 |
|
|
|||||
As outrage spreads over energy-hungry data centers, politicians from President Donald Trump to local lawmakers have found rare bipartisan agreement over insisting that tech companies and not regular people must foot the bill for the exorbitant amount of electricity required for artificial intelligence. But that might be where the agreement ends. The price of powering data centers has become deeply intertwined with concerns over the cost of living, a dominant issue in the upcoming midterm elections that will determine control of Congress and governors offices. Some efforts to address the challenge may be coming too late, with energy costs on the rise. And even though tech giants are pledging to pay their fair share, there’s little consensus on what that means. Fair share is a pretty squishy term, and so its something that the industry likes to say because fair can mean different things to different people, said Ari Peskoe, who directs the Electricity Law Initiative at Harvard University. It’s a shift from last year, when states worked to woo massive data center projects and Trump directed his administration to do everything it could to get them electricity. Now there’s a backlash as towns fight data center projects and some utilities’ electricity bills have risen quickly. Anger over the issue has already had electoral consequences, with Democrats ousting two Republicans from Georgia’s utility regulatory commission in November. Voters are already connecting the experience of these facilities with their electricity costs and theyre going to increasingly want to know how government is going to navigate that, said Christopher Borick, a pollster and director of the Muhlenberg College Institute of Public Opinion. Energy race stokes concerns Data centers are sprouting across the U.S., as tech giants scramble to meet worldwide demand for chatbots and other generative AI products that require large amounts of computing power to train and operate. The buildings look like giant warehouses, some dwarfing the footprints of factories and stadiums. Some need more power than a small city, more than any utility has ever supplied to a single user, setting off a race to build more power plants. The demand for electricity can have a ripple effect that raises prices for everyone else. For example, if utilities build more power plants or transmission lines to serve them, the cost can be spread across all ratepayers. Concerns have dovetailed with broader questions about the cost of living, as well as fears about the powerful influence of tech companies and the impact of artificial intelligence. Trump continues to embrace artificial intelligence as a top economic and national security priority, although he seemed to acknowledge the backlash last month by posting on social media that data centers must pay their own way. At other times, he has brushed concerns aside, declaring that tech giants are building their own power plants, and Energy Secretary Chris Wright contends that data centers don’t inflate electricity bills disputing what consumer advocates and independent analysts say. States moving to regulate Some states and utilities have started to identify ways to get data centers to pay for their costs. They’ve required tech companies to buy electricity in long-term contracts, pay for the power plants and transmission upgrades they need and make big down payments in case they go belly-up or decide later they dont need as much electricity. But it might be more complicated than that. Those rules can’t fix the short-term problem of ravenous demand for electricity that is outpacing the speed of power plant construction, analysts say. What do you do when Big Tech, because of the very profitable nature of these data centers, can simply outbid grandma for power in the short run? Abe Silverman, a former utility regulatory lawyer and an energy researcher at Johns Hopkins University. That is, I think, going to be the real challenge. Some consumer advocates say tech companies’ fair share should also include the rising cost of electricity, grid equipment, or natural gas thats driven by their demand. In Oregon, which passed a law to protect smaller ratepayers from data centers’ power costs, a consumer advocacy group is jousting with the state’s largest utility, Portland General Electric, over its plan on how to do that. Meanwhile, consumer advocates in various states including Indiana, Georgia, and Missouri are warning that utilities could foist the cost of data center-driven buildouts onto regular ratepayers there. Pushback from lawmakers, governors Utilities have pledged to ensure electric rates are fair. But in some places it may be too late. For instance, in the mid-Atlantic grid territory from New Jersey to Illinois, consumer advocates and analysts have pegged billions of dollars in rate increases hitting the bills of regular Americans on data center demand. Legislation, meanwhile, is flooding into Congress and statehouses to regulate data centers. Democrats bills in Congress await Republican cosponsors, while lawmakers in a number of states are floating moratoriums on new data centers, drafting rules for regulators to shield regular ratepayers and targeting data center tax breaks and utility profits. Governors including some who worked to recruit data centers to their states are increasingly talking tough. Arizona Gov. Katie Hobbs, a Democrat running for reelection this year, wants to impose a penny-a-gallon water fee on data centers and get rid of the sales tax exemption there that most states offer data centers. She called it a $38 million corporate handout. Its time we make the booming data center industry work for the people of our state, rather than the other way around, she said in her state-of-the-state address. Blame for rising energy costs Energy costs are projected to keep rising in 2026. Republicans in Washington are pointing the finger at liberal state energy policies that favor renewable energy, suggesting they have driven up transmission costs and frayed supply by blocking fossil fuels. Americansare not paying higher prices because of data centers. Theres a perception there, and I get the perception, but its not actually true, said Wright, Trump’s energy secretary, at a news conference earlier this month. The struggle to assign blame was on display last week at a four-hour U.S. House subcommittee hearing with members of the Federal Energy Regulatory Commission. Republicans encouraged FERC members to speed up natural gas pipeline construction while Democrats defended renewable energy and urged FERC to limit utility profits and protect residential ratepayers from data center costs. FERC’s chair, Laura Swett, told Rep. Greg Landsman, D-Ohio, that she believes data center operators are willing to cover their costs and understand that its important to have community support. Thats not been our experience, Landsman responded, saying projects in his district are getting tax breaks, sidestepping community opposition and costing people money. Ultimately, I think we have to get to a place where they pay everything. Marc Levy, Associated Press
Category:
E-Commerce
After 25 years of obsessing over Mars, Elon Musk announced that SpaceX has shifted focus from invading the Red Planet to invading the Moon. He claims he will build a self-sustainable lunar metropolis in less than a decadea sharp contrast to his proposed Mars colony, which he says would now take at least 20 years. Both timelines are as fictional as Star Trek, but at least now his plan makes sense. It is a jarring plot twist from January 2025, when Musk dismissed the Moon as a “distraction.” Now, he says, the satellite is the “overriding priority” to secure civilization. Musk argues a lunar base is necessary because a “natural or man-made catastrophe” on Earth could cut off the supply lines a Mars colony would need to survive. Musk might actually be making sense this time. As Harvard physicist Avi Loeb points out, Musk is right to pivot. The Moon is closer, making it faster to get to, and it aligns with the geopolitical objectives of the United States (the government pays a lot of SpaceXs bills). It makes sense financially, opening the opportunity for the return of investment that may come from mining the lunar surface and orbiting asteroids, as well as his absurd plan to put one million AI satellites in orbit (made and launched from the Moon, no less). The financial aspect is the key. Really, its the whole end game. By choosing a target thats more accessibleand lucrativethan Mars, Musk is crafting a realistic illusion for investors and bull analysts. He needs to inflate the immediate financial expectations of SpaceX, so his company can get as much money as possible in its programmed 2026 IPO. Hard limits The unavoidable fact that forced him to pivot from Mars is, above everything, basic physical limitations. [The Moon] is much more practical to bring people back and forth,” Loeb told NewsNation. Musk described his “self-growing city” on X as a settlement that would be capable of expanding rapidly using local resources. It’s not something that that has ever been demonstrated. Still, Loeb argues that “the moon makes much more sense” before we attempt to leap into the deep void of the solar system. The physics of space travel don’t care about Musks marketing tweets. The Moon is simply a more forgiving target. Musk says that SpaceX can launch to the Moon every 10 days, allowing for rapid iteration; whereas Mars missions are shackled to planetary alignments that only occur every 26 months. [Rendering: SpaceX] The commute is also drastically different: a two-day hop versus a six-month deep-space haul exposed to radiation and all sorts of space dangers. As Quentin Parker, a professor of astrophysics at the University of Hong Kong, points out: “If you have some issue or emergency, youre a few days away from Earth. Youre months away if youre on Mars.” Thats the difference between a rescue mission and a lot of funerals. Whenever it is ready, Starships massive capacity to haul over 110 tons of cargo makes it a powerful workhorse to send everything Musk needs to build his fabled city as fast as possible. Base alpha and lava tubes Musk is calling his proposed self-sustaining lunar city “Moon Base Alpha,” a direct homage to the 1970s British-Italian science fiction television series Space: 1999. In the show, Moonbase Alpha is a high-tech scientific research center located in the lunar crater Plato. Musk citys hardware is radically different from the series’ shiny sets and spaceships. The workhorse for his plan is the Starship Human Landing System, a modified version of the regular Starship stripped of its heat shield and flight flaps since it will never need to re-enter Earths atmosphere. Instead of massive engines at the base, this ship uses smaller hull-mounted thrusters for touchdown to avoid blasting a crater into the landing zone and kicking up lethal dust. [Rendering: SpaceX] Once landed, a massive elevator would lower crews and cargo from the high-altitude cabin. The sheer volume of a Starship offers nearly 35,000 cubic feet of pressurized spacedwarfing the Apollo Lunar Modules cramped 160 cubic feetwhich allows for actual living quarters rather than just survival pods. SpaceX also envisions landing and tipping Starships horizontally and burying them under five meters of regolith to shield crews from cosmic radiation. Powering this buried city requires overcoming the lunar night, which lasts for two weeks of freezing darkness. For that, SpaceX will need nuclear reactors like those designed by Kilopower, 10-kilowatt fission units that can run continuously for a decaderather than relying solely on solar. The city will still need solar arrays, especially in the initial phase. NASA and SpaceX are developing Vertical Solar Array Technology (VSAT)32-foot-tall masts designed to capture sunlight that grazes the horizon at the lunar South Pole. To move around, astronauts wont just be walking; they will live inside pressurized rovers, essentially mobile habitats that allow them to explore for weeks without returning to base. But the ultimate goal is to go underground. Musks engineers are exploring another old idea for lunar bases: lava tubes. These massive natural tunnels were formed millions of years ago by ancient lunar volcanic flows. They offer ready-made protection with stable temperatures of around 63 degrees.Inside these subterranean cathedrals, SpaceX can build habitats using regolith-based 3D printing tech, like those imagined by 3D-printing construction companies Luyten and Icon. Giant rovers can also weave fibers from moon dust to construct inflatable module supports inside the lava tubes. [Rendering: Icon] However, the chasm between rendering and reality remains vast. SpaceX targets a 2026 orbital refueling test for Starshipa critical prerequisite for any lunar missiona date that has been pushed repeatedly and doesnt look like its going to happen. Aside from part of its elevator, the company hasnt delivered most of the hardware for Starship HLS. Its all on the drawing board, which is why NASA reopened the bids for the lunar lander in 2025 after SpaceX failed to progress on their promised milestones. Icons lunar construction system is still in R&D, and the Kilopower nuclear reactors, while promising, are still in the ground-testing phase with deployment unlikely before the 2030s. Musks less than a decade timeline assumes a flawless convergence of technologies that, right now, exist mostly on paper. Follow the money But we know Musks pivot isnt about practicality. Its about business and valuation. On February 2, SpaceX merged with Musks artificial intelligence venture, xAI, creating a corporate titan valued at $1.25 trillion. Sources indicate SpaceX is preparing for a mid-June 2026 IPO that could target a valuation as high as $1.5 trillion, potentially the largest listing in history. Investors love many fast catalysts, not multi-decade pipe dreams. Musk wants to dominate AI, and he knows he needs raw power so he plans to build orbital data centers to feed this AI obsession, allegedly bypassing the power and cooling constraints of terrestrial facilities. Musks narrative is selling the idea that the only way to put one million xAI servers in orbit is to exploit the Moons resources. Mine it for silicon and oxygen. Build the factories to make the servers. Build a magnetic cannon system to launch the servers into Earth orbit. The Moon becomes the construction site for the “vertically-integrated innovation engine” he promised during the xAI merger announcement. Musk appears to believe he can build this infrastructure before his life ends. It doesnt matter that multiple experts think thats impossible. It doesnt matter that hes basically proposing building a potential weapon of mass destructiona cannon satellite launcheron the Moon. It doesnt matter that he wants to put a one-million satellite orbital minefield around Earth. And it doesnt matter that thermodynamics makes his idea of cooling xAI servers in space extremely hard. [Rendering: SpaceX] Space is not “the cheapest place to put AI in 36 months or less,”as Musk has said. In fact, according to Lluc Palerm, research director for satellite at consulting firm Analysys Mason, Musks plan to make money out of space servers has the same magnitude of challenge as a Mars mission. Still, building a lunar city aligns perfectly with NASAs Artemis programwhich recently saw the SLS rocket lift off for the first time in 50 yearsand offers immediate revenue potential through government contracts that a distant Mars colony simply cannot match. The Bezos threat Which brings us to the second player in this Moon race: Musk is suddenly battling a competent Jeff Bezos. For years, SpaceXs factory in Texas stood unrivaled. But Blue Origin has finally started delivering. Its landing its New Glenn rocket and planning a Blue Moon Mark 1.5 lander that doesn’t require complex orbital refueling. Thats competition to Musk in terms of actual Earth dollars. Multiple sources have told Ars that Bezos has told his team to go all in on lunar exploration, writes Ars Technicas space editor Eric Berger This creates a genuine threat that Blue Origin could put humans on the lunar surface before Starship gets there. Bezos isn’t just playing catch-up either. He is building a parallel industrial machine. Blue Origin has successfully tested Blue Alchemist, a technology that melts lunar regolith to autonomously manufacture solar cells and transmission wire without needing any materials from Earth. The company has also launched Project Oasis, a mission to map lunar water ice and helium-3 using low-orbit satellites equipped with neutron spectrometers. To cut costs, Blue Origin is developing Project Jarvis, a reusable stainless-steel upper stage for the New Glenn rocket, mirroring the reusability of Starship. Bezoss vision is not to build underground cities on the Moon but to build O’Neill colonies, massive orbiting habitats. He sees the Moon not as a colony, but as the mine that will build them (again, a crazy long-term plan). So Musks pivot appears to be a calculated move to seize the commercial opportunity of the Moon before his rival does. Red Moon rises While Musk tweets about future cities and Bezos powers up, the most dangerous enemy for the United States space hegemony is on a fast collision course. China is the last part of Musks wild turn. The Asian giant is executing a concrete, century-long roadmap. On January 29, the China Aerospace Science and Technology Corporation (CASC) officially launched the Tiangong Kaiwu program, a massive national plan named after a 1637 Ming explorers encyclopedia to extend Chinese industrial dominance across the solar system. The plan treats space not as a scientific frontier, but as an economic zone. According to academic Wang Wei, who architected the proposal, the goal is to secure strategic minerals from near-Earth objects to fuel Earth’s sustainable development. According to the regimes official China Space Daily, among the 1.3 million asteroids in our solar system about 700 are relatively close to Earth and estimated to be worth over $100 trillion U.S. dollars each. Taking technical feasibility and cost-effectiveness into consideration, 122 of them are economically suitable for mining and use. This is a four-stage industrial conquest. CASCs roadmap dictates that by 2035, China will establish a lunar resource development system and begin mining near-Earth asteroids. By 2050, operations will expand to Mars and the main asteroid belt. The timeline extends to 2075 for the exploitation of Jupiter and Saturn, aiming for a fully operational solar-system-wide resource network by 2100. Unlike Musks private ventures, this is state policy: verify feasibility by 2030, build the supply chain by 2035, and dominate the market by mid-century. China has already poured concrete for this launchpad. They have successfully deployed the Queqiao satellite constellationincluding the recently launched Queqiao-2creating the worlds first permanent communication relay for the lunar far side. This network is the backbone for future autonomous mining operations. Furthermore, the plan includes a gigawatt-class space-based digital infrastructure that integrates cloud computing and space debris monitoring, essentially creating a space traffic control system that China intends to manage. While Musk is pivoting his company to catch up, Beijings machine has been methodically laying the tracks for decades. And it has a plan to catch up and surpass the U.S. While Musk may want us to believe that only he has the key to our future and that his new Moon plan is now what we all need, there are clearly other people on the planet who think otherwise. For now, what we really have is yet another erratic plot twist, a radical course change masquerading as The New Way to Save Humanity while he makes lots of money in the process.
Category:
E-Commerce
Part of a figure skaters job is to make their routine look as effortless and graceful as possible, as if theyre floating on ice and soaring into the air through sheer force of will. In reality, theyre often launching themselves multiple feet into the air with what amounts to sand bags on their feet; generating hundreds of pounds of centripetal force through rotations; and landing on a blade thats just 3/16 of an inch wide. At the 2026 Winter Games in Milan and Cortina, Italy, NBC is using an AI tool developed by a former MIT researcher to help audiences understand just how mind-boggling the feats of todays Olympic athletes are. Jerry Lu [Photo: Bryce Vickmark/MIT] Jerry Lu is a 2024 MIT graduate and the founder of OOFSports, a sports analytics company that uses AI to analyze program footage, document performance data in real time, and allow commentators to give viewers a more concrete understanding of athletes feats. At Milan Cortina, hes partnering with NBC Sports on its figure skating, snowboarding, and skiing programming, collecting data like the height of jumps, athletes speed, and their rotational paths. As figure skaters continue to break new ground in the sportlike landing more and more jumps with quadruple rotations (see American skater Ilia Malinins first-ever quad axel landed at the Olympics), Lus AI-powered tech can help make sense of their routines, moment by moment. [Video: Jerry Lu] A big ask from NBC Lus career in sports analytics began with his own interest in competitive swimming. During his undergraduate studies at the University of Virginia, he worked with the mathematician Ken Ono to develop a wearable device that let the schools swimmers analyze their strokes, which helped them to increase propulsion and reduce drag. Lu later served as a technical consultant for five swimmers who won medals at the Tokyo Olympics in 2020, followed by 16 medalists at the Paris Olympics in 2024. During his time at MIT in its dedicated sports lab, Lu began experimenting with sports analytics technology for other fields, including a program designed to help Australias BMX freestyle team optimize its strategy. Following the Paris Olympics, he says, NBC approached him directly to ask if he could create a data analytics system for figure skating in Milan Cortina. At that point, some of the artistic sports were missing this data-driven storytelling abilityif you watch hockey on TV, it looks slow, but if you watch it in person, it looks fast, Lu says. Similarly, he explains, if one were to watch American figure skater Amber Glenn perform a jump on screen, it might not look mind-blowingbut in person, she would be soaring unbelievably high in the air. NBC needed a way to bridge the gap between those two experiences. Building an AI model for the Olympics For Lu and his teamnone of whom are skatersthe first step toward building this tool was jumping on a call with former Olympic skaters and longtime NBC analysts Tara Lipinski and Johnny Weir. Unlike sports like swimming or track and field, the judging parameters for figure skating can involve quite a bit of grey area, meaning that Lus team needed a full run-down of what the judges would be looking for. They essentially taught us the sport, Lu says. They taught us exactly what they were looking for, what the judges are looking for, what, from their understanding, is a virtue, and whats a vice. We needed to come up with ways to quantify those and essentially give them the metrics with which they can compare across athletes. Making a tool for analyzing figure skating required a completely different system from swimming, Lu says. Whereas propulsion and drag were the two main variables in that sport, figure skating is all about the speed and rotation needed to complete complicated jumps. To calculate those metrics without wearables, his team trained an AI model to analyze program footage and identify a variety of rotational points on the athletes body, from their head to shoulders, elbows, hips, and ankles. Using those data points, the team then taught the model to categorize different jumps based on body positioninglike the toe loop, luxe, and axeland, further, to count the athletes total rotations in order to classify the jumps as a double, triple, or quad. By understanding exactly where the skater is at any given point, the AI model can calculate statistics like their speed when entering a jump, total jump height, jump exit speed, and the ground they cover across the rink; all crucial elements of their performance. These kinds of numbers can help commentators like Lipinsky and Weir paint a much more detailed picture for this years Olympic viewers. Will AI ever replace Olympic figure skating judges? This researcher says no Outside of his collaboration with NBC, Lu has turned his figure skating model into an app called OOFSkate, which lets skaters of any level film their routines and instantly understand their own stats. The app became an official partner of U.S. Figure Skating in December 2025. Lus next step is creating a version of this technology that not only tracks skaters routines, but also scores them. Right now, he already has a model in the works, which he plans on debuting some time during the skating off-season. Ultimately, he says, the model will be able to assist in evaluating technical performance on a select number of skills, but it will never replace human judgements on athletes artistic performance. Figure skating is this very unique blend of artistic and technical abilities, u says. The Olympics is all about athletes going higher, faster, strongerotherwise you dont deserve to be here. Figure skating has a part of that, which is that the bigger jumps get awarded bigger points, which is correctif you did a quad and I did a triple, you should get more points. But at the same time, this artistic element is also part of the thesis of figure skating.
Category:
E-Commerce
All news |
||||||||||||||||||
|
||||||||||||||||||