CAA ML AI SIG

Machine Learning & Artificial Intelligence in Archaeology

Recent trends in machine learning and artificial intelligence indicate that automated analysis and predictive modeling are becoming more and more important across different archaeological applications. We believe that this transformation will fundamentally reshape how we analyze and interpret archaeological data. In fact, AI tools have already become accessible for archaeological research. But the question is whether the trend is headed in the right direction? What are the threats and opportunities connected with the democratization of such powerful tools?

The main goal of the SIG is to start a broad discussion on this topic amongst international experts with the aim of establishing standards and guidelines that may serve as “good practice” for ML/AI usage in archaeology. Within “good practice” we would like to explore topics such as:

  • Why ML/AI is necessary for archaeology
  • What archaeological questions can be addressed via ML/AI and how
  • What could be efficient workflows for automated artifact classification, site detection, and pattern recognition
  • How to use ML/AI at different scales of analysis: from individual artifacts to landscape-level patterns
  • What archaeologists expect from ML/AI integration with other tools, such as: remote sensing, 3D scanning, GIS, and database systems
  • What is the ground-truth of AI-generated results (accuracy, interpretability, and validation of model outputs) and what is needed for training robust archaeological models
  • Ethics and transparency of archaeological AI applications (bias in training data, black-box algorithms, data privacy, intellectual property of AI-generated insights)
  • The problem of dataset quality and representativeness, how to ensure models generalize appropriately across different archaeological contexts

Additionally, this group will be directly involved in technical testing of newly available ML/AI tools that improve research efficiency and organize “hands-on” workshops.

The long-term goal of the SIG is to try to influence the development community to create tools that archaeologists actually need. The SIG would also be directly involved in the development of new AI applications for archaeologists that address real research questions. We encourage all interested CAA members to join the SIG. We hope that sharing our experiences with ML/AI will help to exchange ideas, establish ethical frameworks, and standardize approaches that rely on artificial intelligence.

SIG Coordinators:

If you are interested in joining this SIG, please contact the coordinators.