Hi folks, we're looking for a couple more papers to round out this session. Please submit your abstracts by next Thursday, 10/24. Thank you! Ryan CFP American Association of Geographers 2025 (Detroit, MI) Organizers: Ryan Burns (University of Washington Bothell and University of Calgary) and Katherine Harrington (National Science Foundation) The advent of generative Artificial Intelligence (AI) has undeniable environmental impacts. In training Chat GPT-3, its parent company, OpenAI, used as much water as an average family uses in almost 2 years (DeGeurin 2023) for server cooling and electricity generation, and each 100 words it generates consumes an average of 3 bottles of water (Verma and Tan 2024). Quantifying carbon emissions produced by AI is difficult, but some estimates show that training a transformer model such as GPT-3 emits as much carbon as a single airplane flight from San Francisco to New York City (Patterson et al 2021), and the cumulative use of such models emits as much as 25 times this amount each year (Chien et al 2023). As of 2022, data centers accounted for about 2% of all global energy use, and the raw amount of energy consumed by data centers doubles roughly every 4-8 years (International Energy Agency 2022), leading some tech companies to secure sole access to energy infrastructure such as rebooting Three Mile Island nuclear reactor (Wade and Bass 2024). These staggering figures underscore the potentially dire impacts of AI on the environment in general (Hogan 2024) and, particularly, undermine global attemps to reduce carbon emissions under “net zero” schemes. To invoke Jevon’s paradox, the remarkable growth of generative AI and its future evolutions could offset any gains from efficiency innovations. The effects of AI's environmental implications, of course, go beyond the natural-physical spheres, as communities can be dispossessed of their access to water (Tironi and Albornoz, forthcoming), indigenous groups’ livelihoods can be disrupted (Lally, Kay, and Thatcher 2019), and labor laws and safety norms can be undermined at sites where key technological component materials like silicon and lithium are mined (Arboleda 2020). Taken together, this means that scientists, policymakers, humanists, and climate activists need to take a broad perspective to contend with AI’s growing environmental costs. At the same time, there are many who maintain that AI is the key to addressing the massively complex “wicked problem” of global climate change (Masterson 2024). Many of the most important tasks rely heavily on mathematical and computational decision-making, such as climate modeling and prediction and optimizing resource management, which AI is better suited to do than human cognition alone. There is also the possibility that AI can reduce greenhouse gas emissions of energy production and consumption (Kaack et al 2022) - particularly within urban settings (Townsend 2018) – and some have even suggested that futher research can reduce the water requirements of AI (Li et al 2023). This panel seeks to facilitate the growing, increasingly broad-ranging dialogue on the socio-natural, political-ecological, and environmental implications of AI. We seek to advance geographers’ understanding of the new spatial configurations of politics, resources, knowledge, and economy with which emerging AI trajectories are aligning. And we hope to deepen this conversation in order to work in small ways toward climate justice in this more deeply compromised context. Among other related questions, we encourage submissions of paper abstracts asking: * Is averting the climate crisis, let alone achieving global climate justice, even possible in a world with increasingly resource-intensive AI models and services? How do we conceive of climate justice within this emergent context? * In what ways are the energy, water, minerals, and manufacturing needs of AI reconfiguring spatial political-economies? * How are researchers measuring, qualifying, and theorizing the socio-natural, political-ecological, and environmental impacts of AI development and use? * Do our theorizations of AI’s ecological implications sufficiently adapt to different use cases (e.g., smart cities or medical imaging), models (e.g., generative AI or climate modeling), and sociospatial context? * How is the growth of AI shifting spatial labor relations, particularly with regard to its need for minerals, manufacturing, domestic and caring labor in the tech industry, maintenance and repair, and workplace norms? * What are the intersections of AI’s environmental implications with other forms of sociospatial difference such as gender, race, indigeneity, social class, state-sanctioned labor regimes, and dis/ability? * How do we reimagine and rearticulate glimmers of hope in this anxiety-riddled contemporary context? How is resistance being enacted? Where are the limits of AI’s ecological harms? How do we collectively act in strategic ways to end AI’s contributions to climate genocide and the unfolding climate crisis? We encourage submissions of 250-word paper abstracts by October 17, 2024 sent to Ryan Burns (ryan.burns1@ucalgary.ca) and Katherine Harrington (katherine_harrington@alumni.brown.edu). We will notify of acceptances within a week, and please then be prepared to submit your abstract to the AAG's system by October 31. Works Cited Arboleda, Martín. Planetary Mine: Territories of Extraction under Late Capitalism. Brooklyn, New York: Verso Books, 2020. Chien, A., L. Lin, H. Nguyen, V. Rao, T. Sharma, & R. Wijayawardana. 2023. Reducing the Carbon Impact of Generative AI Inference (today and in 2035). In Proceedings of the 2nd Workshop on Sustainable Computer Systems (1-7). https://dl.acm.org/doi/pdf/10.1145/3604930.3605705 (Last accessed 5/3/2024). DeGeurin, M. 2023. 'Thirsty' AI: Training ChatGPT Required Enough Water to Fill a Nuclear Reactor's Cooling Tower, Study Finds. Gizmodo. https://gizmodo.com/chatgpt-ai-water-185000-gallons-training-nuclear-1850324... (Last accessed 5/3/2024). Hogan, M. 2024. AI Is a Hot Mess. Training the Archive, 33–54. International Energy Agency. 2022. Electricity 2024: Analysis and forecast to 2026 (Revised version). https://iea.blob.core.windows.net/assets/6b2fd954-2017-408e-bf08-952fdd62118... (Last accessed 5/3/2024). Kaack, L.H., P. Donti, E. Strubell, G. Kamiya, F. Creutzig & D. Rolnick. 2022. Aligning Artificial Intelligence with Climate Change Mitigation. Nature Climate Change 12, 518–527. https://doi.org/10.1038/s41558-022-01377-7 (Last accessed 5/5/2024). Lally, N., K. Kay, & J. Thatcher. Computational Parasites and Hydropower: A Political Ecology of Bitcoin Mining on the Columbia River. Environment and Planning E: Nature and Space, 2019, 1–21. <https://doi.org/10.1177/2514848619867608> https://doi.org/10.1177/2514848619867608. Li, P., J. Yang, M. Islam, S. Ren. 2023. Making AI Less "Thirsty": Uncovering and Addressing the Secret Water Footprint of AI Models. arXiv preprint. https://arxiv.org/abs/2304.03271 (last accessed 9/20/2024) Masterson, Victoria. 2024. 9 Ways AI is Helping Tackle Climate Change. World Economic Forum: Emerging Technologies. https://www.weforum.org/agenda/2024/02/ai-combat-climate-change/ (last accessed 5/5/2024). Patterson, D., J. Gonzalez, Q. Le, C. Liang, L.-M. Munguia, D. Rothchild, D. So, M. Texier, and J. Dean. 2021. Carbon Emissions and Large Neural Network Training. ArXiv. https://arxiv.org/pdf/2104.10350 (last accessed 5/3/2024). Tironi, M., & C. Albornoz. Forthcoming. Divergent futures in a damaged territory: The rise of data centers and water controversies in Santiago de Chile. Journal of Urban Technology. Verma, P., & S. Tan. 2024. A bottle of water per email: the hidden environmental costs of using AI chatbots. Washington Post. https://www.washingtonpost.com/technology/2024/09/18/energy-ai-use-electrici... (last accessed 9/20/2024). Wade, W., and D. Bass. 2024. Microsoft AI Needs So Much Power It's Restarting Site of US Nuclear Meltdown. Bloomberg News. https://www.bloomberg.com/news/articles/2024-09-20/microsoft-s-ai-power-need... (last accessed 9/20/2024). Best, Ryan -- Ryan Burns, PhD, FRCGS University of Calgary University of Washington - Bothell Associate Editor, GeoJournal Vice-chair, Digital Geographies Specialty Group of the AAG https://burnsr77.github.io/ https://cal.com/rlburns (to request a meeting) He/his pronouns The University of Calgary is located in southern Alberta on the traditional territories of the the Treaty 7 peoples--Blackfoot Confederacy (comprised of the Siksika, the Piikani, and the Kainai First Nations), the Tsuut’ina First Nation, and the Stoney Nakoda (including Chiniki, Bearspaw, and Goodstoney First Nations). The City of Calgary is also home to the Metis Nation, Region 3. I am privileged, grateful, and indebted to be allowed to work within these lands.