Skip to content

What Does Palantir Actually Want? Power, Data, and Control

M
Marcus Webb
June 17, 2026
11 min read
Business & Money
What Does Palantir Actually Want? Power, Data, and Control - Image from the article

Quick Summary

Palantir's stock is up 1,400% since IPO. But what does the company actually want? A deep dive into its ideology, platforms, and global ambitions.

In This Article

Palantir's Stock Is Up 1,400%. But Money Isn't the Point.

Palantir Technologies went public in September 2020 at $9.50 per share. Today, it trades above $140 — a return of more than 1,400% in under five years. Its market cap sits around $330 billion, placing it inside the top 40 most valuable companies on Earth. It holds government contracts on multiple continents. It counts six branches of the US military and over three dozen federal agencies among its clients.

So here is the obvious question: if Palantir already has the money, the contracts, and the political access, what else does it want?

The short answer is this: Palantir wants to become the default AI and data infrastructure layer for institutional power across Western civilisation. But the longer answer is stranger, more ideologically charged, and frankly more consequential than most people realise.

This is not a story about a software company optimising for shareholder returns. It is a story about a group of people who believe they are on the right side of a civilisational conflict — and who are building the tools to win it.


Two Platforms, Two Very Different Realities

Palantir confuses a lot of observers because the company operates two structurally distinct platforms that serve radically different purposes.

Foundry is the commercial arm. It ingests operational data from private-sector clients and turns it into real-time decision intelligence. The client list reads like a Fortune 500 roll call: Airbus, Ferrari's F1 team, Nike, Morgan Stanley, United Airlines, Honda, Ford, BP, PwC, and dozens more. Foundry is used to optimise supply chains, flag financial anomalies, and accelerate product development. By most measures, it is a legitimate and powerful enterprise software product.

Gotham is something else entirely. Palantir describes it as an "operating system for global decision-making," which is deliberately vague. In practice, Gotham is a surveillance and targeting platform built for government agencies. Its capabilities include:

  • War targeting — selecting and prioritising military targets for destruction
  • Domestic mass surveillance — aggregating data streams from law enforcement, immigration, health records, tax databases, and license plate readers into a single searchable profile on any individual
  • Battlefield analysis — providing real-time intelligence to troops in active conflict zones

Researcher Nicole Bennett describes Gotham as a system that "takes fragmented data scattered across various agencies stored in different formats and transforms it into a unified searchable web." That is not a neutral capability. Unified, searchable profiles on citizens — built from Medicare records, IRS data, travel history, biometric data, phone location data, and license plate tracking — represent a qualitative shift in state power.

For context: there are over 80,000 automated license plate recognition cameras on US roads and highways. That data is being fed into Palantir systems. If you drive in the United States, your movements, your vehicle, and your patterns are almost certainly already part of that dataset.


The Ideology Behind the Infrastructure

Most enterprise software companies do not have a publicly stated worldview. Palantir does.

CEO Alex Karp — who holds a PhD in social theory from Frankfurt, famously wears tracksuits to investor meetings, and stores his phone in a Faraday cage — has been unusually candid about what he believes. He has argued that Western defence and intelligence institutions need advanced AI-powered military software for one explicit reason: maintaining global superiority. Not parity. Not deterrence. Superiority.

Karp has said on record: "Our product is used on occasion to kill people." He has floated the idea of reintroducing military conscription. He flew personally to Tel Aviv in 2024 to sign a deal with the Israel Defense Forces, supplying automated decision-making for targeting in the Gaza conflict.

This is not the language of a CEO managing quarterly earnings. This is the language of someone who believes his company is fighting a war — and that the war is worth winning at almost any cost.

The ideological framework matters because it shapes how Palantir makes product decisions, chooses clients, and draws — or declines to draw — ethical boundaries. When Karp says "we are fighting for what we believe and putting in a product to give the people that agree with us a superior position," that is not marketing copy. It is a mission statement.


What Does Palantir Actually Want? Power, Data, and Control

How Palantir Locks In Clients — and Why Leaving Is Almost Impossible

From a pure business mechanics standpoint, Palantir's model is built around a concept researchers call vendor lock-in — and it executes it more thoroughly than almost any other enterprise software company.

Here is how it works:

  1. Deep embedding at the point of implementation. Palantir does not just sell software. It sends employees — sometimes for months — to work inside client organisations. Early in the company's history, staff embedded with active military units in conflict zones, working on laptops while troops exchanged gunfire nearby. For commercial clients, employees would spend months in bank operations offices just to understand the workflow before writing a single line of code.

  2. Layered use cases over time. Once a client sees value in one application, Palantir proposes adjacent use cases. Each new layer increases the cost of switching, because more of the client's decision-making becomes dependent on Palantir's data architecture.

  3. Proprietary data structures. As Nicole Bennett and others have noted, Gotham transforms disparate, incompatible data sources into a unified format. But that unified format is Palantir's format. Competitors like Snowflake, Databricks, Microsoft Azure, and AWS offer alternative infrastructure, but none of them replicate Palantir's depth in complex, multi-source government environments.

The result: a company with approximately 4,400 employees is worth more than $330 billion. That ratio — revenue and valuation per employee — is a direct function of how deeply embedded Palantir becomes in the organisations it serves.

Karp has noted, with characteristic bluntness, that most of Palantir's best clients started as its most vocal sceptics. The sceptics ask the hard questions. When the product answers those questions, the sceptics become the most committed buyers.


Project Maven and the Automation of Lethal Force

Beyond Gotham, Palantir is a key contractor on Project Maven — the US Department of Defense's AI-powered targeting initiative. Maven is, in essence, a faster, more automated version of Gotham's targeting capabilities. It identifies and prioritises military targets with reduced human involvement in the decision loop.

In Ukraine, the practical consequences of this technology have become visible in ways that would have seemed like science fiction a decade ago. Ukrainian forces have deployed AI-assisted drone systems capable of autonomous takeoff, navigation, target recognition, tracking, and terminal strike — the lethal endpoint of the flight. In at least one documented case, a Russian position was captured exclusively by unmanned ground and aerial vehicles, with no Ukrainian infantry involved and no Ukrainian casualties.

Three Russian soldiers surrendered — not to a human combatant, but to a robot.

This is not a future scenario. It is documented recent history. And Palantir's data and targeting infrastructure is part of the architecture that made it possible.

The strategic implication is significant: as autonomous lethal systems become more capable, the software layer that controls targeting decisions becomes the most consequential point of leverage in a conflict. Palantir is positioning itself to own that layer.


The Centralisation Risk Nobody Is Talking About Enough

The most underappreciated risk Palantir poses is not what it does today. It is what its infrastructure makes possible tomorrow.

Consider what Palantir is currently integrating across US federal agencies: Medicare records, IRS tax data, Social Security information, immigration and travel records, health data, police records, biometric data, phone location data, online history, and license plate tracking. Each of these datasets, in isolation, is tightly regulated and legally siloed. Together, in a unified, searchable system, they constitute something that has no real precedent in liberal democratic governance.

Free Weekly Newsletter

Enjoying this guide?

Get the best articles like this one delivered to your inbox every week. No spam.

What Does Palantir Actually Want? Power, Data, and Control

Alex Karp himself has acknowledged the historical parallel. Running the East German Stasi — one of the most comprehensive surveillance states in modern history — consumed roughly 20% of GDP. Today, the same functional capability costs a fraction of that, delivered via AI, cameras, and smartphones.

His argument is that this power is safer in Western hands than in the hands of authoritarian governments. That may be true. But the infrastructure Palantir is building does not come with a constitutional guarantee about who controls it next, or under what political conditions it will be used.

The architecture of surveillance states is not built in a single moment. It is built incrementally, one integration at a time, until the cost of dismantling it exceeds any political will to do so. That is precisely the dynamic Palantir's vendor lock-in model creates — not just commercially, but institutionally.


What Palantir's Trajectory Means for Investors, Professionals, and Citizens

For different audiences, the Palantir story carries very different practical implications.

For investors: The 1,400% return since IPO is real, and the structural drivers remain intact. US government AI spending is accelerating, not contracting. Palantir's commercial Foundry business is growing in parallel. The risk factors are regulatory and reputational — not technological or competitive in the short term. The stock is expensive by traditional multiples, but Palantir is not a traditional company.

For enterprise technology professionals: Palantir's embedding methodology — going deep into client workflows before deploying software — is one of the most effective enterprise sales and retention strategies in the industry. The lesson is applicable well beyond defence contracts.

For citizens: If you live in a country that uses Palantir infrastructure — and that list includes the US, UK, Germany, and Australia — your data is almost certainly part of a system you cannot opt out of, did not consent to, and have limited legal visibility into. That is worth understanding clearly, regardless of your political views.

The question Palantir's existence forces is not whether advanced governments should use AI for defence and security. They will, and in many cases they should. The question is who controls the infrastructure, what values are embedded in it, and what accountability mechanisms exist when it is misused.

Right now, those questions do not have satisfying answers.


Frequently Asked Questions

What does Palantir actually do? Palantir operates two main platforms. Foundry serves commercial clients — companies like Airbus, Nike, and Morgan Stanley — helping them process and analyse large operational datasets. Gotham serves government and defence clients, aggregating data from multiple agencies to enable surveillance, intelligence analysis, and military targeting. Gotham is the more controversial of the two.

Why is Palantir's stock so valuable despite having relatively few employees? Palantir's valuation reflects the strategic depth of its client relationships rather than headcount. Once a government agency or large enterprise integrates Palantir's infrastructure, switching costs become extremely high. This creates durable, recurring revenue streams and pricing power. With roughly 4,400 employees and a market cap exceeding $330 billion, the company's value per employee ratio is one of the highest in enterprise software.

Who is Alex Karp and why does his worldview matter? Alex Karp is Palantir's co-founder and CEO. Unlike most tech executives, he holds a PhD in social theory rather than a business or engineering background. He has been publicly explicit that Palantir's mission is to ensure Western civilisational dominance through AI-powered military and intelligence software. His worldview is not incidental to Palantir's strategy — it is the strategy. Understanding Karp is essential to understanding why Palantir makes the product and client decisions it does.

What is Project Maven and how does Palantir relate to it? Project Maven is a US Department of Defense initiative that uses AI to automate the identification and prioritisation of military targets. Palantir is a major contractor on Maven. The system reduces human involvement in targeting decisions and has been used in active conflicts. It represents the most advanced and controversial application of Palantir's government-facing technology.

Is Palantir's surveillance infrastructure legal? In most jurisdictions where Palantir operates, its activities are technically legal because they occur within the framework of government contracts and national security law. However, legal and ethical are not the same thing. The aggregation of previously siloed data sources — health records, tax data, location data, biometric data — into a single searchable system raises significant civil liberties concerns that existing law was not designed to address. Several privacy researchers and civil liberties organisations have called for new legislative frameworks to govern this type of infrastructure.

Frequently Asked Questions

Palantir's Stock Is Up 1,400%. But Money Isn't the Point.

Palantir Technologies went public in September 2020 at $9.50 per share. Today, it trades above $140 — a return of more than 1,400% in under five years. Its market cap sits around $330 billion, placing it inside the top 40 most valuable companies on Earth. It holds government contracts on multiple continents. It counts six branches of the US military and over three dozen federal agencies among its clients.

So here is the obvious question: if Palantir already has the money, the contracts, and the political access, what else does it want?

The short answer is this: Palantir wants to become the default AI and data infrastructure layer for institutional power across Western civilisation. But the longer answer is stranger, more ideologically charged, and frankly more consequential than most people realise.

This is not a story about a software company optimising for shareholder returns. It is a story about a group of people who believe they are on the right side of a civilisational conflict — and who are building the tools to win it.


Two Platforms, Two Very Different Realities

Palantir confuses a lot of observers because the company operates two structurally distinct platforms that serve radically different purposes.

Foundry is the commercial arm. It ingests operational data from private-sector clients and turns it into real-time decision intelligence. The client list reads like a Fortune 500 roll call: Airbus, Ferrari's F1 team, Nike, Morgan Stanley, United Airlines, Honda, Ford, BP, PwC, and dozens more. Foundry is used to optimise supply chains, flag financial anomalies, and accelerate product development. By most measures, it is a legitimate and powerful enterprise software product.

Gotham is something else entirely. Palantir describes it as an "operating system for global decision-making," which is deliberately vague. In practice, Gotham is a surveillance and targeting platform built for government agencies. Its capabilities include:

  • War targeting — selecting and prioritising military targets for destruction
  • Domestic mass surveillance — aggregating data streams from law enforcement, immigration, health records, tax databases, and license plate readers into a single searchable profile on any individual
  • Battlefield analysis — providing real-time intelligence to troops in active conflict zones

Researcher Nicole Bennett describes Gotham as a system that "takes fragmented data scattered across various agencies stored in different formats and transforms it into a unified searchable web." That is not a neutral capability. Unified, searchable profiles on citizens — built from Medicare records, IRS data, travel history, biometric data, phone location data, and license plate tracking — represent a qualitative shift in state power.

For context: there are over 80,000 automated license plate recognition cameras on US roads and highways. That data is being fed into Palantir systems. If you drive in the United States, your movements, your vehicle, and your patterns are almost certainly already part of that dataset.


The Ideology Behind the Infrastructure

Most enterprise software companies do not have a publicly stated worldview. Palantir does.

CEO Alex Karp — who holds a PhD in social theory from Frankfurt, famously wears tracksuits to investor meetings, and stores his phone in a Faraday cage — has been unusually candid about what he believes. He has argued that Western defence and intelligence institutions need advanced AI-powered military software for one explicit reason: maintaining global superiority. Not parity. Not deterrence. Superiority.

Karp has said on record: "Our product is used on occasion to kill people." He has floated the idea of reintroducing military conscription. He flew personally to Tel Aviv in 2024 to sign a deal with the Israel Defense Forces, supplying automated decision-making for targeting in the Gaza conflict.

This is not the language of a CEO managing quarterly earnings. This is the language of someone who believes his company is fighting a war — and that the war is worth winning at almost any cost.

The ideological framework matters because it shapes how Palantir makes product decisions, chooses clients, and draws — or declines to draw — ethical boundaries. When Karp says "we are fighting for what we believe and putting in a product to give the people that agree with us a superior position," that is not marketing copy. It is a mission statement.


How Palantir Locks In Clients — and Why Leaving Is Almost Impossible

From a pure business mechanics standpoint, Palantir's model is built around a concept researchers call vendor lock-in — and it executes it more thoroughly than almost any other enterprise software company.

Here is how it works:

  1. Deep embedding at the point of implementation. Palantir does not just sell software. It sends employees — sometimes for months — to work inside client organisations. Early in the company's history, staff embedded with active military units in conflict zones, working on laptops while troops exchanged gunfire nearby. For commercial clients, employees would spend months in bank operations offices just to understand the workflow before writing a single line of code.

  2. Layered use cases over time. Once a client sees value in one application, Palantir proposes adjacent use cases. Each new layer increases the cost of switching, because more of the client's decision-making becomes dependent on Palantir's data architecture.

  3. Proprietary data structures. As Nicole Bennett and others have noted, Gotham transforms disparate, incompatible data sources into a unified format. But that unified format is Palantir's format. Competitors like Snowflake, Databricks, Microsoft Azure, and AWS offer alternative infrastructure, but none of them replicate Palantir's depth in complex, multi-source government environments.

The result: a company with approximately 4,400 employees is worth more than $330 billion. That ratio — revenue and valuation per employee — is a direct function of how deeply embedded Palantir becomes in the organisations it serves.

Karp has noted, with characteristic bluntness, that most of Palantir's best clients started as its most vocal sceptics. The sceptics ask the hard questions. When the product answers those questions, the sceptics become the most committed buyers.


Project Maven and the Automation of Lethal Force

Beyond Gotham, Palantir is a key contractor on Project Maven — the US Department of Defense's AI-powered targeting initiative. Maven is, in essence, a faster, more automated version of Gotham's targeting capabilities. It identifies and prioritises military targets with reduced human involvement in the decision loop.

In Ukraine, the practical consequences of this technology have become visible in ways that would have seemed like science fiction a decade ago. Ukrainian forces have deployed AI-assisted drone systems capable of autonomous takeoff, navigation, target recognition, tracking, and terminal strike — the lethal endpoint of the flight. In at least one documented case, a Russian position was captured exclusively by unmanned ground and aerial vehicles, with no Ukrainian infantry involved and no Ukrainian casualties.

Three Russian soldiers surrendered — not to a human combatant, but to a robot.

This is not a future scenario. It is documented recent history. And Palantir's data and targeting infrastructure is part of the architecture that made it possible.

The strategic implication is significant: as autonomous lethal systems become more capable, the software layer that controls targeting decisions becomes the most consequential point of leverage in a conflict. Palantir is positioning itself to own that layer.


The Centralisation Risk Nobody Is Talking About Enough

The most underappreciated risk Palantir poses is not what it does today. It is what its infrastructure makes possible tomorrow.

Consider what Palantir is currently integrating across US federal agencies: Medicare records, IRS tax data, Social Security information, immigration and travel records, health data, police records, biometric data, phone location data, online history, and license plate tracking. Each of these datasets, in isolation, is tightly regulated and legally siloed. Together, in a unified, searchable system, they constitute something that has no real precedent in liberal democratic governance.

Alex Karp himself has acknowledged the historical parallel. Running the East German Stasi — one of the most comprehensive surveillance states in modern history — consumed roughly 20% of GDP. Today, the same functional capability costs a fraction of that, delivered via AI, cameras, and smartphones.

His argument is that this power is safer in Western hands than in the hands of authoritarian governments. That may be true. But the infrastructure Palantir is building does not come with a constitutional guarantee about who controls it next, or under what political conditions it will be used.

The architecture of surveillance states is not built in a single moment. It is built incrementally, one integration at a time, until the cost of dismantling it exceeds any political will to do so. That is precisely the dynamic Palantir's vendor lock-in model creates — not just commercially, but institutionally.


What Palantir's Trajectory Means for Investors, Professionals, and Citizens

For different audiences, the Palantir story carries very different practical implications.

For investors: The 1,400% return since IPO is real, and the structural drivers remain intact. US government AI spending is accelerating, not contracting. Palantir's commercial Foundry business is growing in parallel. The risk factors are regulatory and reputational — not technological or competitive in the short term. The stock is expensive by traditional multiples, but Palantir is not a traditional company.

For enterprise technology professionals: Palantir's embedding methodology — going deep into client workflows before deploying software — is one of the most effective enterprise sales and retention strategies in the industry. The lesson is applicable well beyond defence contracts.

For citizens: If you live in a country that uses Palantir infrastructure — and that list includes the US, UK, Germany, and Australia — your data is almost certainly part of a system you cannot opt out of, did not consent to, and have limited legal visibility into. That is worth understanding clearly, regardless of your political views.

The question Palantir's existence forces is not whether advanced governments should use AI for defence and security. They will, and in many cases they should. The question is who controls the infrastructure, what values are embedded in it, and what accountability mechanisms exist when it is misused.

Right now, those questions do not have satisfying answers.


Frequently Asked Questions

What does Palantir actually do? Palantir operates two main platforms. Foundry serves commercial clients — companies like Airbus, Nike, and Morgan Stanley — helping them process and analyse large operational datasets. Gotham serves government and defence clients, aggregating data from multiple agencies to enable surveillance, intelligence analysis, and military targeting. Gotham is the more controversial of the two.

Why is Palantir's stock so valuable despite having relatively few employees? Palantir's valuation reflects the strategic depth of its client relationships rather than headcount. Once a government agency or large enterprise integrates Palantir's infrastructure, switching costs become extremely high. This creates durable, recurring revenue streams and pricing power. With roughly 4,400 employees and a market cap exceeding $330 billion, the company's value per employee ratio is one of the highest in enterprise software.

Who is Alex Karp and why does his worldview matter? Alex Karp is Palantir's co-founder and CEO. Unlike most tech executives, he holds a PhD in social theory rather than a business or engineering background. He has been publicly explicit that Palantir's mission is to ensure Western civilisational dominance through AI-powered military and intelligence software. His worldview is not incidental to Palantir's strategy — it is the strategy. Understanding Karp is essential to understanding why Palantir makes the product and client decisions it does.

What is Project Maven and how does Palantir relate to it? Project Maven is a US Department of Defense initiative that uses AI to automate the identification and prioritisation of military targets. Palantir is a major contractor on Maven. The system reduces human involvement in targeting decisions and has been used in active conflicts. It represents the most advanced and controversial application of Palantir's government-facing technology.

Is Palantir's surveillance infrastructure legal? In most jurisdictions where Palantir operates, its activities are technically legal because they occur within the framework of government contracts and national security law. However, legal and ethical are not the same thing. The aggregation of previously siloed data sources — health records, tax data, location data, biometric data — into a single searchable system raises significant civil liberties concerns that existing law was not designed to address. Several privacy researchers and civil liberties organisations have called for new legislative frameworks to govern this type of infrastructure.

Z

About Zeebrain Editorial

Our editorial team is dedicated to providing clear, well-researched, and high-utility content for the modern digital landscape. We focus on accuracy, practicality, and insights that matter.

More from Business & Money

Related Guides

Keep exploring this topic

Explore More Categories

Keep browsing by topic and build depth around the subjects you care about most.