# Plundered Earth LLMS index: [llms.txt](/en/llms.txt) --- In Santiago, Chile, on days after a good rain, the smog washes away and reveals a stunning view of the Andes mountains. The Andes run all the way up and down this thin sliver of a country, from Patagonia at its southern tip over 2,600 miles north—twice the length of Miami to Boston—to Chile’s upper border. In this part of the country, the urban landscapes of the capital turn into an endless expanse of desert, which, save parts of Antarctica, is the driest place on earth. The mountains come into sharper focus, their colors more vibrant under the naked sun. Before Chile was Chile, the Atacama Desert, as it is named, was home to many Indigenous groups. Where others may have seen a punishing barren landscape, they coaxed the desert into a home, tapping the water and minerals deep under the earth to grow crops, raise livestock, and perform ancestral rituals. Then the Spanish arrived. They cut up the region into administrative units with borders. To this day Indigenous elders still warn in oral histories of a repression that grew so violent the Spanish cut off the tongues of anyone who dared to continue speaking their native language. From there, the empire established the country’s relationship with the rest of the world: It would provide raw resources —land, water, energy, minerals—to strengthen the political and economic agendas of other nations. Today nearly 60 percent of Chile’s exports are minerals, primarily found in the Atacama Desert, chiefly copper, a highly conductive metal used in all kinds of electronics, and more recently lithium, the essential ingredient for lithium-ion batteries. Those and other resource exports drive the country’s economy. In Santiago everyone knows someone who lives by the rhythms of the mining industry: During their work “shifts” they live in the north; during their days off, they come back to the capital. The country has struggled to build any other industry to serve as its economic engine. Long after the Spanish empire ended, the US infamously played a key role in ensuring this trajectory. In the 1950s and ’60s, as developmental economics took root in Chile and Uruguay, favoring strong government regulation and an inward focus on industrialization as the path to maturing a developing economy, American multinationals who had made billions from their holdings in Chilean mines began to chafe against the growing state taxes and restrictions. The US government subsequently embarked on a quest to refashion Chile’s economic policies to be more favorable to American business interests, launching a program in 1956 called “the Chile Project” to educate a hundred Chilean students at the University of Chicago under the intellectual tutelage of American economist Milton Friedman. Friedman was a towering figure in economics who would go on to receive a Nobel Prize in 1976 and whose ideas could best be summed up by the title of his influential 1970 op-ed in The New York Times: “The Social Responsibility of Business Is to Increase Its Profits.” Friedman stood for everything that developmental economics did not: zero government regulation, unfettered freedom for profit-driven companies, a path to the economic maturation of developing countries defined by facing outward, such as through liberal exports. As Naomi Klein details in her 2007 international bestseller The Shock Doctrine, the Chile Project was not education but indoctrination. At the University of Chicago, Chilean students—and later students from other countries in Latin America—were explicitly taught to critique their country’s economic policies and the fatal flaws of Latin American developmentalism. As each batch of graduates, known as “the Chicago Boys,” returned to Santiago, Friedman’s neoliberal ideas percolated through Chile’s intellectual elite, until they became part of ruling ideology. In 1973, Chile’s left-wing, democratically elected president was overthrown by his military general Augusto Pinochet in a coup d’état under conditions fomented in part by the CIA. The coup became the start of Pinochet’s nearly two decades of brutal dictatorship and a new neoliberal economic agenda: the dictatorship appointed the Chicago Boys to write its economic policies. Under Pinochet’s rule, Chile privatized nearly everything—education, health care, the pension system, even water. The strategy produced economic growth; it also fueled stunning inequality. Chile is among the most unequal countries in the world today, with nearly a quarter of the country’s income concentrated among a few powerful families in the 1 percent. Having never meaningfully industrialized, it also remains tethered to the extraction economy that makes it relevant to higher geopolitical powers. And so, as the AI boom arrived, Chile would become ground zero for a new scale of extractivism, as the supplier of the industry’s insatiable appetite for raw resources, not just its copper and lithium in the north but also its land, water, and energy resources for a growing crop of data centers in the Santiago metropolitan region. In May 2024, the government proudly announced that the country would welcome twenty-eight new data centers, on top of its existing twenty-two, over the coming years, bringing in $2.6 billion of foreign investment. While the government’s stance that its role as a resource provider to technology development represents progress for the nation, Chile has also become home to some of the fiercest resistance globally against this narrative. Communities across the country are vehemently fighting against the dispossession of their land, water, and other resources in service of Global North visions that do not include or benefit them. Through street protests and courtroom battles, their efforts have stalled company projects and caught the government’s attention. They have also inspired people in other countries to rise up in solidarity. Martín Tironi Rodó, a professor at Catholic University in Santiago and director of the Chilean research think tank Futures of Artificial Intelligence Research, which studies AI through an interdisciplinary and Latin American lens, summarizes the sentiment that I heard repeatedly from these communities as I traveled up and down the country to meet them. The central question these movements are asking is how to imagine a different path for AI development not rooted in extraction, he says. “If we are going to develop this technology in the same way that we used to, we are going to devastate the earth.” --- “Digital” technologies do not just exist digitally. The “cloud” does not in fact take the ethereal form its name invokes. To train and serve up AI models requires tangible, physical data centers. And to train and run the kinds of generative AI models that OpenAI pioneered requires more and larger data centers than ever before. Before AI, data centers were already growing and sprawling. They were once small and distributed enough to be tucked away in urban environments, a few shelves of computers hidden in a back-office closet or a few dozen racks in a repurposed building. In the aughts, tech giants began trending in a different direction, consolidating all of their computing infrastructure into massive warehouses of servers in rural communities. The data center world became divided: There were the hyperscalers and there was everyone else. The four largest hyperscalers—Google, Microsoft, Amazon, Meta—now spend more money building data centers each year than almost all the others, relatively unknown developers like Equinix and Digital Realty, combined. It’s difficult to imagine what a hyperscale data center looks like if you’ve never seen one. Mél Hogan, an associate professor at Queen’s University in Canada who studies AI, infrastructure, and the environment, used to use football fields to describe them when she began to write about them roughly a decade ago. “Now football fields don’t even come close to the imaginaries of the required size,” she says. Hyperscalers call their data centers “campuses”—large tracts of land that rival the largest Ivy League universities, with several massive buildings densely packed with racks on racks of computers. Those computers emanate an unseemly amount of heat, like a laboring laptop a million times over. To keep them from overheating, the buildings also have massive cooling systems —large fans, air conditioners, or systems that evaporate water to cool down the servers. The equipment all together creates a cacophony of humming, whirring, and crackling that can—especially in underdeveloped communities—be heard for miles, twenty-four hours a day, creating a relentless and body-warping source of noise pollution. Now developers use a new word to distinguish the scale of what’s coming in the post-ChatGPT AI era: megacampus. The word refers not just to the land area but to the sheer amount of energy that will be required to run them. A rack of GPUs consumes three times more power than a rack of other computer chips. And it’s not just the training of the generative AI models that is costly, it is also serving them: According to the International Energy Agency, each ChatGPT query is estimated to need on average about ten times more electricity than a typical search on Google. Until recently, the largest data centers were designed to be around 150-megawatt facilities, meaning they could consume as much energy annually as close to 122,000 American households. Developers and utility companies are now preparing for AI megacampuses that could soon require 1,000 to 2,000 megawatts of power. A single one could use as much energy per year as around one and a half to three and a half San Franciscos. Few places on the planet exist that can produce and deliver that much energy to any single location. Developers are working with utility companies around the world to build more power plants and expand the roster of options. After the last decade of flatlined energy demand in the US, a Goldman Sachs analysis described the sudden new wave of data centers as driving “the kind of electricity growth that hasn’t been seen in a generation.” Utility companies are now delaying the retirement of gas and coal plants and the transition to renewable energy; Microsoft restarted Three Mile Island, a nuclear plant near Middletown, Pennsylvania, that had a partial nuclear meltdown in the late 1970s, the worst commercial nuclear accident in US history. By 2030, at the current pace of growth, data centers are projected to use 8 percent of the country’s power, compared with 3 percent in 2022; AI computing globally could use more energy than all of India, the world’s third-largest electricity consumer. This scale—the mega-hyperscale—has created startling environmental consequences. And yet, in the very same moment, corporate obfuscation of that impact has reached new heights. Since Emma Strubell’s paper and Gebru’s citation in “Stochastic Parrots,” tech giants have hidden away even more of their models’ technical details, making it exceedingly hard to estimate and track their carbon footprints. At the same time, those companies have amped up their public and policymaker influence campaigns with powerful counternarratives: Data centers will grow so efficient, their impact will stop being a problem; generative AI will unlock new climate innovation; AGI will solve climate change once and for all. While the last claim is impossible to prove, the first two are highly misleading, says Sasha Luccioni, a research scientist and climate lead at opensource AI firm Hugging Face. The second is especially pernicious: There are indeed many AI technologies, as cataloged by the initiative turned nonprofit Climate Change AI, that can accelerate sustainability, but rarely are they ever generative AI technologies. “What you need for climate are supervised learning models or anomaly detection models or even statistical time series models,” says Luccioni, who is also a founding member of the initiative. All of these models are previous generations of AI technologies—primarily machine learning tools —that are small and energy efficient, and in some cases could even run on a powerful laptop. “Generative AI has a very disproportionate energy and carbon footprint with very little in terms of positive stuff for the environment,” she adds. Luccioni says her past collaborators within closed-off companies no longer receive approval from their employers to cowrite papers with her about AI’s environmental impact. Instead, she has worked with external collaborators and academics like Strubell, whose research she was first inspired by. In one paper, together with Hugging Face machine learning and society lead Yacine Jernite, the two measured the carbon footprint of running open-source generative AI models as a proxy to what closed companies are building. They found that producing one thousand pieces of text from generative models used as much energy on average as what it would take to charge the standard smartphone nearly four times. Generating one thousand images used on average as much energy as 242 full smartphone charges; in other words, every AI-generated image could consume enough energy to charge a smartphone by roughly 25 percent. Luccioni’s and Strubell’s papers are among the few that still provide quantifiable measures of the carbon behind generative AI models. It’s a constant uphill battle. At one point, Luccioni says she reached out to over five hundred authors of the most recent machine learning papers to request basic information about their model training. “I barely got any answers,” she says. “People were just not even responding or saying that this is confidential information.” Even as hyperscalers have spoken loudly in public about the sustainability of their computing infrastructure, executives at Microsoft have admitted internally that the intermittent availability of renewable energy just doesn’t cut it when data centers need to operate 24/7. Nonstop operation is considered so crucial that Google, Amazon, and most recently Microsoft now build their campuses in threes to have a backup for the backup in case any facility goes down. During Hurricane Irma in Florida and Hurricane Harvey in Texas, even as millions of people lost power, some hospitals evacuated patients, and hundreds of thousands of homes and businesses faced damage and destruction, the data centers in those areas continued to hum along—so well that the displaced families of one facility’s employees moved into it for the duration of the natural disaster. The land and energy required to support these megacampuses are but two inputs in the global supply chain of data center expansion. So, too, is the extraordinary volume of minerals including copper and lithium needed to build the hardware—computers, cables, power lines, batteries, backup generators— and the extraordinary volume of potable—yes, potable—water often needed to cool the servers. (The water must be clean enough to avoid clogging pipes and bacterial growth; potable water meets that standard.) According to an estimate from researchers at the University of California, Riverside, surging AI demand could consume 1.1 trillion to 1.7 trillion gallons of fresh water globally a year by 2027, or half the water annually consumed in the UK. Those effects will not be felt evenly. Another study found that in the US, one-fifth of data centers were already drawing that water before the generative AI boom from moderately or highly stressed watersheds due to drought or other factors. And in Global South countries like Chile, it’s often the most vulnerable communities who have borne the brunt of these accelerating economies of extraction. As more and more communities have watched data centers affect their lives, a growing number have pushed back vehemently against their unfettered development. In response, data center developers have grown more sophisticated with tactics to maintain business as usual: They’ve entered communities in secrecy under shell companies; they’ve donated to community programs to dampen resistance; they’ve made promises to cities about the sustainability of their facilities before walking them back one by one after projects have broken ground and are more difficult to reverse. In one case in Virginia, a group of residents protesting against several massive data centers was shocked to uncover an email from a lawyer to a developer suggesting to place them under surveillance. “We need a mole or 2 in this group,” the lawyer wrote. --- From the moment they committed to the idea of scaling, OpenAI sought to secure an unprecedented amount of computing infrastructure. “In AI, whoever has the biggest computer gets the most benefits,” Brockman told me in 2019. So with Microsoft, Altman developed a plan for how the tech giant would meet OpenAI’s exponentially growing hunger for compute. The two companies would work together to design and deliver dozens of supercomputers for research— data centers that would need to be equipped with Nvidia chips for training various AI models in the course of OpenAI’s explorations. Crucially, Microsoft would also build a series of ever-larger, ever-more-powerful supercomputers for training each subsequent generation of OpenAI’s models. Altman began referring to the series of supercomputers as “phases,” at one point showing employees a slide to illustrate the size of each phase, with Phase 5, the last one planned, breathtakingly larger than all the others. The supercomputer that OpenAI had trained GPT-3 on was Phase 1. Equipped with ten thousand V100s, the facility had been built in West Des Moines, Iowa, which Microsoft first entered in 2012. It had made nice with city officials through a “staggering” sum of tax payments, according to the then mayor, helping the city make major improvements to its public infrastructure. Over more than a decade, the company also invested some $2.5 million in local community programs, most of them nonprofits. Microsoft code-named the Phase 1 supercomputer Odyssey; OpenAI called it Owl, after a convention it had started early on of naming each of the Research division’s compute clusters alphabetically after animals. When it ran out of all twenty-six letters, it would switch to naming them after periodic elements, ordered by atomic number. Phase 2 was also in Iowa and used to train GPT-4. It had started with the eighteen thousand A100s referenced in the 2021 research road map and ultimately grew to around twenty-five thousand by the end of the model’s training process. To Microsoft, Phase 2 was Telemachus, named after the son of Odysseus in Greek mythology. To OpenAI, it was Raven. Phase 3 shifted to Arizona. With its cheap land, good tax breaks, and close proximity to California, Arizona had fast become a preferred hub for data center development among all of the cloud providers. After carefully cultivating favorable relationships with the governments of two underdeveloped cities right outside Phoenix, Microsoft had bought three tracts of land in 2018 and 2019 and similarly donated to various community organizations to quiet any objections from residents. The tracts had a combined land area of nearly 600 acres, or more than 450 football fields. (Microsoft would expand that area with another 283 acres in 2024.) Each tract would host a new data center campus that would serve Azure customers alongside OpenAI in the cloud region “West US 3.” Phase 3—code-named Inglewood at Microsoft and Whale at OpenAI— would cost several billion to build and house hundreds of thousands of Nvidia H100s, the generation after A100s, to train what OpenAI at the time believed it would likely brand as GPT-4.5 and GPT-5. For Phase 4, planned in Wisconsin, Altman expected costs to hit $10 billion using Nvidia’s latest B100s, yet another generation after H100s. That was a staggering amount, considering the most expensive hyperscale data centers then hovered around $1 to $2 billion. He didn’t plan to stop there, casually floating the idea of the $100 billion supercomputer for Phase 5. After the blowout success of ChatGPT, Altman tempered his expectations. With so many chips locked up in serving ChatGPT to customers, Microsoft was struggling to acquire chips fast enough to keep up with finishing Phase 3’s development. Whale split into three separate clusters— Beluga, Narwhal, and Orca—with plans to complete the build-out sometime in 2024. No one within Microsoft or OpenAI even knew whether Phase 5 was technically possible. In Microsoft and OpenAI’s design plans, The Information later reported, the $100 billion facility could need as much as 5,000 megawatts, nearly matching the average power demands of all of New York City. While the plan didn’t seem the most financially sound as a business investment, money wasn’t the main bottleneck. It was energy. “We’re running out of land and power,” an OpenAI employee says. Within both companies, it was understood that Phase 5 would only become possible with some amount of innovation. Either Microsoft and OpenAI would need to split the supercomputer into multiple campuses to distribute the energy demands and figure out how to train an AI model across distant locations, or, as Altman sometimes liked to say, the problem would solve itself with a future breakthrough in nuclear fusion. At times he would give OpenAI employees optimistic updates about Helion Energy, the nuclear fusion startup that represented his largest personal investment and for which Microsoft had already committed to buying power from once a plant, with a target generation of 50 megawatts, was up and running. The Wall Street Journal would later report that OpenAI and Helion were also in talks to strike a deal, from which Altman had recused himself. Altman and other executives never brought up the data centers’ environmental toll in company-wide meetings. As OpenAI trained GPT-4 in Iowa, the state was two years into a drought. The Associated Press later reported that during a single month of the model’s training, Microsoft’s data centers had consumed around 11.5 million gallons, or 6 percent, of the district’s water. GPT-4 had trained there for three months. (A Microsoft spokesperson said the company is working to increase its water efficiency by 40 percent above its 2022 baseline and to replenish more water than it consumes across its global operations by 2030, with a focus on the water-stressed regions where it works.) Arizona, too, faces a severe water crisis. In 2022, as Microsoft laid the groundwork for Phase 3, a study in Nature Climate Change found that the Southwestern US had been facing the worst drought it had seen in over a thousand years. That drought, combined with severe mismanagement, has drained the Colorado River, which Arizona and six other states rely on for fresh water, to dangerously low levels. Without drastic action, the river could cease to flow. The shortage compounds a power crisis, as climate change has slammed the region with relentless record-breaking temperatures and families have cranked up their air-conditioning. The region relies in part on hydropower from the Hoover Dam and water-cooled nuclear power plants. In other words, it needs water to produce more energy. In 2023, the Phoenix metro area hit multiple new heat records as well as the worst year for heat-related fatalities, which surged at least 30 percent from 2022 to over six hundred dead. “All things,” says Tom Buschatzke, the director of the Arizona Department of Water Resources, “are converging in a challenging direction.” What Altman did bring up was his impatience. In March 2024, after sleeping through the early years of the generative AI race, Meta would come out swinging with aggressive new plans to have 350,000 H100s up and running as part of an even larger infrastructure build-out to support its sudden burst of generative AI investments. This was more GPUs than OpenAI had at its disposal. Altman wasn’t happy. Microsoft was being too slow, he felt, and it was costing OpenAI its competitive advantage. --- When Sonia Ramos was a child, she witnessed an accident that would shape the rest of her life. She was born into a mining family in Chile. Her father worked for an American copper company; she grew up among the children of the other workers. In 1957, a part of the Chuquicamata mine collapsed, killing several people and injuring dozens more. Though her father was spared, she remembers watching the wretchedness of the aftermath unfold around her: affected families spiraling into abject poverty, children wasting away from hunger. Four decades later, as Ramos began to protest mining, becoming one of the most active and outspoken Indigenous voices in Chile shining a light on its social, cultural, and environmental destruction, she would remember the lessons she learned in the tragedy: The mining industry is a system, and that system, left to its own devices, will seek profit at any cost. “The worker doesn’t exist,” she says. None of the victims received any ceremony or commemoration; none of their families received compensation. “In that place, there is no humanity.” Chile is the world’s largest producer of copper, accounting for a quarter of the global supply. Since the beginning, copper mining has reshaped not just the land but the societies that rely on it. Sometimes the effects are visible: Chuquicamata today is the largest open-pit copper mine in the world, a gaping wound in the earth that explosives regularly deepen. That displaced rock, piled up in towering mountains that monumentalize the cavities they came from, is slowly burying the remains of a town that was abandoned after the copper mining began to swallow it. The mining has also drained the region of water to process the copper. At one point a foreign multinational corporation consumed so much water it depleted an entire basin in a nearby salt flat, or salar, destroying its rich ecosystem. Less visible are the trails of arsenic that the industry leaves in the air and water, which has increased rates of cancer throughout the north of the country, and the ways mining has restructured Indigenous life and sowed divisions among different communities. With their lands depleted of water and minerals, the Atacameños, the name that ties together all of the distinct Indigenous groups who share this region, can no longer sustain themselves by growing their own crops or raising their own livestock. The shift has plunged their towns into deep poverty. Crime has risen along with depression, alcoholism, and delinquency. They don’t have enough food, running water, proper health care, or educational resources, having seen little benefit from the billions in profits that their land has generated for someone else. Instead, many are forced to work for the very industry that seized their territories and receive health care from the small clinics it sponsors. Where there was once greater unity among them, the Indigenous groups now squabble over diminishing resources. Lithium is a more recent discovery there, stumbled upon by an American company in the 1960s as it searched for the water it needed for copper mining. When it drilled into the salares, it found high concentrations of lithium floating in an oily brine beneath the surface, opening up a new front of extraction and accelerating the depletion of more ecosystems. Today Chile produces roughly a third of the world’s lithium, second only to Australia. The material is primarily extracted out of the Salar de Atacama, the largest salt flat in the country, by pumping its brine out into shimmering pools of turquoise and waiting for the sun to evaporate and crystallize the solution into lithium and other by-products. The salares were once home to flocks of pink flamingos, which the Atacameños consider their spiritual siblings. Now the flamingos are gone; the young daughter of one Indigenous leader in the Peine community has only her ancestors’ stories and a flamingo plushie by which to remember them. Over the years, the Atacameños have heard many narratives used to justify all of this extraction. In 2022, as the European Union set new policies around the energy transition and the demand for lithium skyrocketed, both companies and politicians in Chile and the rest of the world lauded the importance of the country’s mining industry in propelling forward a better future. Indigenous communities watching their land and their communities get ripped apart asked: A better future for whom? “Local people never have the ability to think about their own destiny outside the forces of economics and international politics,” says Cristina Dorador, a microbiologist who lives in the north and studies its rich biodiversity. Now the same narratives are being recycled with generative AI. The accelerated copper and lithium extraction to build megacampuses—and to build the power plants and thousands more miles of power lines to support them—is, in Silicon Valley’s account, also ushering in a better and brighter future. To block that extraction is thus to block fundamental progress for humanity. But it is not the mining that Indigenous communities resist. “Our ancestors were miners,” says Ramos. They were the ones who discovered the copper in the first place. The problem, she says, is the scale. That scale has consumed everything. It has made the north and the rest of Chile completely dependent on the industry and not allowed for the emergence of other economies. It has choked off the country’s—and the rest of the world’s —ability to imagine different paths where development could exist without plundering natural resources, Ramos says. By enabling the production of massive generative AI models, that scale has also led to the perpetuation of racist stereotypes about the Indigenous peoples already suffering from how the technology was physically built. In Brazil, a 2023 art exhibition coproduced by a Chilean university showed the vast chasm between the reality of Latin America’s rich Indigenous cultures and the woefully bereft depictions of them spit out by Midjourney and Stable Diffusion as primitive, technologically inept peoples. In recent years, the Atacameños have mounted more and more resistance. They fly black flags on their houses to denounce the exploitation of their lands and their community. They’ve organized protests to physically block the roads that company buses and trucks must take to get to the mines. They’ve contracted lawyers to assert their legal rights as Indigenous peoples under international law, which protects their cultural and territorial sovereignty. As companies and the Chilean government have been forced to invite them to negotiations, central to Indigenous demands are the need for the government to conduct research into the health of the Atacama Desert’s ecosystems and to quantify the water loss and any irreparable damage. Ramos, too, has her own foundation, bringing together “the ancestral and non-ancestral,” she says, to promote and conduct scientific research into the natural wealth that the Atacama Desert has to offer. Due to its uniquely extreme conditions, it is home to many microbial communities—potentially useful for medicines or new sources of energy—that don’t exist anywhere else. For the same reasons, the desert has also been studied for decades as an analogue to Mars’s climate. Ramos hopes that any discoveries will help prove the value of preserving her beautiful homeland. Against the narratives of high-speed progress used to fuel extraction, she searches for new conceptions of progress that promote healing, sustainability, and regeneration. --- As Ramos’s fight continues in the north, a different battle is waging in the heart of Chile, over the government’s embrace of the tech industry’s data centers themselves. The faster the hyperscalers have expanded, outpacing the supply of land and power in their typical regions of operation, the more aggressively they have pushed to lay claim to those resources in new territories globally. Microsoft alone spent more than $55 billion in fiscal year 2024, nearly a quarter of its reported revenue, to build what SemiAnalysis described as “the largest infrastructure buildout that humanity has ever seen.” Google, meanwhile, said in its third 2024 quarterly earnings call that it planned to crank up its data center expenditures to reach around $50 billion for the fiscal year. Meta said it would likely round out the fiscal year with up to $40 billion in data center and infrastructure expansions, which it estimated would rise the following year. On a rare misty afternoon in June 2024, Alexandra Arancibia directs our car in Quilicura, a municipality on the outskirts of Santiago where she lives and serves as a council member, to what she sees as the defining symbol of the yawning power differential between American tech giants and her community. Less than a thirty-minute drive away from Santiago’s most picturesque neighborhoods, packed with European-style cafés and vegan restaurants, the roads in front of us are crumbling from poor maintenance, mountains of trash strewn alongside in illegal dumps controlled by a local mafia. Past a graveyard dedicated to deceased pets, we pull up to what looks like an abandoned plot of grassland with tufts of shrubs and a scattered handful of nutrient-starved trees jutting out of the soil. Most days the land is so parched it looks like parts of the Atacama Desert; today the rain is turning everything into mud. In the middle of the plot, a purple sign announces in Spanish, “Welcome to the Quilicura Urban Forest,” a project, it explains, that Google began in 2019 to give back to the community for hosting its data center. The sign includes a diagram to elaborate on the “forest’s” benefits: on the left side is an illustration of Quilicura as an industrial zone, packed with factories producing greenhouse gases and air pollution; on the right is an illustration of the forest flourishing under the generous rain pouring out from a big cloud labeled “SMOG.” Google boasts about this forest on its website and in its PR releases. When I ask the company’s country spokesperson for an interview about Google’s development in Chile, she sends me instead some polished briefing materials, later adding that the company creates a community impact program for each new data center to support local projects such as in education, sustainability, internet access, and health; for Quilicura, Google has invested over $1.2 million. In her materials, the part about the forest talks about residents using the green space. There are no residents. The place is too far from any bus line, and there are no homes in the surrounding area to speak of. Outside the modest plot, too small to fit Google’s data center itself, a dozen stray dogs meander around, barking and rummaging through the trash. The spokesperson said the forest is being updated to “evolve the experience for the community.” Arancibia smiles wryly as we take the scene in. “Do you feel like you’re in Silicon Valley?” --- Arancibia had just started college when she realized that Quilicura was a place where things were discarded. She was commuting each day to her university through parts of the municipality—piled high with refuse—that she hadn’t known existed. She had never thought of Quilicura as “home”—it was simply the place she lived, an underdeveloped and unremarkable working-class town that her parents moved to when she was little. But something about seeing it treated as a literal dump stirred within her a deep desire to revitalize the land to its former beauty. Only two decades ago, when Arancibia was a child, Quilicura was mostly country: rolling pastures and glistening wetlands, home to a different yet just as rich biodiversity as the Atacama Desert—birds, beasts, and varieties of flora. Then the trash mafia arrived, allowing anyone from the rest of Santiago to dump their waste in Quilicura for a price. Some dumps have operated for so long that grass and weeds have grown over them, making them look like eerie deformed hillsides closing in on the landscape. Next came different industries, including beer companies and real estate developers, who siphoned off more land and extracted water from the wetlands for their purposes. Today only tiny, interspersed pockets of green in this twenty-two-square-mile municipality offer a window into what Quilicura once was. Against this backdrop Google came to Quilicura in 2012 to build its first data center in Latin America. On Google’s web page, the company proudly presents the data center, which became operational in January 2015, as one of the most efficient and environmentally friendly on the continent. At the time, no one in Quilicura paid attention to the project; certainly no one in the better-off epicenter of Santiago was paying attention to Quilicura. Neighborhood residents who passed by the data center every day on their bus route to work assumed that it was a factory producing beer or food and providing jobs to the local community. Google’s data center—like most data centers—did not provide many jobs beyond its initial construction. A job posting from 2024 for a mechanical technician, one of the few long-term positions available, was advertised on Google’s job board only in English; the posting stated that Google wouldn’t consider résumés submitted in any other language. The data center—as activists point out—did not provide much other benefit to the local community either. Nearby, public schools still lack good internet or devices for students to access it. The data center’s arrival marked Quilicura and the rest of Santiago as a desirable destination for Silicon Valley’s physical expansion. In 2019, Google announced that it would build its second Latin American data center in the Santiago metropolitan area. Soon enough, Microsoft and Amazon announced that they were coming too. The Chilean government was quick to welcome them, positioning the country as a safe and stable haven for foreign direct investment in Latin America, which otherwise suffers a reputation for unreliable governments and social and economic instability. In 2020, the government went a step further. It announced a project to build a new underwater cable, akin to a data highway, for connecting the Asia Pacific straight to the Americas through Chile’s central coast, not far from Santiago. Chile would become a global hub for digital infrastructure. Google backed the partnership. But in July 2019, as Google began the paperwork for its second data center in Chile, a group of residents was watching. The company had chosen Cerrillos for its new location, another working-class municipality bordering Santiago. Like Quilicura, Cerrillos has a long history of being overlooked and abandoned. From the 1930s to the 1990s, a cement factory that belonged to a Belgian company contaminated the community with lethal levels of asbestos, leading to what one Chilean historian called “the largest industrial genocide” in the country. To this day, residents still die from higher than average rates of cancer. But Cerrillos is also special—in a country where water is privatized, the municipality is home to the nation’s only public water service, which serves up the local groundwater to neighboring communities and, in emergency situations, to other parts of Chile. This unique combination—a history of neglect and a precious water source —created fertile ground for the blossoming of several environmental activist groups who were used to being watchdogs and were fiercely protective against the extraction of their resources. That summer, as Google filed a report with Chile’s environmental agency for approval of its data center—a largely rubber stamp process—MOSACAT, a water activist group, began combing through all 347 pages of the filing. Buried in its depths, Google said that its data center planned to use an estimated 169 liters of fresh drinking water per second to cool its servers. In other words, the data center could use more than one thousand times the amount of water consumed by the entire population of Cerrillos, roughly eighty-eight thousand residents, over the course of a year. MOSACAT found this unacceptable. Not only would the facility be taking that water directly from Cerrillos’s public water source, it would do so at a time when the nation’s entire drinking water supply was under threat. In 2019, as with Iowa and Arizona, Chile was already nine years and counting into a devastating and historically unprecedented megadrought. --- Tania Rodríguez, a member of MOSACAT, hands me all 347 pages of Google’s environmental filing, printed out and spiral-bound between two blue plastic protectors. The tome drops into my lap with a thud, a physical manifestation of the way Silicon Valley wields technical knowledge to justify its centralized decision-making. Jutting out from the bottom are carefully labeled Post-it notes. “Agua potable” (potable water) reads one in Spanish, bookmarking the pages that discuss the data center’s need to consume fresh water for cooling. MOSACAT was founded in 2019 after activists from several different movements fighting for women’s, housing, workers’, and environmental rights joined in solidarity to form a unified collective. Many had met while protesting an illegal mining project. MOSACAT’s activism successfully chased out the miners, shut down the project, and designated the land a protected nature reserve, the group says. It was shortly thereafter that a friend of the group, who now serves as a member of Chile’s national congress, tipped them off about Google’s data center project and urged them to look at its projected water consumption. MOSACAT’s members are not technologists. But they read through every page of dense diagrams and arcane terminology, took copious notes, and memorized the ins and outs of data centers and their cooling systems to prepare themselves to go up against Google. Rodríguez lets out a spirited laugh when I ask her how they were able to digest all of the information. “It took all of us,” she says—referring to more than a dozen volunteers who make up MOSACAT’s membership and do the work in stolen hours between jobs and family obligations. In most cases, projects in Chile that require water take a long time to receive approval. In Google’s case, the approval came quickly, even in the midst of a series of drought-related water emergencies. At first, MOSACAT sought to contest the project through Google’s local partner, a Chilean investment and services firm named Dataluna. The initial meeting went badly, MOSACAT says: The Dataluna representatives seemed to have little understanding of the project and denied that it would use fresh water. From there, MOSACAT went to the local government. The mayor and city council had themselves been meeting with Dataluna, the group remembers, and held the false impression that the data center needed only wastewater for its cooling. After MOSACAT briefed them, the government, alarmed, demanded an explanation from Dataluna. The matter escalated from Dataluna to Google’s Chile division all the way to Google’s headquarters in Mountain View, California. In October 2019, Google sent two engineers and a lawyer to Cerrillos to present to the community. The day they arrived, MOSACAT plastered the streets with protest signs along the route the Googlers would drive to get to the meeting location. At the meeting itself, MOSACAT didn’t come alone. Among the roughly two dozen residents who attended, six other activist and community groups were represented. The Google engineers were gringos, MOSACAT remembers, tall and able to speak only English. They gave a highly technical presentation, and Google’s lawyer doubled as a translator. During the discussions, MOSACAT members who spoke English say they could hear the lawyer mistranslating their words. At another point, the Google representatives sought to assuage the community by offering to plant an urban forest just like the one the tech giant had given to Quilicura. The show left MOSACAT and the other groups unimpressed: Google wasn’t here to truly engage with and hear what the community wanted. “They came to intimidate us,” says a MOSACAT member Alejandra Salinas, who also serves as a council member for Cerrillos’s neighboring municipality, Maipú. “Think about it. They come offering us trees while drying out our earth.” Residents of Cerrillos didn’t need trees. They didn’t need Google to build them a park—as if the Santiago metropolitan area were such a backward place that it didn’t already have parks. They needed Google to stop treating their land as a place to plunder precious water and other resources; they needed the company to stop dismissing the community as bystanders instead of participants in the development of its local projects. “We know that we feed the world, that we provide raw materials like copper and lithium,” Salinas says. “Nobody is saying our treasure is ours alone and we won’t share it. Yes, we can help each other. But they are not going to come and use the water, which is vital for life, and leave us with nothing.” At the time, Chile was in the midst of a massive, monthslong political upheaval, known as the Estallido Social. Explosive and at times violent protests were erupting every week, beginning that same month in October 2019—a collective outcry over unemployment, privatization, and deep inequality that left thousands injured and dozens dead. In late 2019, as communities across the country began to hold referendums in response to the movement to reform local and national politics, MOSACAT piggybacked on the referendum in Cerrillos to add a question about whether residents agreed for Google to build a facility that would consume so much of the community’s water. They mobilized to broadcast the true nature of the project, handing out flyers on street corners, knocking on people’s doors, and posting signs all over the municipality. In December 2019, MOSACAT won; the referendum to build the data center was rejected with a slim majority of the vote. The following year, the government joined MOSACAT in filing a lawsuit against the project with an environmental court. In the meeting with Google, its representatives ultimately put on a friendly face, MOSACAT says, presenting a greater willingness to negotiate once it became clear that some residents could understand them. But as both Julian Posada, the assistant professor at Yale, and Mercy Mutemi, the lawyer for Mophat Okinyi, told me, the risk of pushing back as a Global South country is always that a Silicon Valley company will pick up and take its money somewhere else. As the project continued to stall and Google’s desire for more compute intensified, the company announced that it would shift its next planned data center in Latin America from Chile to another country. --- In Uruguay, a small country of 3.4 million, the national telecommunications company Antel has three data centers that help provide internet and cell services to the entire country. Cumulatively, they take up only some five thousand square meters. One, less than one thousand square meters, sits nestled into a typical residential street in Montevideo, taller and wider than its neighboring buildings but integrated into its surroundings. The data center runs on two power lines that are separate from the rest of the neighborhood’s. Thirty percent of the space is filled with computers, 70 percent with administrative offices, electrical closets, and mechanical rooms. The computers get hot but not so hot that they can’t still be cooled with air instead of water. They produce a low hum barely audible during the bustle of the day and just enough of a nuisance in the quiet of the night that two neighboring families have come and knocked on the data center’s door to complain with some regularity. The manager of the data center, Javier Echeverria, looks sheepish as he admits this. He says he is now working with local researchers on solutions to reduce the noise and has already modified the cooling system to be less noisy. The responsiveness is a far cry from the rigmarole that MOSACAT had to go through to get an American company’s attention. A thirty-minute drive away, right outside the city limits, the government delineated a large expanse of land to develop a science park, Parque de las Ciencias, which operates as a zona franca, a free-trade zone. Some cheekily call it zona America, with a hint of bitterness, for housing mostly American companies that do not pay taxes to the government. The park even looks somewhat like America, its lush, manicured lawns, symmetric design, and majestic sundial-shaped fountain reminiscent of the stately aesthetic of the National Mall in Washington, DC. The home page of the park’s website advertises a politically and economically stable country, plenty of land, and an abundant supply of power and water. So it came to pass that in 2021, as GPT-3 spurred new interest in dramatically scaling AI models, Google purchased twenty-nine hectares of land here, fifty-eight times the size of Antel’s total data center footprint, to establish a different home for its second data center in Latin America. But at the time, Uruguay did not in fact have an abundance of water. Like Chile, like Iowa, like Arizona, it was also experiencing a devastating drought. The water shortage was so severe that farmers were losing their entire harvests, costing the country over $1 billion in agricultural losses; by the summer of 2023, the Montevideo government would start mixing contaminated salt water into the city’s drinking supply. Families opening their taps saw a putrid brownish fluid pouring out that smelled intensely of chemicals. Those who could afford it purchased bottled water for drinking and bathed with their windows open to avoid breathing in too many carcinogens. Those who couldn’t drank the tap anyway, leading many to suffer stomach pains, skin rashes, an aggravation of existing health conditions, and the agony of a growing rate of miscarriages. In the aftermath of a raging pandemic, there were many who couldn’t afford bottled water. Where Silicon Valley had ascended, with Google’s and Microsoft’s market capitalizations both peaking above $2 trillion, in part from companies going remote and increasingly relying on cloud services, illegal housing settlements in Uruguay had grown by orders of magnitude. Ollas, the local equivalent of soup kitchens, were turning children away hungry. Those running the ollas were themselves in poverty and barely surviving. Fabiana, the boisterous head of an olla who lives in an illegal settlement and is affectionately called Reina Madre (Queen Mother), grows quiet as she remembers it. “To have to say, ‘I don’t even have a little plate for your child…’ ” She trails off. “It was horrible.” Even after a lifetime of poverty that included sweeping the floors of a brothel at seven years old for survival, she finds the pandemic and drought years stand out in her mind as some of the worst in her life. The water crisis emerged from the compounding effects of climate change and a failure of the state’s allocation of freshwater resources: In Uruguay, more than 80 percent of the country’s fresh water goes to industry instead of human consumption—most notably, cash crop agriculture. These include industrial farms for soybeans and rice, and for trees that feed into paper production. Most such farms are run not by local companies but by multinationals that export what they grow and show little accountability for Uruguay’s natural environment. Their activities deplete the nutrients in the soil, making it more difficult to grow actual food, and pollute the country’s water streams with a volume of fertilizers that makes Uruguay one of the world’s largest per capita fertilizer consumers and causes unusually high rates of cancer. Daniel Pena, a sociology researcher at the Universidad de la República in Montevideo who has for years studied the politics of this environmental extractivism, draws a direct connection to Uruguay’s colonial history. He drives around the country in a beat-up pickup truck to interview farmers and residents of the poorest neighborhoods, to document up close how they’re squeezed by industry. As with Chile, the foreign multinationals still exist above locals in the political pecking order. During the drought, industry continued to use water unabated, drawing what it needed directly from the main river, Río Santa Lucía, that also feeds Montevideo’s public water system. Two decades ago, after significant environmental activism, Uruguay became the first country in the world to recognize water as a human right in its constitution. Now, in a bitter irony, it was the drinking water rather than water for industry that saw the most severe cutbacks during the shortage. So when Google arrived, Pena was vigilant. During his regular scans of the Uruguayan environmental ministry’s website, which lists major industrial projects, he came across the company’s proposal for the data center. Pena had read about hyperscalers using potable water, even during major droughts, and the activism of communities like MOSACAT that had resisted the projects. But when he downloaded the details of the project, the water numbers were marked as confidential. After submitting a public information request, which he had successfully done around twenty times, the ministry continued to withhold the numbers, saying they were proprietary information. Pena wondered what they were hiding and worried about the precedent it would set for other cloud companies that would inevitably begin to eye Uruguay, following Google’s lead, for their own expansion. So he evoked the water clause in the constitution. With the help of a lawyer friend who was willing to work pro bono, he sued the ministry. In March 2023, four months later, Pena won the case in a surprising victory. The environmental ministry revealed that Google’s data center planned to use two million gallons of water a day directly from the drinking water supply, equivalent to the daily water consumption of fifty-five thousand people. With much of Montevideo receiving salt water in their taps not long after, the revelations were explosive. Thousands of Uruguayans took to the streets to protest Google and all of the other industries that had led the government to squander the country’s precious freshwater resources. The slogans of resistance are still scrawled across the city’s walls and roadside barriers during my visit in June 2024. “This is not drought,” reads one. “It’s pillage.” Pena sees data centers in the same way that Ramos sees mining. It’s not the infrastructure itself that’s the problem, but the scale at which Silicon Valley is trying to build it. That scale is what drove companies like Google and Microsoft to expand in Chile and Uruguay even as the countries suffered from a severe lack of resources. That scale is what makes them require fifty-eight times more land than Antel and operate with much less accountability to the local population. “They are extractivist projects that come to the Global South to use cheap water, tax-free land, and very poorly paid jobs. And then they don’t contribute to our country; they don’t improve our internet access,” he says. Near the end of 2023, Google silently updated its proposed data center in Uruguay to use a waterless cooling system and said it would reduce the facility to a third of its size. Pena says the fight is still not over: The government is now withholding the projected energy consumption of the latest proposal as a commercial secret. Pena also sent a petition to the ministry, with over four hundred signatories, demanding a more extensive environmental and social impact study of the full supply chain of the data center: where the minerals for producing its hardware are being extracted, how the labor involved is being treated, how much carbon will be emitted, how the generated e-waste will be disposed of in a way that doesn’t leach chemicals into someone’s community. Most of these other impacts won’t befall Uruguay, but Pena feels a solidarity with the other countries where they will. They are “generally all from the Global South,” he says. “We all end up with the same consequences, but from different links in the global supply chain.” In 2024, Chile’s environmental court ruled that Google cannot build a water-using data center in Santiago. The Google Chile spokesperson said the company remains committed to the country and Latin America, and plans to begin the permitting process for an air-cooled data center in Cerrillos when needed. But that hasn’t slowed down other hyperscalers from entering Latin America. In 2022, Microsoft finalized the location for its data center in Chile, shortly after its second investment into OpenAI—right back in Arancibia’s hometown, Quilicura. --- As data center developers go, Microsoft was a late bloomer. Before its investments in OpenAI, Google was well ahead in the number of facilities it was constructing around the world. But with the sudden explosion of demand for more computing infrastructure to support its AI ambitions, Microsoft adopted Google’s playbook and followed it into the same regions. The Chile that Microsoft entered was different from the Chile that had greeted Google. By 2022, more than two years had passed since the Estallido Social, which had left an indelible mark on the country’s politics. After months of protests, Chile had undergone a remarkable experiment to rewrite its constitution, with the broad participation of regular citizens, to replace the one that had carried over since Pinochet’s brutal dictatorship. Ultimately the new drafts of the constitution didn’t pass; two separate processes resulted in two partisan documents that failed to gain broad support. But it reinvigorated the youth and working-class families in particular with a new optimism for democracy. The upswell of leftist ideas and support led, in a dramatic turn of events, to the election of a millennial left-wing president, Gabriel Boric Font, only thirty-five years old. In his victory speech, Boric repeated a slogan of the protests that slammed the legacy of the Chicago Boys during Pinochet’s rule: “If Chile was neoliberalism’s cradle, it will also be its grave.” Boric, who began his term in March 2022, had himself been a student protester. It emboldened youth across the country to take what they learned from the social upheaval about organizing and protest to establish a new generation of progressive activist organizations. They became part of the rhythms of each community, meeting regularly to dream up big-picture visions about what they wanted for the future of Chile. It was during this period that Arancibia cofounded her own activism group with another young Quilicura resident, Rodrigo Vallejos. The two had met during the organizing of the social upheaval and quickly bonded over their deep passion for the environment. They called their group Resistencia Socioambiental Quilicura—the Socio-environmental Resistance of Quilicura—drawing upon a well-established concept in Latin America that the social and the environmental are inextricably linked. Upon Microsoft’s entrance into Quilicura, Vallejos and Arancibia did what MOSACAT and Daniel Pena had before them: They began to pore over whatever materials they could find that Microsoft had made available and to extensively research the project. Vallejos, a law student at Universidad Diego Portales, worked late into the nights in between his schoolwork to read technical documentation and teach himself about how data centers work. Microsoft projected that it would need a significantly lower amount of water than Google had in Cerrillos. Even still, Vallejos worried. The drought in Chile had only gotten worse and was expected to last until 2040. Quilicura’s wetlands were suffering acutely, on top of industrial encroachment, from accelerating desertification. On its website, Microsoft boasted about new cutting-edge innovations in data center cooling systems that would mean its facilities didn’t need to use water. If Microsoft had the capability to build waterless data centers, why wasn’t it doing so in Quilicura? “It is deeply striking that a company with as much reputation as Microsoft publicly presents an environmentally friendly discourse, but in reality does not comply with global innovation standards in a third world country like Chile,” Vallejos later wrote in an article. Microsoft would subsequently explain that the innovations it had advertised were still under development and only being piloted in a place in the US. “Then why do they promise these things” on their website? Vallejos asks. Vallejos caught the attention of local and international researchers, including Marina Otero Verzier, a director of research at Nieuwe Instituut, the Dutch institute for Architecture, Design and Digital Culture, and Serena Dambrosio and Nicolás Díaz Bejarano, researchers at FAIR, the think tank co-led by Martín Tironi Rodó. Otero was moved by the passion of Vallejos and Arancibia, and their exhaustion. They had worn themselves thin reading Microsoft’s long technical documents, writing critical articles, and protesting continuously, but had struggled to get an audience from either the company or the government. Otero pondered ways to help them. How could she get them in a room to negotiate with the right people? Otero knew she had one thing Vallejos did not: affiliations with prestigious universities like Harvard and Columbia that would command Microsoft’s and the government’s attention. She began to mount a multipronged campaign, growing so deeply involved that she quit her job: She spoke at high-profile conferences about the environmental impacts of data centers and Resistencia’s fight against them; she forged connections with the Chilean Ministry of Science, Technology, Knowledge and Innovation and with representatives at Microsoft and Google; she connected Vallejos and Arancibia to other international researchers to elevate their profile. With Dambrosio and Díaz, Otero also developed a more speculative project. All three had architectural backgrounds and had been studying the infrastructure of modern digital technologies through the lens of the built environment. They began to wonder: What if they treated data centers as architecture structures and fundamentally reimagined their aesthetic, their role in local communities, and their relationship with the surrounding environment? Díaz liked to visit national libraries during his travels—beautiful venues that seek to capture the grandeur of a country’s memories and knowledge. It struck Díaz that data centers, too, could be thought of as libraries—with their own stores of memories and knowledge. And they, too, could be designed to be welcoming and beautiful instead of ugly and extractive. This represented a sharp departure from Microsoft’s and Google’s definitions of what it means to give back, such as through the latter’s community impact programs, with what Díaz calls their “schizophrenic” initiatives, which tend to be divorced from how communities are actually affected by the companies’ facilities. Together with Vallejos and Arancibia, the three researchers applied for funding and put together a fourteen-day workshop, inviting architecture students from all around Santiago to reimagine what a data center could look like for Quilicura. The students designed stunning mock-ups. One group imagined making the data center’s water use more visible by storing it in large pools that residents could also enjoy as a public space. Another group proposed tossing out the brutalist designs of the typical data center in favor of a “fluid data territory” where data infrastructure coexists with wetland, mitigating its damaging impacts. The structures of the data center would double as suspended walkways, inviting Quilicura residents to walk through the wetland and admire the ecosystem. Plant nurseries and animal nesting stations would be interspersed among more traditional server rooms to rehabilitate the wetland’s biodiversity. The data center would draw polluted water from the wetland and purify it for use before returning it. The computers themselves would collect and process data about the health of the wetlands to accelerate the local environment’s restoration. “We’re not fixing the problem, but we’re imagining other types of relationships between data and water,” Díaz says. At the end of the workshop, the students presented their ideas to residents and other community members. “It was an incredible conversation,” Otero says. “You can see how much knowledge the community has. They had so much to offer.” --- Three years into his four-year term, Boric is under pressure to get “quick wins” for the economy—even more so as a young, left-wing president. That means pressure to expand the mining industry, pressure to see through the arrival of twenty-eight new data centers. On the day Boric announced a plan to develop a national data center strategy, Vallejos texted me a video of the press conference and an emoji: the disoriented face with spiral eyes. But in fairness, the coalition of activists in northern Chile and Santiago with researchers domestic and abroad has clearly made a mark. As part of the data center plan, the Ministry of Science, which oversees its creation, has for the first time formed a committee of activists to consult with regularly as part of the drafting process, and invited Vallejos, Arancibia, and members of MOSACAT to join them. The Ministry of Science is also overseeing an AI bill that articulates how Chile wants to approach AI development, application, and regulation. Whereas before the ministry’s discussions cast AI as a universal positive, the tone has since shifted to acknowledge the social and environmental costs of the technology. Chile, like many Global South countries, has learned from hard experience that it should not wait around for the Global North to decide how it will build digital technologies. “The way we build technology responds to a certain cultural framework and historical framework,” says Aisén Etcheverry, the head of the ministry. Where the global internet was shaped without Chile, the country now has an opportunity to shape AI on its own terms. Tironi pushes this one step further. It’s very clear that the AI industry today is rooted in a colonial ideology, he says: It imposes its worldview and its technology—what is AI, what is good AI, what it means to create an industry of AI—on the rest of the world. Chile could be a leader in resisting that imposition. After centuries of extractivism, the country intimately understands what it means for its land to be hollowed out, dispossessed, and destroyed under a banner of progress. It could use those experiences as a wellspring from which to generate fundamentally new conceptions—decolonial conceptions—of AI. “In the planetary market of AI, we as a country are playing a specific role: giving materials to develop this technology,” he says. “Many companies are trying to extract a lot of material from us to create AI. “So we need to think from this position, this geopolitical position, this terrestrial position. We can think of another way to relate technological innovation with the earth.” It is a noble ambition, and the forces arrayed against it are mighty.