Her project, which aims to reconstruct spatial behavior patterns in early cities of Central Asia using generative AI, has been awarded funding of more than 400,000 Euros. The grant will support research activities scheduled to begin in autumn of this year. The work will be carried out in collaboration with project partners: Puay Peng Ho at the National University of Singapore and Ye Zhang at Tsinghua University in China; the precise institutional affiliation for the project is currently being finalized.
”Everyone is focused on AI at the moment, but I think that a much more crucial game-changer is how the humanities ride the waves of these technological transformations in knowledge production", Annie Chan says. ”I am looking forward to learning through this project—besides, of course, the archaeology of ancient cities— about the ways we can ask bolder and more impactful questions by strengthening how the humanities interface with AI.
This Volkswagen funding initiative is very timely because it calls for this ‚opening-up‘ strategy and mindset we ought to embrace facing the challenges of our times. In our project, we are innovating techniques to better reconstruct ancient urban environments, but as a means to deepen the investigation of what is fundamentally human—how the built spaces we created have shaped our evolution.”
New insights into the social dynamics of early urban life
The project investigates how generative AI can expand the interpretive toolkit of archaeologists studying ancient built environments. Its aim is to uncover patterns of movement, social behavior, and spatial use in ancient cities by reconstructing features that are hidden or only partially preserved in the archaeological record. In doing so, the project seeks to generate new insights into the social dynamics of early urban life.
At the core of the research is the use of machine-learning models, including Generative Adversarial Networks (GANs), which will be trained on archaeological site plans and conditioned with multimodal data—such as drawings, textual descriptions, and photographs—from six well-documented early urban sites in Central Asia.
These sites are characterized by long settlement histories and diverse architectural forms. The models will generate simulated and synthetic settlement layouts in order to reconstruct architectural features and spatial configurations that have been lost over time.
Using configurational and network-analytical methods, the research team will visualize and simulate patterns of movement and social behavior within the generated layouts, allowing them to identify culture-specific spatial dynamics at these ancient sites. After validation and refinement by subject specialists, the models will be applied to generate building data for additional sites, demonstrating the scalability of the approach.
Beyond exploring the potential of AI in archaeological research, the project will also engage local stakeholders and communities at the investigated sites. This participatory approach is intended to ensure that AI-based methods and findings remain accessible and beneficial for cultural heritage preservation, while also contributing to the development of technological expertise in the regions concerned.
Supporting Research in the Humanities and Cultural Studies
The Volkswagen Foundation’s funding initiative “Open Up – New Research Spaces for the Humanities and Cultural Studies” supports research teams that explore new and largely uncharted areas within the humanities, cultural studies, and theoretical social sciences. The program specifically promotes highly innovative and exploratory projects that approach complex research questions from multiple perspectives.
Projects may run for up to 1.5 years and can receive funding of up to 300,000 Euros for collaborations involving two partners, or up to 400,000 Euros for projects involving three partners. A certain degree of project risk is explicitly encouraged. At the same time, proposals must demonstrate that they genuinely open up new fields of inquiry rather than expanding already established areas of research.