This paper uncovers the “realities” of AI with an emphasis on the machine learning technologies that drive the new surveillance economy and its characteristic structures, social relations, and onto-epistemological conditions of possibility. I dwell on large language models (LLMs) because these systems for generating human-like text are the subject of heightening commercialization and debate, and I discuss them in relation to Ted Chiang’s The Lifecycle of Software Objects (2010). Though this novella follows a long line of fictional works that render "AI" in terms of an anthropomorphised technology that does not exist, its near-future storyworld is illuminating of today's data-driven systems for prediction and optimization, and their relation to the material conditions and "lifecycle" of writing subjects.

Prof Lauren M.E. Goodlad is Distinguished Professor of English and Comparative Literature at Rutgers, New Brunswick. She is the chair of Critical AI @ Rutgers and the editor-in-chief of Critical AI, an interdisciplinary journal published by Duke UP that will be launched in 2023. Goodlad's work on language models overlaps with a new project, Genres that Matter: The Ontological Work of Nineteenth-Century Fiction, and a recent (December 2020) co-edited special issue of MLQ, What Is and Isn't Changing: Critique after Postcritique.  She is the lead US PI for an NEH-funded international collaboration between Rutgers and ANU which has centered on data ethics and data ontologies.

ANU COVID safe event: social distancing / masks mandatory / COVID safety officer

Online: Zoom Meeting ID: 861 8419 0389 / Password: 2022