20260401-004, AI, Data Centers, Electricity, and Water Presentation — original pdf

Backup

AI, Data Centers, Electricity, and Water Dave Sullivan, Ph.D. The University of Texas Center for Energy and Environmental Resources March 2026 3/26/2026 AI, Data Centers, Electricity, and Water 1 Ironically…. • WASHINGTON – On February 4, 2025, U.S. EPA Administrator Lee Zeldin announced the agency’s Powering the Great American Comeback Initiative, to achieve the agency’s mission while energizing the greatness of the American economy. • Pillar 4: Make the United States the Artificial Intelligence Capital of the World • “As we rapidly advance into this new age of AI, it is important that the United States lead the world in this field. Those looking to invest in and develop AI should be able to do so in the U.S., while we work to ensure data centers and related facilities can be powered and operated in a clean manner with American-made energy. Under President Trump’s leadership, I have no doubt that we will become the AI capital of the world,” said Administrator Zeldin. 3/26/2026 2 Ezra Klein, NYT, 3/4/2025 • “If you’ve been telling yourself this isn’t coming, I really think you need to question that. It’s not web3. It’s not vaporware. A lot of what we’re talking about is already here, right now. • I think we are on the cusp of an era in human history that is unlike any of the eras we have experienced before. And we’re not prepared in part because it’s not clear what it would mean to prepare. We don’t know what this will look like, what it will feel like. We don’t know how labor markets will respond. We don’t know which country is going to get there first. We don’t know what it will mean for war. We don’t know what it will mean for peace. • And while there is so much else going on in the world to cover, I do think there’s a good chance that, when we look back on this era in human history, A.I. will have been the thing that matters.” 3/26/2026 3 Ezra Klein… • “What’s fascinating to me is that this is the first revolutionary technology that is not funded by the Department of Defense, basically. And if you go back historically, over the last hundred years or so, nukes, space, the early days of the internet, the early days of the microprocessor, the early days of large-scale aviation, radar, the global positioning system — the list is very, very long — all of that tech fundamentally comes from Department of Defense money.” • “It took a long time to lay down electricity. It took a long time to build railroads. A.I. is going to come really quickly.” 3/26/2026 4 Ezra Klein… • “The reason I’m pushing on this is that we have seen this coming for a while. But as I look around, I do not see a lot of useful thinking here. And I grant that we don’t know the shape of it. But, at the very least, I would like to see some ideas on the shelf for what we should do if the disruptions are severe.” • “I have a lot of disagreements with the Trump administration, to say the least. But I do understand the people who say: Look, Elon Musk, David Sacks, Marc Andreessen, JD Vance — at the very highest levels of the Trump administration are people who have spent a lot of time thinking about A.I. and have considered very unusual thoughts about it.” 3/26/2026 5 3 books… Ben Buchanan, former special adviser for AI under Pres. Biden • “The Structure of Scientific Revolutions” by Thomas Kuhn, the book that coined the term “paradigm shift,” which is basically what we’ve been talking about throughout this whole conversation — a shift in technology and scientific understanding and its implications for society. • “Rise of the Machines” by Thomas Rid, which tells the story of how machines that were once the playthings of dorks like me became in the ’70s and the ’80s things of national security importance. • “A Swim in the Pond in the Rain” by George Saunders, taking great human short stories, and he’s giving a modern interpretation of what those stories mean. And when we talk about the kinds of cognitive tasks that are a long way off for machines, I hope, at some level, this is one of them — that there’s something fundamentally human that we alone can do. I’m not sure if that’s true, but I hope it’s true. 3/26/2026 6 Max Read on Substack.com, • “The AI Backlash Backlash”: • Since the release of ChatGPT in 2022, A.I. discourse has gone through at least two distinct cycles, at least in terms of how it’s been talked about and understood on social media, and, to a lesser extent, in the popular press. • First came the hype cycle, which lasted through most of 2023, during which the loudest voices were prophesying near-term chaos and global societal transformation in the face of unstoppable artificial intelligence, and Twitter was dominated by LinkedIn-style A.I. hustle-preneur morons claiming that “AI is going to nuke the bottom third of performers in jobs done on computers — even creative ones — in the next 24 months.” • When the much-hyped total economic transformation failed to arrive in the shortest of the promised timeframes--and when too many of the highly visible, actually existing A.I. implementations turned out to be worse-than-useless dogshit--a backlash cycle emerged, and the overwhelming A.I. hype on social media was matched by a strong anti-A.I. sentiment. For many people, A.I. became symbolic of a wayward and over-powerful tech industry, and many people who admitted or encouraged the use of A.I., especially in creative fields, were subject to intense criticism. 3/26/2026 7 Max Read… • The problem with using “A.I. insiders” as a guide to A.I. progress is that these insiders have been whispering stuff like this for years, and at some point, I think we need to admit that even the smartest people in the industry don’t have much credibility when it comes to timelines. • In my own capacity as an A.I. skeptic I’m desperate for A.I. to be demystified, and shed of its worrying reputation as a one-size-fits-all solution to problems that range from technical to societal. I think--I hope, at any rate--that widespread use may help accomplish that demystification: The more people use A.I. with some regularity, the more broad familiarity they’ll develop with its specific and consistent shortcomings; the more people understand how LLMs work from practical experience, the more they can recognize A.I. as an impressive but flawed technology, rather than as some inevitable and unchallengeable godhead. 3/26/2026 8 Max Read… • It’s an off-the-cuff interview and I wouldn’t want to make to much of it, but I thought it was interesting that Klein’s immediate illustrative example didn’t involve a way that A.I. might replace him, but a way it might replace people who work for him. Whatever else you can say about this technology, it has a way of making people think like bosses. 3/26/2026 9 Other recent tomes… • Dario Amodei – Machines of Loving Grace, How AI Could Transform the World for the Better • Makes strong arguments that AI improvements to health & human well- being. • “Mitigating climate change. Climate change will be felt much more strongly in the developing world, hampering its development. We can expect that AI will lead to improvements in technologies that slow or prevent climate change, from atmospheric carbon-removal and clean energy technology to lab-grown meat that reduces our reliance on carbon-intensive factory farming. Of course, as discussed above, technology isn’t the only thing restricting progress on climate change—as with all of the other issues discussed in this essay, human societal factors are important. But there’s good reason to think that AI-enhanced research will give us the means to make mitigating climate change far less costly and disruptive, rendering many of the objections moot and freeing up developing countries to make more economic progress.” 3/26/2026 10 https://www.datacentermap.com/usa/texas/ austin/ 3/26/2026 11 3/26/2026 12 3/26/2026 13 From a friend at the TX Adv. Comp. Ctr. • TACC: “national data, which you will find horrifying... new data centers for AI built this year alone will consume 650TW-hours per year. 650TW hours = 650 x 1012 watt hours, or 650 x 106 MWh.” • “This is like adding a new Texas in terms of power use every year” • More new data centers with more compact denser processing systems that consume less space. • Most emphasis on getting online ASAP as opposed to higher efficiency, or greener resource use. 3/26/2026 AI, Data Centers, Electricity, and Water 14 https://www.statista. com/statistics/12284 33/data-centers- worldwide-by- country/ 3/26/2026 15 Civil Engineering Source Mar. 4, 2024, “Engineers often need a lot of water to keep data centers cool” • Although the 24/7 operation of these data centers, which are often housed in massive buildings, makes the digital world possible, the industry also consumes enormous amounts of resources, especially water. That water is used in two primary ways: indirectly, to generate the electricity that the data centers need to operate, and directly, as a liquid coolant to dissipate the heat generated by the servers and other data center equipment. • In The Dalles, Oregon, for example, a lengthy legal battle ultimately revealed that Google data centers in the region consumed more than 355 million gal. of water in 2021 — an amount that had tripled since 2016 — representing more than one-quarter of the town’s annual water consumption, according to the February 22, 2023, article “Google’s water use is soaring in The Dalles, records show, with two more data centers to come” on The Oregonian’s OregonLive website. 3/26/2026 16 From Michael Enger at AE • “On water impact, data centers use a lot of power (which depending on type, uses large amounts of water, like nuclear) and use cooling systems that also use water. The amount per day will vary depending on generation supply, cooling systems and size. • …more efficient cooling systems are being tested and implemented.” 3/26/2026 17 ERCOT reported peak energy use by year, 2000 - 2023 • 16% increase 2021-2023 • 2nd biggest 2-year jump was 9.5% (2001-2003) 3/26/2026 18 3/26/2026 19 Texas Tribune 1/24/2025 • https://www.texastribune.org/2025/01/24/texas-data-center-boom-grid/ • “In Texas, the U.S. Energy Information Administration predicted that demand from large users — including but not limited to data centers — would grow by 60% this year, making up around 10% of the total forecast demand on the state’s main grid. • Large users requiring 5,496 MW of power have been approved by ERCOT to connect to the grid, according to a September report. The EIA expects that by the end of this year, ERCOT will have approved 9,500 MW in total large-user demand — a 73% increase. • Nationally, data centers are expected to consume between 11% and 12% of total U.S. power demand by 2030 — up from around 3% and 4% of demand today, according to an analysis by McKinsey.” 3/26/2026 20 ABJ, Feb. 20, 2025 • Data centers — massive hubs for computer processing power commonly called "the cloud" — are proliferating throughout the Austin suburbs. In fact, Austin is among the fastest-growing data center markets in the country, according to August data from CBRE based on the first half of 2024. The Austin Business Journal has identified more than a dozen in various stages of completion, in cities ranging from Hutto and Georgetown to Taylor and Uhland, and in counties that include Caldwell, Bastrop and Williamson. 3/26/2026 AI, Data Centers, Electricity, and Water 21 ABJ, Feb. 20, 2025, continued… • “But as the local market grows, developers and city leaders are keeping a handful of potential challenges top of mind: • the availability of electricity and water, • workforce development, • land availability and • the huge cost of building these centers. • They have drawn criticism for straining electricity grids because of the huge amounts of power they require. • While still uncommon, some are starting to use natural gas as an alternative energy source to help alleviate the issue.” 3/26/2026 AI, Data Centers, Electricity, and Water 22 As reported by Austin Business Journal (thanks Justin Sayers!) • Taylor – ABJ 8/9/24 “What might ultimately be a $1 billion data center development is planned for 52 acres on the east side of Taylor. It's another high-tech score for a rural Texas suburb growing fast thanks to Samsung's new factory, and it further solidifies the Austin area's status as an emerging hub for data centers.” • Georgetown – ABJ 9/25/24 “Austin-based Blueprint Projects won approval from the Georgetown City Council on Sept. 24 for a property tax abatement for its planned $160 million project on a 10-acre parcel at 1201 Westinghouse Road.” • Uhland – ABJ 10/8/24 “A company that builds master-planned campuses for data centers is eyeing what could be a large project in Caldwell County at the site where Micron Technology Inc. once looked at building a multibillion-dollar semiconductor factory. Denver-based Tract disclosed during an Oct. 2 meeting of the Uhland City Council that it's under contract for up to around 2,000 acres along Farm to Market Road 2720, about 30 miles south of Austin.” • Hutto – ABJ 12/6/24 “A planned $500 million data center at a 118-acre industrial park rising about 30 miles northeast of Austin has been given a green light by the Hutto City Council.” • Bastrop County – ABJ 12/9/24 “A Virginia-based data center operator has secured the go- ahead from the Bastrop County Commissioners Court for a property tax abatement to build what could be a billion-dollar data center campus east of Austin. 3/26/2026 AI, Data Centers, Electricity, and Water 23 ABJ 12/18/2024 – “What is happening in Austin is just a reflection of the data center industry overall, experts say. A huge demand for data center space fueled by the boom in artificial intelligence has caused companies to look outside typical markets like Northern Virginia, Dallas, Silicon Valley and others.” 3/26/2026 AI, Data Centers, Electricity, and Water 24 From Austin Energy • Many of the datacenters I have been talking to are also interested in carbon free supply, so the demand of new data center growth likely leads to additional renewable and battery development as the timing of new nuclear may be difficult, although likely preferred due to the baseload nature of output and their steady consumption profile. • Additional load growth would likely increase GHG emissions in the near term as capacity factors for less efficient generators should increase to help meet the load and balance intermittency. You could try to look at some of this through ERCOT generation stack and model assumptions based on previous year wind and solar profiles but that would likely take someone a few hours of modeling. • On water impact, data centers use a lot of power (which depending on type, uses large amounts of water, like nuclear) and use cooling systems that also use water. The amount per day will vary depending on generation supply, cooling systems and size. 3/26/2026 25 https://www.datacenterdynamics.com/en/new s/vistra-in-discussions-with-data-center- operators-to-increase-nuclear-power-output/, • US utility Vistra is looking to secure data center customers hungry for power. • In its Q3 earnings call, the Texan energy firm revealed that it is seeking agreements with data center developers in Texas and the PJM region to meet the industry’s growing energy demand. • Vistra indicated that it had entered into discussions with several high- profile data center operators to increase the output of its nuclear power projects. • Vistra operates four nuclear power plants, including the 2.4GW Comanche Peak nuclear power plant in Texas, which has an operating license until 2053. 3/26/2026 26 6 new data centers recently reported by ABJ in 2024 3/26/2026 27 ABJ Mar. 4, 2025 • “Since 2019, Austin's industrial inventory has grown to 91 million square feet, a staggering increase of 73.7%, the second-highest rate in the country behind only Savannah, Georgia, with a 99.7% growth rate, according to data from Jones Lang LaSalle Inc. • That includes 10 million square feet of speculative industrial space, and 1.5 million square feet of build-to-suit and owner-built product delivered in 2024.” 3/26/2026 AI, Data Centers, Electricity, and Water 28 Austin Industrial Market 2019 - 2024 3/26/2026 AI, Data Centers, Electricity, and Water 29 Questions…. • Where will data centers get electricity? Some may use their own natural gas-powered generators (bad), others may use small modular reactors (SMR = mini-nukes) (good), which are small and safe. Advances in battery storage technology are also being made (at UT). • How about water for cooling? Most use closed loop systems with H2O but also propylene glycol or other coolants, also closed loop. • Is anyone in charge of resource / environmental effects, or paying attention? 3/26/2026 AI, Data Centers, Electricity, and Water 30