We build
AI for History
Supported by Schmidt Sciences
We're looking for collaborators — historians and AI researchers.
philip.torr@eng.ox.ac.ukIt is 2050. Artificial intelligence can now explain its reasoning, simulate complex societies, and help humanity learn from ten thousand years of recorded history. Historians and anthropologists work with AI collaborators that reconstruct lost evidence, test competing explanations, and reveal new insights into how cultures evolve and societies endure. This project lays the foundation for that future.
Humans have long been fascinated by the study of history and the lessons it might teach us. Historians and philosophers have repeatedly emphasized the importance of understanding the past as a key to self-knowledge and societal insight. While recent years have seen remarkable advances in AI scientists and AI scientific co-pilots, the humanities have, by comparison, been largely neglected. This raises a central question: if AI is to benefit all of society, how can we extend its capabilities into the interpretive and causal reasoning domains of the humanities? In particular: could we design an AI historian, a co-worker that assists historians in their reasoning, perhaps even approaching human-level interpretive skill? The aim of this project is to work closely with historians to develop such an AI collaborator, modelling their workflows and reasoning practices. The goal is to make this AI historian co-worker fully open source and publicly available, serving as a new kind of scientific instrument for historical scholarship worldwide.
This project is made possible by AI2050, a Schmidt Sciences initiative launched by Eric Schmidt to support researchers working on the hard problems that must be solved for AI to be hugely beneficial to society by the year 2050.
-
From dusty scans to one chart: how we traced industrial self-generation in Spain, 1958–1966
This guest article is by Mar Rubio-Varas , Professor of Economic History in the Department of Economics at the Universidad Pública de Navarra (UPNA) and a research fellow at the Institute for Advanced Research in Business and Economics (…
Read on Substack → -
Mapping Frankfurt’s Merchants of 1812 using AI
Working with an archive is, for the most part, an exercise in patience. A single volume can hold several hundred pages of blackletter type or handwriting, and the facts a researcher actually wants - who lived where, which goods were traded…
Read on Substack → -
Towards AI Economic History
The nature of low-level work in economic history is changing. Projects that once required armies of research assistants can now be completed by a single economic historian leveraging AI…
Read on Substack →
Workshops and talks organised or attended by members of the AI for History Team.
Upcoming
-
EPFL, Lausanne, Switzerland
Xiaoxi Luo, Niclas Griesshaber -
AI in the Historical Humanities and Social Sciences
University of Regensburg, Germany
Lorenz Hufe, Niclas Griesshaber -
Digital History Seminar
Berlin, Germany
Niclas Griesshaber, Lorenz Hufe
Past
-
Application of AI for the Digitalization and Analysis of Historical Data
Public University of Navarre (UPNA), Spain
Lorenz Hufe, Niclas Griesshaber -
Schmidt Sciences · August 2026
Philip Torr -
Accelerating Historical Research with AI Co-Historians
Computational Economic History Workshop, Vienna, Austria
Lorenz Hufe -
OxDSS × TVG: AI for Digital Scholarship Workshop
University of Oxford, UK 2nd edition
Niclas Griesshaber, Lorenz HufeOrganisers · ESRC-funded -
Accelerating Historical Research with AI Co-Historians
BIFOLD Workshop “AI to Accelerate Scientific Understanding”, Berlin, Germany
Lorenz Hufe -
AI for Digital Historical Scholarship
Aarhus University, Denmark
Lorenz Hufe, Niclas Griesshaber -
Warwick University, UK
Niclas Griesshaber -
OxDSS × TVG: AI for Digital Scholarship Workshop
University of Oxford, UK 1st edition
Niclas Griesshaber, Lorenz HufeOrganisers · ESRC-funded -
Machine Learning and Big Data in Economic History
University of Hohenheim, Germany
Niclas Griesshaber
AI for History Team
-
Philip Torr FRS FREngProfessor of EngineeringUniversity of OxfordSchmidt Sciences AI2050 Senior Fellow -
Christian Schroeder de WittPI of the Oxford Witt Lab (OWL)University of OxfordSchmidt Sciences AI2050 Early Career Fellow -
Niclas GriesshaberCo-PI NFDI4Memory Incubator FundsDPhil Student in Economic and Social HistoryUniversity of Oxford -
Lorenz HufeELLIS PhDFraunhofer HHIUniversity of Oxford -
Sebastian Oliver EckDPhil Student in Music HistoryUniversity of Oxford -
Pieter FrancoisProfessor of Cultural EvolutionDirector of the Computational Humanities LabUniversity of OxfordThe Alan Turing Institute -
Gavin GreifCo-PI NFDI4Memory Incubator FundsDPhil Student in Economic and Social HistoryUniversity of Oxford -
Zhenfei (Jeremy) YinPostdocUniversity of Oxford -
Zachary ShinnickVisiting StudentAustralian Institute for Machine Learning (AIML)The University of Adelaide
External Collaborators
-
Yao LuAssistant Professor in NLPUniversity College London -
Freda ShiAssistant Professor in Computer ScienceUniversity of Waterloo and Vector InstituteCanada CIFAR AI Chair -
Yixuan WangPhD Student in Computer ScienceUniversity of Waterloo and Vector Institute -
Xiaoxi LuoMaster's StudentUniversity of Waterloo and Vector Institute -
Rob IliffeProfessor of History of ScienceGeneral Editor of the Newton ProjectDirector of the Koch History CentreUniversity of Oxford -
Georgiana HedesanPostdocFaculty of HistoryUniversity of Oxford
The AI Historian project is funded by AI2050, a Schmidt Sciences initiative launched by Eric Schmidt to support researchers working on the hard problems that must be solved for AI to be hugely beneficial to society by the year 2050. The support is awarded through Philip Torr, a Schmidt Sciences AI2050 Senior Fellow.
Follow us: