The Architecture of Control

The political project described in Gil Durán’s The Nerd Reich and Jacob Silverman’s Gilded Rage is no longer best understood merely as an aspiration to subordinate government to a Silicon Valley faction’s will. It has achieved meaningful, though incomplete, institutional power: through executive action, public procurement, AI deployment, immigration enforcement, defense contracts, donor access, and attacks on the independent civil service. It has advanced fastest where presidential directives, procurement authority, and discretionary enforcement can move quickly; it is less successful where courts, federalism, public resistance, fiscal constraints, technical failure, and the sheer complexity of governing impose limits.

The immediate danger is not a fully consolidated techno-dictatorship. It is a government increasingly reorganized so that public authority serves a narrow fusion of executive power, national-security priorities, extraction, venture-backed technology, and ideologically aligned wealth—while democratic oversight and public-serving capacity weaken. Evidence from democracy monitors, federal actions, AI procurement, and immigration technology indicates that this shift is well underway.

Here, “advance” does not mean total control over the state. It means measurable movement in four areas: political control over administration, integration of private technical systems into coercive functions, public underwriting of favored technology sectors, and erosion of democratic checks and avenues for challenge.

The project’s advance can be measured in concrete institutional changes. The administration created Schedule Policy/Career, transferring roughly 8,000 policy-influencing career positions into a new excepted-service category and substantially weakening incumbents’ civil-service protections, making removal easier. It has used federal leverage to challenge state-level AI regulation, including an executive order on a national AI policy framework allowing agencies to condition discretionary grants on states’ agreement not to enforce AI rules deemed inconsistent with national policy. At the same time, private technology firms are becoming embedded in coercive government functions: ICE awarded Palantir contracts to develop and maintain ImmigrationOS, a system designed for enforcement targeting, self-deportation tracking, and more efficient removal logistics. Meanwhile, the Army granted Palantir a ten-year enterprise agreement with a ceiling of up to $10 billion, consolidating about 75 existing software and data contracts into a single procurement channel. These developments do not demonstrate that a single company or faction controls the government. They demonstrate something more durable: public administration is being reorganized around executive loyalty, proprietary technical infrastructure, and private firms whose commercial interests increasingly align with the state’s surveillance and immigration-enforcement priorities and, as one advocacy analysis argues, its wider military-security architecture.

The Project: Not Abolishing Government, But Changing Its Purpose

The language of Silicon Valley’s anti-state right can be confusing because it often celebrates “exit,” decentralization, crypto, private cities, deregulation, and freedom from bureaucracy. But the most influential version of this politics does not actually seek the disappearance of government. It seeks a state with different priorities and different beneficiaries.

Its priorities are more visibly concentrated in border enforcement, defense technology, energy and data buildout, and deregulation than in labor enforcement, environmental protection, social provision, or an independent professional civil service. The governing model favored by this coalition concentrates state capacity in securing borders, purchasing surveillance tools, funding defense technology, protecting favored investment sectors, expanding energy and data infrastructure, and shielding favored industries from regulation. By contrast, it has been more willing to weaken or subordinate the institutions and protections that constrain concentrated power: labor enforcement, environmental safeguards, AI regulation, public housing, social insurance, an independent professional civil service, and democratic checks on executive and corporate authority.

That is the bridge between Gil Durán’s The Nerd Reich and Jacob Silverman’s Gilded Rage. Durán explains the ideological vocabulary: elite sovereignty, technological destiny, the rejection of democratic equality, the fantasy of “exit,” and the apocalyptic promise that disruption will liberate superior individuals from public constraints. Silverman supplies the institutional and material account: tech wealth converts itself into political donations, personnel networks, land acquisition, defense contracting, data infrastructure, and privileged access to executive power.

The result is not a government captured by one company or one billionaire. It is a coalition project. Its participants do not agree on every issue, and their interests can conflict. But they share a broad preference for executive speed, weak regulation, anti-union policy, technological acceleration, national-security spending, immigration enforcement, and an administrative state made more vulnerable to the removal or marginalization of officials able or willing to resist presidential directives and corporate influence.

What Has Already Changed

By the available evidence, the movement has advanced furthest in four interlocking areas: control over administrative capacity, the expansion of surveillance and immigration enforcement, public subsidization of AI infrastructure, and the weakening of democratic checks.

1. The civil service is a central battlefield

A democratic state requires more than elections. It requires an administrative apparatus that can deliver services, preserve expertise, enforce law consistently, and resist being transformed into the personal instrument of a president or donor class. That is why the federal workforce is such a strategic target.

Staff reductions, hiring constraints, agency restructuring, and efforts to expand political control over policy-influencing roles do more than shrink government. They can alter what government is able to do. A state that loses inspectors, scientists, engineers, public-health workers, benefits administrators, attorneys, and regulators may remain powerful in coercive domains while becoming incapable of providing reliable services or enforcing obligations against powerful private actors.

The administration has openly treated the federal workforce as an object of executive remaking. Independent measures of American democratic health have registered sharp deterioration: Pew Research summarized multiple indices showing that the United States’ 2025 democratic ratings reached recent or historic lows; Freedom House rated the country at 81 out of 100, citing legislative dysfunction, executive overreach, restrictions affecting expression, and efforts to weaken anticorruption safeguards.

This does not mean that the civil service has been entirely conquered. Federal workers retain legal protections; courts can block unlawful action; agencies possess institutional memory; unions, whistleblowers, inspectors general, journalists, and state governments can still expose abuses. But the project does not require absolute conquest to succeed. It only needs to make resistance harder, expertise scarcer, enforcement less independent, and policy more vulnerable to executive command.

2. The immigration system is becoming a test case

The clearest example of technology serving coercive state power is immigration enforcement. Here, the abstract language of “AI,” data integration, and operational efficiency becomes concrete: systems identify people, prioritize targets, merge datasets, guide field operations, and scale detention or removal decisions.

Palantir’s role is especially revealing. ICE’s longstanding sole-source relationship with the company had already exceeded $145 million before the expansion of ImmigrationOS, a platform intended to support enforcement operations, including target identification, self-deportation tracking, and removal logistics. Separately, reporting on Palantir’s ELITE tool indicates that it draws on government data to identify and prioritize enforcement targets through advanced analytics, generating dossiers and address-confidence assessments for people potentially subject to arrest or deportation. Together, these systems illustrate how immigration enforcement is being reorganized around proprietary data infrastructure: political decisions about whom to pursue are translated into technical processes that are difficult for the public to inspect, contest, or appeal.

This is not simply a technical upgrade. It changes the operational scale of enforcement. A government that combines commercial data, databases, predictive or prioritization tools, facial-recognition systems, license-plate-reader networks, and broad administrative discretion can make immigration control more pervasive, rapid, and difficult to challenge.

The civil-liberties issue is not that computers make mistakes—though they do. It is that technical systems can convert political decisions into administratively opaque processes. Who is designated high priority? What data are used? What errors are tolerated? How can a person know they have been flagged? Can they correct the record? Who is accountable when a system directs state attention toward the wrong person? The risk is not hypothetical. In 2026, Kimberlee Williams spent six months in jail after Maryland police relied on a faulty facial-recognition match to identify her as a suspect; she was later identified as the fourteenth publicly known person in the United States wrongfully arrested after police reliance on the technology. The case exposes the accountability vacuum at the center of algorithmic enforcement: when a technical system directs state attention toward the wrong person, responsibility can be dispersed among the agency that supplied the data, the officers who acted on it, the vendor whose system produced the lead, and the officials who authorized its use.

The same basic danger extends beyond immigration. Once infrastructure exists to assemble personal information, identify targets, and automate prioritization, its techniques can migrate across institutional domains. The Electronic Frontier Foundation notes that algorithmic decision systems already shape consequential decisions made by police, courts, public agencies, employers, landlords, and insurance companies, often with insufficient transparency or meaningful recourse for affected people. This does not guarantee universal repression; it identifies the architecture through which selective enforcement can become routine, portable, and difficult to contest.

3. AI is being installed before democratic governance catches up

The AI agenda is perhaps the movement’s most consequential long-term achievement. The federal government is not merely adopting a few productivity tools. It is rapidly creating procurement pathways, cybersecurity partnerships, defense deployments, scientific-data integrations, and administrative systems that make private AI firms integral to public power.

A tracker maintained by the Revolving Door Project reported nearly 3,000 AI uses across 29 federal agencies in 2025—an increase of more than 75 percent over 2024. It describes AI’s use in analyzing sensitive data and delivering public services, raising the prospect that some functions previously handled by public employees may be automated or supplemented. Tech Policy Press reported that the Pentagon rolled out Google’s Gemini for Government as its largest generative-AI deployment to date, while the IRS began rolling out Salesforce’s Agentforce; it also noted an executive order directing federal agencies to challenge state AI regulations deemed “onerous.”

The administration’s AI-security initiatives explicitly envision closer collaboration between government and industry. A June 2026 White House fact sheet announced mechanisms for AI-enabled cybersecurity tools and services across federal agencies, state and local authorities, critical-infrastructure operators, and “trusted partners,” while resisting mandatory licensing, pre-clearance, or permitting for the development and distribution of AI models.

There are defensible reasons for governments to use advanced computing. Cybersecurity, scientific research, accessibility, medical analysis, fraud detection, and emergency logistics can all benefit from accountable tools. The issue is governance: whether the technology is transparent, auditable, appealable, privacy-protective, publicly accountable, and deployed only where its risks are justified.

The present direction favors speed, industry access, centralized technical power, and national-security framing. It risks producing a government that becomes dependent on proprietary systems it cannot fully inspect, explain, or replace. That is a form of structural capture. A contractor need not dictate every policy if agencies become operationally reliant on its infrastructure, personnel, data architecture, and definitions of risk.

4. Public resources are underwriting private technological power

The most basic contradiction of the tech-right program is that it profits from the public systems it disparages. SpaceX, Palantir, cloud providers, chip manufacturers, AI laboratories, defense startups, and surveillance companies all depend, in varying degrees, on public research, contracts, subsidies, energy systems, land, legal protection, and security policy.

The emerging arrangement is not laissez-faire capitalism. It is a form of state-backed technological corporatism: public risk, private ownership, public procurement, private data infrastructure, and national-security justification.

The scale matters. The administration’s policies have increasingly described AI dominance as a national priority, which turns private-sector buildout into a public mission. Analysts have warned that an “AI First” strategy embeds AI expansion in the bureaucracy, uses public procurement to reinforce the industry’s business model, and frames military and surveillance demand as necessary to win a geopolitical race.

This is exactly the transformation that Gilded Rage anticipates. A venture-capital and defense-tech ecosystem does not remain on the outside of the state, selling optional products into a neutral market. It becomes a governing partner, shaping which problems are legible, which solutions are funded, and whose expertise becomes indispensable.

How Well Is The Project Going?

The short answer is: better than its critics should be comfortable admitting, but much less completely than its supporters may imagine.

It is progressing rapidly because its actors have three advantages.

First, they possess concentrated wealth. Political donations, super PACs, think tanks, legal organizations, media ownership, lobbying networks, and personal relationships can influence agendas before the public ever votes on a policy.

Second, they possess institutional access. Technology companies supply systems the government increasingly believes it needs for defense, cybersecurity, intelligence, logistics, communications, AI research, and administration. This access becomes more powerful under an executive branch that treats regulation as a barrier rather than a democratic safeguard.

Third, they possess an emergency narrative. China, immigration, “wokeness,” crime, bureaucratic inefficiency, AI competition, energy scarcity, and national decline can all be presented as crises requiring speed, discipline, deregulation, security contracting, and obedience to technologically fluent elites.

But the project has profound weaknesses.

The Elites Cannot Escape Material Reality

The Long Emergency supplies the material limit on elite fantasies of control. A government can expand surveillance, automate paperwork, enrich contractors, militarize borders, and build data centers. It cannot automate its way out of water scarcity, extreme heat, ecological degradation, failing infrastructure, supply-chain fragility, public-health crises, or the social breakdown produced by inequality.

Complex societies depend on trust, skilled labor, public health, repair capacity, education, stable communities, energy systems, water, food, housing, and legitimate institutions. A regime that weakens these foundations while concentrating resources in defense, extraction, and private technological infrastructure may gain short-term command while making the underlying society less governable.

This is the central contradiction of selective protection. Elite enclaves, secure facilities, private security, air conditioning, data centers, and alternative jurisdictions can provide temporary insulation. They cannot independently reproduce the broad social and ecological conditions necessary for a sophisticated society to function. The people who build, maintain, transport, teach, heal, clean, farm, repair, and govern remain necessary.

Courts, States, and Civil Society Still Matter

American institutions have been damaged, but they have not disappeared. Federal courts, state governments, local governments, nonprofit organizations, unions, investigative journalism, university researchers, public-interest lawyers, civil servants, and organized communities retain real capacity to contest executive action.

This resistance is uneven. It is expensive, slow, and often defensive. But it matters. Federalism prevents total uniformity. Courts can delay or invalidate actions. State privacy, labor, environmental, and consumer-protection laws can constrain some forms of corporate power. Local resistance can obstruct private-city schemes, surveillance programs, and infrastructure projects. Public backlash can alter the political price of extremism.

The struggle is not between omnipotent elites and powerless citizens. It is between a highly organized coalition with exceptional resources and a fragmented but still substantial architecture of democratic resistance.

Corporate Interests Are Not Identical

Silicon Valley is not one party. Defense contractors, consumer platforms, chip companies, cloud providers, crypto firms, AI labs, electric-vehicle manufacturers, venture funds, and social-media companies have overlapping but not identical interests. Some benefit from deregulation; others need stable public investment, international markets, skilled immigration, research universities, grid reliability, and consumer confidence. Nationalist trade policies, political chaos, resource shortages, and international conflict can harm the very businesses that seek influence.

This makes the coalition powerful but unstable. Its members may share hostility toward regulation and democratic constraint, yet they disagree about how much state intervention they want, where public money should go, which technologies should dominate, and how much instability they can tolerate.

The Deeper Measure of Success

The project’s success should not be measured solely by whether a particular executive order survives review, a contract is awarded, or a billionaire’s preferred candidate wins an election. Its deeper success would be cultural and institutional: making it seem natural that technology companies should govern public functions, that the public sector should depend on proprietary systems, that democratic deliberation is an inefficient obstacle, that social protection is fiscally unrealistic, and that security requires ever more surveillance.

That normalization is already far advanced. The language of “innovation,” “efficiency,” “AI leadership,” “national competitiveness,” and “security” often narrows the range of imaginable alternatives. Questions about rights, worker power, privacy, environmental limits, public ownership, due process, and democratic control can be dismissed as nostalgic, naïve, anti-technology, or hostile to national strength.

That is why Durán’s ideological history matters. The essential political victory is not merely creating a private city or installing an AI system. It is convincing the public that those things are inevitable and that resistance itself is irrational.

What Remains Unresolved

The project has not solved its core dilemma: it wants the benefits of a complex, stable, technologically advanced civilization while weakening the public institutions, ecological safeguards, and social reciprocity that make such a civilization possible.

A government increasingly shaped by billionaire networks and defense-tech priorities may become highly capable of monitoring, sorting, and coercing people amid chronic scarcity. Yet it may remain unable—or unwilling—to secure the material preconditions of a livable society: housing, health care, clean water, affordable energy, climate adaptation, and institutions people can trust. In an era of cascading climate shocks, fiscal strain, displacement, and deteriorating public systems, that imbalance is not merely a policy failure. It creates the conditions for permanent emergency governance, in which surveillance and selective enforcement substitute for social provision, and instability becomes something to contain rather than resolve.

The most realistic danger is not a seamless, all-powerful technocratic regime. It is a brittle and unequal order: technologically advanced in protected sectors, highly capable of sorting, monitoring, and coercing targeted populations, yet increasingly unable to sustain the ordinary conditions of collective life.

That is the authoritarianism most compatible with the Long Emergency. It does not offer broad prosperity or stable normality. It offers defended enclaves of privilege, automated administration of scarcity, and harsher containment for people and places the system cannot—or chooses not to—protect.