Artificial Intelligence Meets Archaeology: New Frontiers, Challenges, and the Launch of MAIA
AI is aiding archaeology by identifying artifacts, detecting sites, decoding ancient scripts, and reconstructing 3D models. The MAIA network unites experts to improve AI use and data sharing in the field.
Investigating Artificial Intelligence Applications in Archaeology
Artificial Intelligence (AI) has steadily become an integral tool across science and technology, and archaeology is no exception. Its use in archaeology has grown, with dedicated discussions on how AI can support and enhance the field.
AI proves especially useful when handling large datasets or performing tasks that are complex, highly specialised, and time-consuming. Currently, AI assists archaeology in several key areas:
- Artifact Identification: AI uses image recognition to classify and date pottery shards.
- Site Detection: Machine learning analyzes satellite imagery to uncover hidden archaeological sites.
- Language Decoding: AI helps decipher ancient scripts such as Linear B and cuneiform.
- 3D Reconstruction: AI rebuilds damaged statues or buildings from fragmented remains.
- Predictive Modelling: AI forecasts locations where undiscovered artifacts might be found, based on previous data.
These applications offer valuable insights into our archaeological heritage. However, several challenges require attention. Defining specific research questions suited for AI, ensuring access to quality data, addressing ethical and interpretive concerns, and establishing sustainable infrastructure are all critical issues. Each demands focused discussion and practical solutions.
One major hurdle is not algorithm development but the creation of training datasets. Archaeological data, while increasingly digital, often lacks the structure needed for effective AI use. Making data Findable, Accessible, Interoperable, and Reusable (FAIR) in machine-readable formats is essential. Adopting the FAIR Principles can significantly improve data usability in AI projects.
Introducing MAIA COST Action
In September 2024, the Managing Artificial Intelligence in Archaeology (MAIA) network launched, bringing together over 255 experts from 34 countries. This group includes digital and field archaeologists, data curators, computer scientists, museum professionals, and others working at the intersection of AI and archaeology.
MAIA focuses on building a shared understanding of AI’s role in archaeology. The network evaluates current AI applications, identifies limitations, and promotes open archaeological datasets for training AI systems. It also explores important research questions where AI can have meaningful impact, considering resource needs, ethical implications, transparency about biases, and sustainability.
MAIA will also communicate its findings broadly to make the challenges and opportunities clear to a wide audience. As Dr. Gabriele Gattiglia, Chair of MAIA, explains, the initiative offers a unique platform for diverse experts to collaborate on the future direction of AI in archaeology.
Kicking Off in Seville: Shaping the Future of AI in Archaeology
In May 2025, seventy participants from over 30 countries met in Seville for the MAIA Kick-off Meeting at the University of Seville’s Facultad de Geografía e Historia. This event officially launched MAIA’s activities and set the stage for upcoming projects.
In the months ahead, MAIA will organize training schools and working group meetings. These sessions will provide hands-on opportunities to map existing archaeological datasets and study AI applications worldwide. Young researchers, in particular, will gain training to create high-quality datasets suitable for AI tools.
Those interested in the evolving relationship between AI and archaeology can follow MAIA’s progress and explore AI courses that build relevant skills on Complete AI Training.
