the history of simulation:  How have humans attempted to create artificial intelligence or simulate human cognition across different historical periods? What continuities and ruptures exist between pre-digital automata and contemporary AI systems?

Submission Guidelines

For the two-day conference, we invite you to submit proposals by March 1, 2026, for:

1. workshops for (hands-on) presentations of projects, tools, or skills (90 minutes),
2. or individual presentations (20 minutes)

Please submit a short CV and paper abstract of no more than 500 words to brigitte.dolenc@uni.lu by 1 March 2026. The conference will offer a dynamic, inclusive international forum. Although we favor in-person attendance of participants and presenters, facilities for hybrid participation will be provided with the aim of making the event as inclusive as possible.