[with apologies for cross-posting] Dear colleagues, We warmly invite you to submit an abstract to our panel at the next 4S Annual Meeting in Toronto, October 7-10, 2026. P199: From Bubbles to Rubbles: Ruins in the Age of AI Submission link: https://www.4sonline.org/call_for_submissions_toronto.php Panel ID: 199 Deadline: April 30, 2026 Organizers: Benedetta Catanzariti (University of Edinburgh) and Ludovico Rella (Durham University) Panel description: Since the public release of ChatGPT, generative AI has entered an accelerated phase of sociotechnical consolidation, marked by the rapid expansion of computational infrastructures, large-scale capital investment, complex debt and financing arrangements, and intensifying present and anticipated energy demands. The scale of these investments, the reliance on leveraged capital, and the increasingly circular financial and infrastructural entanglements among ostensibly competing actors have unfolded alongside growing public skepticism regarding the material and social returns of AI projects. This conjuncture has led commentators to frame generative AI in terms of speculative excess and the possibility of an AI ‘bubble.’ From an STS perspective, however, the question is not only whether such a bubble will burst, but how AI infrastructures reinforce or precipitate particular political-economic arrangements, ecological burdens, and forms of technological dependence. What kinds of infrastructural ‘ruins’ - data centers, energy systems, extractive supply chains, financial instruments - might be left behind if large-scale AI investment contracts? Conversely, if expansion continues, what enduring social, energy, and environmental footprints will be stabilized? Which actors across hardware and software industries, finance, states, and local communities will benefit or bear the costs of these trajectories? And how might the dynamics of the AI bubble be mobilized as an analytic lens for imagining alternative technological paths, governance regimes, and infrastructural futures beyond extractive and growth-oriented models? This panel invites contributions that engage with these questions and critically examine generative AI as a contested sociotechnical formation. This panel seeks abstracts (no more than 250 words) on topics including (but not limited to): * Ruins of cloud infrastructures and data centers * Energetic, ecological, and social afterlives of AI systems * Material residues (chips, minerals, water, labor) * Infrastructural failures and breakdown * Resistance, refusal, and post-cloud futures * Methods for studying decay, maintenance, and infrastructural ruins We look forward to receiving your submission. Please feel free to get in touch should you have any question! All the best, Benedetta & Ludovico