The AI Raj: How tech giants are recolonizing power

2025-09-15 10:02:40 英文原文

作者:Sara Goudarzi

A brightly coloured illustration which can be viewed in any direction. It has many elements to it working together: men in suits around a table, someone in a data centre, big hands controlling the scenes and holding a phone, people in a production line. Motifs such as network diagrams and melting emojis are placed throughout the busy vignettes.Just as the East India Company’s commercial success gradually justified new powers, today’s AI firms seek to leverage technical prowess to assume public functions by default. Image: Clarote & AI4Media / Power/Profit / Licensed by CC-BY 4.0

On December 31, 1600, Queen Elizabeth I signed a royal charter granting the East India Company exclusive rights to conduct trade in the Indian Ocean region. The document was precise in its limitations: The company could establish trading posts, negotiate with local rulers, and defend its commercial interests. Nothing more.

Seventy-seven years later, the same company had acquired the right to mint currency on behalf of the British crown. By 1765, it controlled the tax collection (ruthlessly enforced by its own private army) for the Indian provinces of Bengal, Bihar, and Orissa—territories containing roughly 20 million people. What began as commercial efficiency had become imperial governance. The transformation was so gradual that few contemporaries even noticed sovereignty shifting in the region from local rule to corporation.

A similar pattern can be seen today with national governments and Big Tech—only this time, centuries of drift have been compressed into months. Where the East India Company deployed trading posts and private armies, today’s technology firms and specifically AI development companies use data pipelines, data centers, and algorithmic systems. The medium has changed; the mechanics of private power assuming public functions remain the same.

Consider the trajectory of Elon Musk’s so-called “Department of Government Efficiency” (DOGE). Established in February 2025 with the stated goal of eliminating bureaucratic waste but an unstated aspiration to vacuum up new data to improve Musk’s companies, DOGE began with access to federal payment systems—ostensibly to identify inefficiencies. Within weeks, reports emerged that DOGE personnel had gained the ability to alter government databases, including Social Security records and contractor payments. The justification remained consistent: To deliver efficiency, one must first seize control.

The parallel extends beyond metaphor. Just as the East India Company’s commercial success gradually justified new powers, today’s AI firms seek to leverage technical prowess to assume public functions by default, implicitly assuming that the reallocation of power will serve human flourishing. Each efficiency gain becomes justification for the next transfer of authority, yet the costs of that automation go uncalculated.

What once took generations now takes quarters; the key difference is the ease with which private digital systems can be aligned with the politics of friends and enemies. Communications systems, financial networks, and governance mechanisms are no longer reshaped through military conquest but by software updates. Increasingly, those same systems are being weaponized against the very allies who helped build them.

From content moderation to infrastructure control to monetary governance, AI companies are taking on public operations. As AI becomes a more prominent feature of everyday life, already existing problems in our public life will proliferate exponentially. The transformation before us is likely to proceed through three variants—algorithmic capture of information systems, weaponization of critical infrastructure, and cryptocurrency’s escape from public accountability. Absent immediate intervention, democratic societies risk permanent subordination to unelected digital sovereigns.

I. The algorithmic Raj

Content moderation as corporate abdication. The transformation begins with what once appeared to be responsible governance. As I show in my forthcoming book, Who Elected Big Tech?, social media platforms have rarely upheld their stated terms of service, instead moderating content to protect the company bottom line. As these systems scaled globally, the pretense of neutral arbitration gave way to something more profitable: the wholesale abandonment of truth as a governing principle.

Meta’s evolution reveals this trajectory most clearly. The platform that once maintained fact-checking partnerships and claimed to combat misinformation abandoned these efforts entirely in January 2025. The company’s CEO and chairman Mark Zuckerberg framed the decision not as corporate surrender, but as a triumph of free speech. The move liberated the platform to optimize entirely for engagement, free of the burden of distinguishing truth from falsehood.

This represents a fundamental shift from governance to profit maximization. Where content moderation once involved difficult decisions about balancing free expression with harm prevention, the new model eliminates such complexities entirely. Truth becomes irrelevant, only engagement matters. The algorithms now amplify anything that maximizes attention—whether rigorously sourced journalism or imaginative fact-free conspiracy.

The mechanics reveal how thoroughly this resembles colonial extraction rather than governance. Facebook still employs many content moderators in developing countries—concentrated in places like Kenya, the Philippines, and India—but their role has fundamentally changed. Instead of adjudicating truth claims or preventing harm, they now remove only the content most threatening to advertising revenue—child sexual abuse material, terrorist propaganda, and little else. The colonial hierarchy persists, but the mission has devolved from quasi-judicial oversight to commercial protection.

This amplification apparatus no longer functions as even a flawed legal system—particularly outside US borders. Instead, it operates as a commercial enterprise disguised as a public service. Platforms develop engagement-optimization rules, test them for revenue impact, and deploy them globally with little public input. The decisions are made unilaterally, without regard for democratic principles, constitutional rights, or even basic accuracy standards.

But content moderation is only the first layer. The deeper story lies in digital infrastructure.

Infrastructure weaponization. The greater transformation occurs through infrastructure capture that operates beyond any single nation’s control—and increasingly, against the very allies who depend on these systems. Elon Musk’s Starlink satellite constellation exemplifies this new form of power: private infrastructure dominance that can supersede governmental authority while being held hostage to its owner’s political preferences.

Starlink’s network now makes up about 62 percent of all active satellites in orbit, creating dependencies that span the globe. During Russia’s invasion of Ukraine, this infrastructure dependence became a geopolitical weapon. When the Trump administration reportedly threatened to withdraw Starlink access from Ukraine unless the country handed over rights to exploit its mineral reserves, this demonstrated how private infrastructure had become an instrument of state coercion.

The weaponization extends far beyond warfare. Earlier this year, when Poland’s foreign minister Radek Sikorski attempted to defend Ukraine’s access to Starlink services, Musk’s response—“Be quiet, small man…”—echoed the tone of East India Company dispatches: a private entity treating sovereign governments as subordinate clients.

The regulatory arbitrage is built into the system’s architecture. Satellites operate in international space, following orbital mechanics rather than territorial boundaries. Traditional telecommunications regulation assumes infrastructure exists within specific jurisdictions, but Starlink’s satellites pass over every nation on Earth while being controlled from private facilities in the United States. No single government possesses the technical capacity to regulate a system that literally operates above their heads.

The F-35 program reveals how military platforms, too, generate dependency. While there may be no literal “kill switch” on an F-35, the reality is more insidious: The aircraft relies on constant maintenance, software updates, and real-time intelligence—all of which flow through US networks and can be shut off. When the United States temporarily cut off intelligence to Ukraine after Trump’s February meeting with Zelensky, it effectively disabled missile systems dependent on encrypted GPS feeds.

The payment integration completes the picture. Musk’s simultaneous control of Twitter/X, Starlink communications, government database access through DOGE, and his aspiration for X to have cryptocurrency capabilities creates potential for end-to-end governance systems that operate beyond any single nation’s regulatory authority while being harnessed to Musk’s personal and political preferences.

Where Starlink exemplifies hard infrastructure weaponization, cryptocurrency represents the soft infrastructure of finance and governance—equally revolutionary and equally unaccountable.

The crypto escape hatch. Cryptocurrency offers the final lever in the shift away from democratic accountability—one that fractures even the private sector it was meant to empower.

When President Trump signed the GENIUS Act into law in July 2025, it created the first federal framework for stablecoins, while conspicuously exempting the president and executive branch from conflict-of-interest provisions that applied solely to members of Congress and their families. That exemption came amid reports that Trump’s family had profited handsomely from token sales and stablecoin ventures like the Trump organization’s World Liberty Financial.

Meanwhile, other legislative efforts long sought by the crypto industry—such as the Digital Asset Market Clarity Act and the Anti-CBDC Surveillance State Act—remain stalled. This selective success has created a rift that is becoming more apparent. The GENIUS Act legitimized Trump’s crypto profits while denying broader protections to the rest of the sector.

The backlash was swift. On July 28, Coinbase quietly changed its terms of service, requiring all users to waive the right to sue, submit to binding arbitration, and acknowledge they transact at their own risk. The platform’s move, viewed by many as legal triage, signals an awareness that fair rules of the game for crypto players and “normies” (the crypto bro name for ordinary investors) alike do not yet exist and underscores the industry’s growing uncertainty about its legal standing.

This growing tension reveals a deeper truth: Despite its claims to be doing otherwise, the Trump family has no interest in building a coherent, transparent crypto regime for the public good. Its goal is personal enrichment through selective deregulation. By enabling the president to profit while regulatory scaffolding for others lags or stalls, the current system has become one of privatized gain without public reciprocity.

Monetary governance, like so many other domains, has entered the realm of selective empire: one rule for the insiders, another for everyone else.

What colonial charters once granted to private firms—control over currency, commerce, and infrastructure—is now executed digitally. The East India Company had cannons and ships. Today’s digital sovereigns have code, contracts, and exemptions written into law.

II. The democratic deficit

The traditional mechanisms of democratic oversight were designed for a different era, when platforms served users rather than weaponizing dependencies against them. Congressional hearings follow predictable rhythms: subpoenas issued, witnesses called, testimony delivered, reports filed. The entire cycle requires months, sometimes years.

By contrast, software deployments happen overnight, and infrastructure can be tuned to malevolence with a single command.

This temporal mismatch creates what intelligence analysts call a “decision cycle advantage”—the ability to outpace one’s adversaries by acting faster than they can respond. By the time democratic institutions recognize a threat, assess its implications, and formulate a response, the technological landscape has already shifted. The oversight apparatus is left examining yesterday’s problems with yesterday’s tools, while today’s weapons are already in use.

Consider the inspector general system—democracy’s early warning mechanism for institutional capture. On January 20, 2025, the Trump administration dismissed at least 12 inspectors general, a preemptive move that cleared the path for DOGE to operate without scrutiny. When self-proclaimed efficiency experts took control, there were no longer independent watchdogs positioned to question their methods or monitor their access.

AI systems introduce a complexity moat that traditional oversight cannot cross. When DOGE personnel modify Social Security payment algorithms, the changes are technically reversible—but practically irreversible. The new systems integrate with other federal databases, create dependencies across multiple agencies, and generate data flows that existing personnel cannot replicate or replace. Musk’s stated goal had long been to make X an everything app, like China’s WeChat. Rolling back the changes would require rebuilding institutional capabilities that may no longer exist, creating space for a private company to fill the gap.

The international dimension reveals how American allies are learning painful lessons about platform dependency. European governments that once built their digital infrastructure atop American technologies now face the hard truth: Platforms like Starlink, Meta, and Microsoft can be turned against them. European Commission officials are scrambling to replace Microsoft’s cloud services, while Danish ministries migrate from Microsoft Office to open-source tools—not because American tech is inadequate, but because of the widespread perception that it is no longer trustworthy.

This erosion of accountability serves a larger vision. Cryptocurrency entrepreneur Balaji Srinivasan calls it “the network state”—virtual nations that form online before establishing physical presence. The model applies startup methodology to nation-building: Develop digital systems and economic relationships, integrate them with existing government infrastructure, then gradually shift authority from elected institutions to algorithmic ones controlled by private interests. In the hands of entities like Musk’s DOGE, the network state becomes less a vision of exit and more a blueprint for capture.

III. Reclaiming we, the people

The current trajectory toward privatized corporate governance is not inevitable—but the window for alternatives is narrowing. As European allies race to build what they call “EuroStack”—an EU-led digital supply chain independent from weaponizable American platforms—democratic societies face a stark choice: reclaim technological sovereignty or accept permanent subordination to private interests with their own agendas.

It can be done. Taiwan offers the most compelling model for democratic technology governance. Rather than choosing between corporate efficiency and democratic oversight, its government has deployed tools like Pol.is—a digital platform designed for facilitating cooperation across difference—that enable thousands of citizens to participate directly in policy formation. The result demonstrates that democracy can move at technological speed—when paired with infrastructure designed to serve the public good, not private profit.

This approach scales beyond transportation policy. These tools can now inform decisions on budget allocation, environmental regulation, and tech policy itself. The key insight is that artificial intelligence can enhance democratic participation rather than replacing it with weaponizable private control. AI can process input, model outcomes, and surface areas of consensus—but the fundamental decisions must remain with the people affected.

The technical requirements are straightforward but urgent. Democratic AI governance requires open-source algorithms, transparent training data, and citizen oversight of system development. Rather than surrendering governmental functions to private companies capable of capturing or manipulating them, democratic societies must build public-private partnerships that preserve public control over essential infrastructure—even as they harness private innovation.

EuroStack represents one such path—slower and more expensive than its American counterparts, but more resilient. As French President Emmanuel Macron declared: “We must reinforce our independence.” The European Union is assembling joint defense funds, satellite networks, and payment systems precisely to escape reliance on American platforms that can be turned against them.

With every new database integrated and dependency hardened, the window of opportunity for democratic options begin to close. The choice is still ours, but the time to act is now. Democracies can reclaim control over critical infrastructure—or continue outsourcing it to corporate entities that increasingly resemble the East India Company: efficient, unaccountable, and sovereign in all but name.

As American allies have discovered, platform dependency is a trap that snaps shut when you least expect it. The question facing democratic societies is whether they will escape this trap while they still can, or whether they will remain subject to the whims of unelected digital sovereigns.

Everything scientists most value—objectivity, truth-seeking, skepticism and transparency—is at stake. These digital sovereigns are no longer merely connecting the world—they are remaking it. Whether this transformation serves public values or corporate profits will decide not only the future of technology—but the fate of self-governance.

“The right to search for truth, implies a duty,” warned Albert Einstein. “One must not conceal any part of what one has recognized to be true.” The true cost of “efficiency” may be democracy itself, which is currently at risk of becoming just another social atavism of the analog age.

关于《The AI Raj: How tech giants are recolonizing power》的评论


暂无评论

发表评论

摘要

The rise of AI firms and tech giants mirrors historical corporate power grabs like those by the East India Company, where commercial efficiency gradually led to imperial governance. Modern technology companies are now leveraging data and software to take over public functions, such as content moderation, digital infrastructure control, and monetary governance through cryptocurrencies. This shift can undermine democratic oversight and accountability, with private entities potentially wielding unchecked influence over critical systems globally. The urgency for democratic societies to reclaim technological sovereignty is emphasized, advocating for open-source algorithms, citizen oversight, and the development of independent digital infrastructures like EuroStack to prevent permanent subordination to unelected digital sovereigns.

相关新闻