Document Type
Conference Paper
Abstract
In 2012 Google's Eric Schmidt, in the infamous creepy line interview, claimed that Google (we) “know where you are, know where you've been and can more or less guess what you're thinking”. Schmidt, no stranger to saying the quiet part aloud, tellingly connected locational knowledge with intent and prediction, and in doing so placed location at the heart of Google's data-extractivist model. This paper argues that as the size and complexity of AI models increases cities are key sites of both training data extraction and model deployment, with technology companies competing to build models that offer machinic understandings of cities as both operational and lived spaces. These efforts to model cities as complex relational spaces are built on understandings of location that evolved from urban theorists, activists and artists such as Henri Lefebvre, Jane Jacobs and Locative Media artists. The paper proposes location data as uniquely characteristic of urban life and individual ways of being. Location contextualises myriad other data comprising the data deluge of the digitally mediated city, binding these data together, connecting bodies, places, and movements to their digital doubles. They qualify, relativise and add value to other data; opening digital actions, gestures and traces however idiosyncratic to algorithmic reasoning. The extractivist data models of the digital economy thus weaponise our relationships with place. Turning the relations and qualities that make cities worth living into parameters for training AI that colonises everyday life. Through tracing a near history of location as data, from locative media art that paralleled the emergence of consumer location aware devices in the early 2000s to the urban vision of Alphabet's (Google's) Sidewalk Labs and the marketplace for location data, the paper demonstrates that primacy of location data in the synthetic city and its role in the erosion of urban privacy.
Recommended Citation
McGarrigle, C. (2023). Where You Are.
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 4.0 International License.