Palantir: When the State Leases Out Its Nervous System

The real question is not whether Palantir has too much power. The real question is how much sovereignty Western states have already transferred to private platforms before admitting what is happenin

Apr 24,2026

This is not just a software company. It is a political infrastructure entering intelligence, healthcare, defense, and borders—while openly publishing a doctrine that calls on Silicon Valley to merge itself back into Western military power.

For years, the public was told the issue was privacy. Then it was told the issue was efficiency. The real issue is much larger: software no longer merely supports the state. It increasingly provides the language through which the state sees, classifies, integrates, prioritizes, and acts. And when the company supplying that infrastructure also publishes its own political doctrine, the question is no longer merely technological. It becomes a question of national sovereignty, democratic resilience, and the protection of citizens.

Palantir should not be read as an ordinary tech firm with strong opinions. It should be read as an actor already embedded in sensitive state functions across multiple countries while openly advancing a worldview: Silicon Valley owes a “moral debt” to the nation, hard power in this century will be built on software, AI weapons are inevitable, and hollow pluralism is no longer enough. That is not a hostile interpretation of the text. It is the substance of the 22-point summary Palantir published from Alex Karp and Nicholas Zamiska’s The Technological Republic, and it is consistent with the book’s own publisher-facing descriptions that frame technology, government, and geopolitical power as inseparable.

The point most people miss: Palantir does not simply sell tools

The standard reassurance is always the same: the data remains under public control. But that addresses only one layer of the problem. One thing is legal ownership or formal custodianship of data. Another is substantive control over the infrastructure that makes data operational. When a public authority buys not only software, but also integration, support, maintenance, deployment assistance, and decision environments, it is not purchasing a neutral utility. It is outsourcing the technical grammar through which fragmented archives are joined, interpreted, ranked, and turned into action.

In plain terms: the data may remain “yours,” but if you need an outside proprietary platform to make that data speak, connect, visualize, and shape operational choices, then sovereignty is no longer intact. It is conditional. And the deeper that platform enters the daily routines of the state, the higher the technical, political, and institutional cost of exit becomes. That is why a debate framed only around privacy is too narrow. The deeper issue is functional lock-in.

France: when emergency hardens into structure

France offers a textbook example of how dependence becomes normalized. Palantir entered the DGSI in 2016, in the aftermath of major terrorist attacks. Reuters reported in 2019 that France wanted to build a domestic alternative, but in the meantime would have to renew its Palantir relationship because there was no equivalent sovereign solution ready in the short term. At the time, DGSI chief Nicolas Lerner said the red line was that all intelligence data had to remain on a closed, top-secret internal network, even while admitting that a French technological alternative would be preferable.

By December 2025, that temporary fix had become a long-running structure. Le Monde, citing AFP and company statements, reported that the DGSI renewed Palantir for another three years. The renewed agreement covered not just the proprietary software platform but also the integration, support, and assistance services necessary for deployment and operational use. In other words: not merely a license, but an operating relationship.

This is how contemporary dependence works. Emergency justifies the initial entry. Lack of alternatives justifies renewal. Renewal produces operational habituation. And eventually the “temporary” exception turns into a durable architecture. French authorities have insisted that the data remains in France and under full national control. That matters. But it does not eliminate vulnerability, because dependence does not arise only from where data is stored. It also arises from who controls the environment that makes data strategically usable.

The United Kingdom: healthcare and defense inside the same ecosystem

In the United Kingdom, the picture is even more revealing because it shows convergence between welfare-state functions and military-state functions. On one side sits the NHS Federated Data Platform, awarded in 2023 in a deal worth up to £330 million over seven years. On the other sits Palantir’s expanding role in defense: a strategic partnership announced by the UK government in September 2025, up to £1.5 billion in promised investment, London designated as its European defense headquarters, and joint work on AI-enabled capabilities already tested in Ukraine to accelerate decision-making, military planning, and targeting.

That combination is politically significant. When the same company becomes central to both public healthcare infrastructure and defense modernization, the issue is no longer a single controversial contract. It is the construction of a systemic position inside the state. Hansard records that Palantir was treated as a “strategic supplier” to the Ministry of Defence, and that the December 2025 contract worth £240 million was awarded directly, without a competitive process. MPs challenged precisely this point: why bind strategic state capacity to one foreign vendor through direct award?

Healthcare as the rehearsal space for lock-in

The NHS case is especially important because it shows how dependence can be built under the banner of public benefit. NHS England says Palantir cannot commercialize NHS data, cannot use NHS data to train AI models, and acts only under NHS instruction rather than as the data controller. Those safeguards matter and should be reported honestly. The official FAQs are explicit on this point.

But this legal reassurance does not settle the deeper sovereignty question. The same NHS material makes clear that the Federated Data Platform is being delivered using Palantir’s technology stack, and critics in Parliament have asked the most important question of all: could the NHS switch smoothly to another provider if it chose to? That is the real test of technological sovereignty—not whether a contract is formally compliant today, but whether a public institution remains genuinely free to leave tomorrow without institutional trauma.

This is why the NHS case matters far beyond healthcare. Once a platform becomes entwined with waiting lists, bed management, diagnostics, and operational planning, the state does not merely use the technology. It begins to organize itself around it. And once that happens, exit becomes politically and operationally expensive. The promise of efficiency then turns into a mechanism of dependence.

The United States: from contractor to war infrastructure

In the United States, the trajectory is already further advanced. Reuters reported on March 20, 2026 that the Pentagon will formalize Palantir’s Maven as a “program of record,” meaning a core military program with stable funding and long-term institutional adoption. Reuters described Maven as a command-and-control platform that analyzes battlefield streams from satellites, drones, radar, and other sensors to identify threats and support targeting and operational coordination. It also reported that the system had already supported thousands of US strikes on Iran in the preceding weeks.

At that point, the claim that Palantir sells mere “support tools” becomes hard to sustain. A platform embedded in command-and-control is part of the grammar of war itself: sensor fusion, visualization, alerts, target identification, operational acceleration, and compressed decision time. Palantir argues that humans remain responsible for final lethal decisions. That defense matters, but it does not close the ethical or democratic question. The issue is not only who presses the final button. The issue is how much of the path leading to that button has already been structured by a private, proprietary, and politically aligned system.

Borders, migrants, dissent: the same architecture, different targets

The model does not remain on the battlefield. Reuters reported in early 2026 that Palantir had won a surveillance-related ICE contract tied to immigration enforcement, and also noted that it had become the agency’s largest single award in that area as of June 2025. Around the same period, Amnesty International argued that AI-powered surveillance tools made by Palantir and Babel Street were being used by US authorities in ways that posed risks to migrants, refugees, asylum seekers, and international students speaking out for Palestinian rights.

That continuity matters. The same technical architecture that promises efficiency in hospitals or superiority on the battlefield can be repurposed for border enforcement, visa scrutiny, protest monitoring, and algorithmic risk classification. The public language changes. The operational logic does not: gather, fuse, classify, prioritize, act. And the more the state becomes accustomed to this chain, the more the citizen is treated not as a rights-bearing subject but as a data-bearing object of administrative processing.

Israel: the piece Western debate too often brackets out

A complete reading cannot stop at Washington, London, and Paris. A crucial part of the ecosystem in which Palantir operates also runs through Israel, and that role is not incidental. Reuters reported in 2024 that Palantir’s work with Israel was serious enough for Storebrand, a major Norwegian investor, to divest over concerns that the company’s activities could expose it to breaches of international humanitarian law and human rights standards. The same Reuters reporting tied the decision directly to Palantir’s support for Israel and its operations in the occupied Palestinian territories.

The point here is not to overstate what open evidence can prove. It is to recognize that the problem is not merely “American.” It is transnational. It is a techno-military bloc in which Big Tech, security services, cloud infrastructure, AI-enabled analysis, and wartime operations converge. When that ecosystem then enters European public institutions as a partner in healthcare, defense, policing, or intelligence, the issue is not only who stores the data. It is which model of power arrives with the platform.

Why the doctrine matters

This is why Palantir’s manifesto matters. If a company already embedded in defense, intelligence, healthcare, and immigration openly declares that Silicon Valley has a moral obligation to serve national defense, that hard power in the twenty-first century will be built on software, that AI weapons are inevitable, and that some cultures are “regressive,” it is not simply offering abstract philosophy. It is supplying a moral vocabulary for the fusion of private innovation, coercive state power, and civilizational hierarchy.

That is also why the word “technofascism,” used by critics and amplified by Al Jazeera, has traction even when used polemically. Not because a label should replace analysis, but because the text itself combines recognizable elements: moral duty to the nation, normalization of technological violence, contempt for “vacant” pluralism, hierarchical judgments about culture, and a call to realign technical elites with military-state power. Al Jazeera’s reporting is useful not because it proves every allegation at the same evidentiary level, but because it correctly identifies that these themes are not external projections onto the text—they are already in the text.

What this means for national sovereignty

Sovereignty is not lost only when a country loses territory, industrial capacity, or fiscal control. It is also lost when it outsources the technical layer through which institutions classify the world, integrate archives, and turn rights, emergencies, and threats into operations. If that grammar is supplied by an external, proprietary, politically aligned, unelected actor, democratic autonomy narrows even when every contract is formally lawful.

For citizens, that has at least three concrete implications. First, protection cannot be reduced to privacy, because even “properly governed” data can sit inside infrastructures that generate strategic dependence. Second, efficiency cannot be the sole test of public value, because a platform that improves outcomes can also become a structural capture point inside the state. Third, transparency on paper is not enough without real public audit capacity, reversibility, and institutional ability to replace the system without paralysis.

The question that remains

The real question is not whether Palantir has too much power. The real question is how much sovereignty Western states have already transferred to private platforms before admitting what is happening. For years, the debate has been softened by a reassuring formula: “the data remains ours.” But that is no longer the decisive metric. The decisive metric is who controls the environment that makes data operational, who updates it, who owns the underlying intellectual property, who shapes public workflows, and how realistic exit still is.

If there is one critical consciousness readers should take away from this, it is this: national sovereignty cannot be defended by outsourcing the state’s nervous system. And citizen protection cannot rest on a model in which the same technical architecture can, depending on context, reduce a hospital waiting list, accelerate military targeting, tighten a border, or classify dissent. Once software enters that level of power, it is no longer neutral. And once the company selling it publishes its own doctrine of civilization, the silence of institutions is no longer prudence. It begins to look like surrender.

What is needed now is not passive outrage but precise public scrutiny. Who controls the technical grammar of the state? What is the exit plan? Which public functions should never be entrusted to a single vendor? What truly independent audit exists over these systems? And how many public decisions are already being prepared, accelerated, or shaped inside operational environments citizens never chose and almost never understand?

This is where real civic awareness begins: reading the contracts, not just the press releases; examining parliamentary exchanges, not just PR claims; comparing promised efficiencies with dependency costs; and refusing the rhetoric that confuses innovation with inevitability. That work is slow. But without it, sovereignty will continue to be hollowed out under the language of modernization.

Sources

Penguin-linked publisher descriptions and public summaries of The Technological Republic; Palantir’s published 22-point summary; Reuters on Pentagon Maven; Reuters on ICE and surveillance growth; Reuters and Le Monde on the DGSI renewal; Hansard on UK Ministry of Defence contracts; NHS England FAQs on the Federated Data Platform; the UK government’s September 2025 Palantir partnership announcement; Reuters on Storebrand’s divestment over Palantir’s work connected to Israel; Amnesty International on AI-powered surveillance targeting migrants and pro-Palestinian students.