By William I. Robinson
This article was originally published by Truthout
AI is deepening worldwide webs of accumulation, ushering in new class and power relations.
The relentless drive by Big Tech corporations to expand data center construction to every continent has sparked a growing movement of mass resistance. From the United States to Ireland, from Chile to Malaysia, grassroots communities are protesting water depletion, soaring utility bills, air and noise pollution, and environmental destruction. Behind the data center boom is the artificial intelligence (AI) revolution. The amount of capital and energy needed to supply AI training and compute power is skyrocketing. The massive increase in compute power in turn requires monumental amounts of electricity along with water for cooling computers. This tsunamic shift in energy requirements has led to a global grab for land on which to locate data centers, energy sources to fuel these centers, and the rare earth and other critical minerals that AI technologies require.
The data center boom has spared no continent. New hyperscale data centers cost more than $100 billion; generate upward of one gigawatt, equivalent to an average nuclear power plant; and sprawl across thousands of acres. China and the United States are experiencing the fastest growth in data centers and investment is also flowing outward from both countries and from Europe to new regions. Southeast Asia is experiencing an unprecedented data center boom, with major hubs in Malaysia, Thailand, Vietnam, the Philippines, Indonesia, and Singapore. Latin America is experiencing its own expansion, mirroring that of Southeast Asia, with hubs in Mexico, Brazil, Chile, Colombia, and Peru, with submarine cables linking hubs to global circuits.
Africa is going through a similar construction boom, with major hubs in South Africa, Kenya, Nigeria, and Morocco. Data centers are proliferating in the Middle East, driven by the availability of cheap land, abundant fossil fuel, and billions of investable dollars in sovereign wealth funds. Saudi Arabia’s $40 billon fund made it the largest investor worldwide in AI in 2024. Both it and the United Arab Emirates announced that year that they would launch $100 billion investment plans for data centers, researchers, and other AI ventures. In the words of OpenAI CEO Sam Altman, “I do guess that a lot of the world gets covered in data centers over time.”
A data center under construction in the U.S. state of Utah gives an idea of the gargantuan dimensions of these mega-investments. Once completed, it will cover twice as much land as the island of Manhattan, require more power than the entire state of Utah uses, and suck in a vast amount of water in the drought-stricken area. A recent United Nations Report projects that by 2030, global AI development will annually demand 945 terawatt-hours of electricity and 9.3 trillion liters of water — equivalent to the water needs of 1.3 billion people — threatening to aggravate shortages and drought. In the United States, about two-thirds of upcoming data centers are set to be built in places that have been among the driest in the country over the past year. Of 809 planned data centers, 517 are in locations that have been in drought conditions throughout the past year.
Pax Silica and the AI Capital Complex
AI is deepening worldwide webs of accumulation, bringing about new class and power relations, and promising to have a transformative effect around the globe. It is fast becoming a general-purpose technology — meaning that, like electricity or the internet, it spreads throughout all branches of the economy and society and becomes built into everything. Nearly 100 percent of U.S. economic growth in 2025 came from investment in digital technologies. To give an idea of the meteoric rise of AI-driven technologies to the very core of the global economy, adjusted annual private fixed investment in these technologies in the United States alone jumped from $17 billion in 1970, to $65 billion in 1980, to $175 billion in 1990, $496 billion in 2000, topped $800 billion in 2020, and then reached $1.5 trillion by early 2026. Worldwide, AI spending was forecast to total $2.5 trillion in 2026, a 44 percent increase over 2025. To put this in perspective, the expansion of AI infrastructure in 2025 represented nearly three percent of the entire global GDP.
Use of AI is also skyrocketing around the world. In 2024, 78 percent of organizations surveyed by Stanford University’s Human-Centered Artificial Intelligence center reported using AI, compared to 55 percent just a year earlier (the percentage using generative AI jumped from 33 to 71 in this same period). China showed the highest rate of growth in the use of AI — 27 percent — from 2023 to 2024. This growing dependence on AI is not limited to the core countries of the world capitalist system. The 78 percent figure is global. The figure was 72 percent in Asia-Pacific, 80 percent in Europe, 82 percent in North America, 75 percent in China, and 77 percent in what the Stanford report refers to as “developing markets,” including India, the Middle East and North Africa, and Central and South America.
The new AI-driven technologies mediate the gamut of human interactions. Many corporations, states, political, military institutions, schools, and hospitals are now dependent on these technologies to function. This dependence of our global economy and society on these technologies gives big tech corporations and the billionaires, financiers, and warlords who control them enormous structural power over states and peoples. A vast expansion of data centers to feed the computational needs of AI underpins this emerging AI capital power bloc. This new ruling class is cohering as a triangulated bloc of transnational capital that brings together the giant tech companies, transnational finance capital, and the military-industrial-repression complex — to put this another way, the bloc brings together Silicon Valley, Wall Street, and the Pentagon.
The AI revolution has greatly enhanced the class power of the transnational capitalist class over the global working class and its structural power over capitalist states. In his final farewell address to the nation on January 14, 2025, outgoing U.S. President Joe Biden warned of a “tech industrial complex” and of a new “oligarchy taking shape” through the “dangerous concentration of power in the hands of a very few ultra-wealthy people” — a warning that reenacted president Dwight Eisenhower’s famous 1961 final address on the dangers of the “unwarranted influence of the military-industrial complex.”
The U.S. State Department has referred to the new AI dispensation as Pax Silica. “If the twentieth century ran on oil and steel, the twenty-first century runs on compute and the minerals that feed it,” declared U.S. Under Secretary of State for Economic Affairs Jacob Helberg in December 2025. Pax Silica involves developing “global supply chains” that are to drive “historic opportunity and demand for energy, critical minerals, manufacturing, technological hardware, infrastructure, and new markets not yet invented.” In the first year of its second term, the Trump government undertook a sweeping deregulation of AI and promised to promote a vast expansion of data centers.
Led by the new hegemonic capital complex, the transnational capitalist class has unleashed a predatory round of digitally driven expansion around the world, pivoting toward more extreme forms of extractivist accumulation as it seizes land, energy, and mineral resources to fuel the demands of AI technology and data centers. This is the big picture behind the global maelstrom, from Venezuela to Iran, from the Congolese and Sudanese civil wars to Washington’s Greenland grab to ICE terrorism in U.S. cities.
AI in the hands of the transnational capitalist class is an instrument of class warfare against the global working and popular classes. AI accelerates the process whereby labor is deskilled, made precarious, and expelled through algorithmic restructuring of the labor process and outright automation. The flip side of the jobs apocalypse this will induce is the degradation and immiseration of those workers still required, including the data workers themselves who are reduced to itinerant laborers. One report tracked down the workers who powered OpenAI’s popular AI chatbot, ChatGPT. It found that the wages were so miserable that a number of them could not afford housing and were homeless.
Warfare, social control, and repression are also being digitalized and automated. Global AI supply chains span everything from mineral extraction to manufacturing to data center operation. Each step along the way involves the displacement of communities, the exploitation of labor, and the repression of resistance. Big tech has sought to embed itself in military and security agencies around the world. The AI state is a warfare state with enhanced capacity to surveil and control. The whole thrust of capital investment and accumulation worldwide is shifting to advanced AI-driven militarization.
Communities Fight Back Against the Data Center Takeover
Seen in this light, escalating resistance worldwide to the data center construction boom constitutes working class struggle from below. As the “Silicon Valley of Europe,” Ireland has grabbed global attention for its growing mass movement against the data center invasion. The country has one of the largest clusters of data centers in the world. Dozens of centers in Dublin alone suck in more electricity that all urban homes combined in the country, accounting now for over 20 percent of the country’s electricity supply. A report by Friends of the Earth found that data centers in Ireland rely heavily on fossil fuels, are driving up energy prices, and are projected to add up to nearly $740 to household utility bills over the next decade.
In Mexico, residents near a Microsoft data center say that, since the plant opened in 2024, power cuts have become more frequent and water outages can last for weeks, affecting schools and hospitals. In Chile, activists shut down Google’s expansion plans in 2024. In the Netherlands, protests blocked access to a construction site for Microsoft’s planned facility this month. In Malaysia, citizens in the state of Johor launched the country’s first data center protest, demanding compensation for dust and noise pollution. Similar protests have taken place in the United Kingdom, Brazil, Spain, Australia, and throughout the United States as a global movement comes into being to push back against the billionaire tech oligarchs.
In the United States, 7 out of every 10 people oppose having data centers located in their communities. The backlash has led, as of June 2026, to over 100 proposed moratoriums on new data center construction at the local, county, state, and national level. At commencement ceremonies on U.S. university campuses graduates now boo their commencement speakers as soon as they mention AI. In May 2026, a group of leading activists, journalists, scholars, and researchers launched the AI Resist List, a publicly accessible global database that documents resistance against the AI capital takeover. The ruling class is taking notice. There is growing evidence that police departments and other state security and intelligence agencies are surveilling and even preparing to criminalize data center critics.
State Ownership and the Future of AI
In March, Sen. Bernie Sanders and Rep. Alexandria Ocasio-Cortez introduced the AI Data Center Moratorium Act, legislation that would enact a “reasonable pause” to the development of AI “to ensure the safety of humanity.” Sanders also called for the public to take a 50 percent ownership stake in AI companies. In early June, Sanders met with OpenAI CEO Sam Altman, at the latter’s request, to discuss the Senator’s proposal. Altman agreed that there should be some measure of public participation but he made clear he did not support Sander’s 50 percent threshold, which would deprive technology corporations of controlling shares.
However, these kinds of public ownership arrangements can operate as a mechanism to subsidize capital just as they can function to impose constraints on capital’s prerogatives, depending on the nature of the arrangement. Thus far, the state has intervened in the private sector less to regulate than to guide and shore up the development of digitally based accumulation and the war economy, suggesting that the emerging Pax Silica world order requires certain forms of state capitalism.
In early 2026, the U.S. government committed an initial $10 billion to acquiring equity stakes in private companies from the AI capital complex, including semiconductor company Intel, the military firm L3Harris Missile Solutions, and rare earth mining companies MP Materials and USA Rare Earth, among others. In addition, the White House issued an executive order to “revitalize the defense industrial base” that obliges private contractors with the federal government to prove they are prioritizing “warfighter capability and readiness.” Senate legislation later codified the order, requiring contractors to submit “qualified defense investment plans.” Whether public AI ownership in the United States would function to bolster capital’s power or force capital to be more responsive to the working class depends on how mounting class and popular struggles play out.
This article was originally published by Truthout and is licensed under Creative Commons (CC BY-NC-ND 4.0). Please maintain all links and credits in accordance with our republishing guidelines.
Editor’s Note: At a moment when the once vaunted model of responsible journalism is overwhelmingly the play thing of self-serving billionaires and their corporate scribes, alternatives of integrity are desperately needed, and ScheerPost is one of them. Please support our independent journalism by contributing to our online donation platform, Network for Good, or send a check to our new PO Box. We can’t thank you enough, and promise to keep bringing you this kind of vital news.
You can also make a donation to our PayPal or subscribe to our Patreon.
Please share this story and help us grow our network!
