What I generally mean by AI

As many others do, I generally use the initialism “AI” in my work to refer to the sociopolitical project by which the tech industry and its sponsors have, over the last few years in particular, sought to forcibly install, normalize, and make visible the presence of commercial technologies (themselves marketed under the generic banner of AI) within virtually all dimensions of life. “AI” refers to the processes and dynamics through which this project is produced and reproduced, as well as to the project’s broader historical and geographical contexts and to its very real—in no small part, physical—short-term and long-term consequences. In short, I’m using “AI” to refer, as Kate Crawford puts it, to “the massive industrial formation that includes politics, labor, culture, and capital,”¹ rather than to AI products in a purely technological sense ²

1. Kate Crawford, Atlas of AI: Power, Politics, and the Planetary Costs of Artificial Intelligence (Yale University Press, 2021), 9.
2. I’d indeed suggest that “AI in a purely technological sense” is an abstration that doesn’t exist in any meaningful way but that is nevertheless invoked in disingenuous observations that “there are many types of AI” and that “you use AI everyday without using it” and demands that we be specific in “what types of AI we’re talking about”)