On Artificial Intelligence as Assistive Technology in Archaeological Scholarship

In archaeological scholarship, the challenge of interpreting large amounts of fragmented evidence can be daunting. This paper will examine how artificial intelligence can assist in this process, mainly through examining visual evidence and classifying material evidence. Drawing from Trevor Paglen's argument of machine vision and Matteo Pasquinelli's history of classification in machines, this paper develops an analysis of two archaeological case studies: archaeological site detection based on satellite imagery in Mesopotamia and deep-learning classification of Tusayan White Ware pottery. The findings of the studies demonstrate that artificial intelligence is capable of expanding preliminary surveys as well as applying typologies consistently. At the same time, they also highlight the importance of human input, coming to the conclusion that artificial intelligence's potential and principal contribution lies in simplifying the archaeological process and directing human attention towards evidence requiring further interpretation, establishing a framework for understanding artificial intelligence as an assistive technology.