Palantir Technologies

Based on Wikipedia: Palantir Technologies

In Tolkien's fantasy universe, a palantír is a seeing stone—a dark crystal orb that lets its user peer across vast distances, spy on enemies, and glimpse possible futures. But the palantíri came with a warning: gaze into one carelessly, and you might find someone gazing back. Worse still, the seeing stones could be used to manipulate what you saw, twisting truth into something that served darker purposes.

When Peter Thiel and his co-founders needed a name for their new company in 2003, they chose Palantir. The reference was intentional. They even named their office locations after places in Middle-earth: The Shire for Palo Alto, Rivendell for McLean, Virginia (conveniently close to the Central Intelligence Agency headquarters), and Minas Tirith for Washington, D.C.

This wasn't whimsy. It was a mission statement.

The Birth of a Surveillance Company

Palantir Technologies emerged from an unlikely source: PayPal's fraud detection systems. When Thiel sold PayPal to eBay in 2002, he walked away with more than money. He had watched his engineers build something remarkable—software called "Igor" (named after a particularly persistent Russian scammer) that could analyze massive amounts of transaction data to catch fraudsters in real time.

The Federal Bureau of Investigation noticed. They wanted something similar to detect financial crimes. But Thiel had bigger ambitions than catching credit card thieves.

Here's where the story gets interesting. According to multiple accounts, Thiel recruited Alex Karp, a former Stanford Law School colleague with a doctorate in philosophy from Frankfurt, to run the company. Karp was an unusual choice for a tech CEO—he practiced Qigong, a Chinese meditative exercise, and spoke often about the tension between security and civil liberties. When asked about Palantir's founding philosophy, Karp described it as a "classic German approach," influenced by the Hegelian dialectic—the idea that opposing forces can be synthesized into something new.

The opposing forces in this case? Preventing terrorism while preserving civil liberties. Palantir's pitch was that they could thread this needle.

The In-Q-Tel Connection

No venture capital firm in Silicon Valley would touch them.

According to Karp, Sequoia Capital's chairman doodled through their entire pitch meeting. A Kleiner Perkins executive reportedly lectured the founders about why their company was destined to fail. The problem wasn't the technology—it was the customer. In the mid-2000s, the government was deeply unpopular in Silicon Valley. The Iraq War was raging, and no one wanted to build tools for the intelligence community.

But someone turned them down and then suggested they try In-Q-Tel.

In-Q-Tel is one of the stranger entities in American capitalism. It's a venture capital firm, but its sole investor is the Central Intelligence Agency. Created in 1999, its purpose is to identify and invest in technologies that might benefit the intelligence community. The investment amounts are typically modest—In-Q-Tel put only about two million dollars into Palantir—but the real value is access. An In-Q-Tel investment means the CIA vouches for you. It opens doors throughout the intelligence world.

For two years after that initial investment, Palantir did something unusual for a tech startup: they continuously revised their software based on feedback from actual intelligence analysts. They weren't building a product and hoping someone would buy it. They were building exactly what their customers needed, because their customers were sitting in the room telling them what to build.

What Palantir Actually Does

To understand Palantir, you need to understand a problem that plagues every large organization: data silos.

Imagine you're an intelligence analyst trying to track a suspected terrorist. You might need information from the FBI's database, the CIA's files, immigration records, financial transaction logs, phone metadata, and social media profiles. Before Palantir, each of these databases was essentially a separate kingdom. Different formats, different access protocols, different user interfaces. Connecting the dots meant manually requesting data from multiple agencies, waiting days or weeks for responses, and then trying to reconcile information that might be organized in completely different ways.

This wasn't just inefficient. It was dangerous. The 9/11 Commission Report identified the inability to share information across agencies as a critical failure that allowed the attacks to succeed.

Palantir's core technology—originally called Gotham, now one of several products—solves this problem. It creates a single interface that can connect to multiple databases simultaneously, normalize the data into common formats, and let analysts search across everything at once. Think of it as a universal translator for databases.

The software doesn't just search, though. It visualizes relationships. If you're investigating a network of people, Gotham can show you who called whom, who transferred money to whom, who traveled to the same locations at the same times. It transforms raw data into a map of connections that human analysts can actually understand and explore.

From Intelligence to Everything

By 2013, Palantir's client list had expanded dramatically. A leaked document revealed they were working with at least twelve groups within the United States government: the CIA, the Department of Homeland Security, the National Security Agency, the FBI, the Centers for Disease Control, the Marine Corps, the Air Force, Special Operations Command, the United States Military Academy at West Point, the Joint Improvised-Threat Defeat Organization (the agency responsible for countering roadside bombs in Iraq and Afghanistan), and the National Center for Missing and Exploited Children.

But Palantir wasn't content to remain a government contractor. They developed a second major product called Foundry, designed for commercial clients. The pitch was essentially the same—connect your siloed databases and make sense of your data—but the applications were different. Morgan Stanley used it for financial analysis. Airbus used it for supply chain management. Pharmaceutical giant Merck used it to accelerate drug development.

The company found an unexpected role during the COVID-19 pandemic. Palantir worked with the National Health Service in Britain and developed Tiberius, a software system that helped the United States allocate vaccine doses across the country. When you're trying to distribute hundreds of millions of vaccines to thousands of locations while accounting for different storage requirements, population demographics, and supply chain constraints, the data integration problem is massive. It's exactly what Palantir was built to solve.

The Controversy Machine

Palantir has never been far from controversy.

Civil liberties organizations, including the American Civil Liberties Union, have criticized the company's work with law enforcement agencies. Multiple police departments have used Gotham for crime analysis, and critics argue this amounts to predictive policing—using data to forecast where crimes will occur or who might commit them, with all the potential for bias and civil liberties violations that implies.

Karp disputes this characterization. He argues that Palantir's software is an analytical tool that requires human judgment, not an autonomous system that independently predicts criminal activity. The distinction matters: a tool that helps detectives analyze evidence from crimes that have already occurred is different from a system that tells police to surveil someone because an algorithm thinks they might commit a crime in the future.

But the controversy runs deeper than policing. Under the Trump administration, Palantir's contracts with Immigration and Customs Enforcement drew intense criticism. The software helped ICE aggregate data and coordinate deportation operations. When the administration's family separation policy made headlines, Palantir found itself associated with some of the most controversial enforcement actions in recent American history.

The company has also faced criticism from within Silicon Valley. When Palantir moved its headquarters from Palo Alto to Denver in 2020, Karp took a parting shot at his former neighbors, dismissing the "engineering elite of Silicon Valley" who think they "know more about how society should be organized or what justice requires."

The Defense Industrial Complex, Reimagined

In 2025, Karp co-authored a book called The Technological Republic with Nicholas Zamiska. The central argument is provocative: American technological dominance requires deeper integration between Silicon Valley and the defense establishment. Karp contends that China operates with fewer ethical constraints than American companies, and that maintaining technological leadership is essential for national security. Deterrence through technological dominance, the authors argue, could prevent many wars.

Critics were not impressed. Bloomberg noted dryly that the atomic bomb—produced by the Manhattan Project, another example of civilian-military technological collaboration—was ultimately used on actual cities full of actual people. The New Republic described Karp's vision as "techno-militarism to advance American global supremacy through hard power and targeted violence."

Palantir responded to these criticisms by leaning in. In 2025, the company launched a fellowship program featuring a four-week seminar on Western civilization and whether it was worth defending. The Wall Street Journal described this as embodying Palantir's "pro-America ethos."

The Valuation Rollercoaster

Palantir's financial history reads like a fever dream.

After raising $30 million in initial funding—mostly from Thiel himself, since venture capitalists wouldn't touch them—the company grew steadily through government contracts. By 2013, they were valued at $9 billion. By late 2015, that number had more than doubled to $20 billion.

Then came the reality check. In 2018, Morgan Stanley valued the company at just $6 billion. The problem wasn't the technology or even the controversy—it was simple business fundamentals. Palantir's software required extensive customization for each client, which meant high costs and limited scalability. They weren't losing money on every sale, but they weren't making money either. The company didn't turn a profit for nearly two decades.

Things changed in 2022, when Palantir reported its first quarter of positive net income under non-GAAP metrics (essentially, accounting that excludes certain expenses). By early 2023, they achieved profitability even under standard accounting rules, earning $31 million in the first quarter.

Then came artificial intelligence.

As large language models and generative AI captured public imagination in 2023 and 2024, investors suddenly saw Palantir in a new light. Here was a company that had spent twenty years building the infrastructure to integrate and analyze data at massive scale. If AI was going to transform government and enterprise, Palantir already had the pipes in place.

By 2025, Palantir's market capitalization exceeded $400 billion, making it one of the most valuable companies in the world. But there was something strange about this valuation. The company's revenue remained modest—nowhere near the $50 billion annually required to qualify for the Fortune 500 list. Its entire workforce numbered just 4,100 people (Karp wanted to reduce it to 3,600).

Palantir had become one of the most valuable companies in the world despite being, by many conventional measures, a small business.

The Machine That Sees Everything

In May 2025, the second Trump administration had spent $113 million on Palantir contracts with the Department of Homeland Security and the Pentagon, with an additional $795 million contract with the Department of Defense in the works. Officials were deliberating contracts with the Social Security Administration and the Internal Revenue Service.

This is where Palantir becomes something more than just another government contractor.

The Social Security Administration maintains records on virtually every American. The IRS knows your income, your assets, your financial relationships. The Department of Homeland Security tracks who enters and leaves the country, maintains databases on immigration status, and monitors potential security threats. The Pentagon has its own vast databases on personnel, logistics, and intelligence.

Palantir's software is designed to connect these databases. To make them speak to each other. To transform isolated pools of information into a unified lake where analysts can see patterns that would be invisible if each database remained separate.

Supporters argue this is exactly what good government should do—use technology to work more efficiently, catch fraud, and protect national security. They point out that Palantir doesn't collect, store, or sell data itself. The company provides software; clients retain control over their own information.

Critics see something darker: the infrastructure for a surveillance state. Once these systems are built and these databases are connected, the temptation to use them for purposes beyond their original intent becomes enormous. The history of surveillance powers in America is a history of scope creep—tools created for one purpose being used for others, safeguards eroding over time, oversight proving inadequate to constrain the systems' growth.

The Philosophical CEO

Alex Karp remains one of the most unusual figures in American business.

He earned his doctorate in social theory from the University of Frankfurt, studying under Jürgen Habermas, one of the most influential philosophers of the twentieth century. He practices Qigong daily. He's known for wandering Palantir's offices in his socks. In interviews, he speaks in philosophical abstractions that would seem bizarre coming from most corporate executives.

But Karp has also proven to be a remarkably effective operator. He maintained founder control of Palantir through years of fundraising by refusing to give board seats to investors—a power move that most startups can't pull off. He navigated the company through controversy after controversy without losing major clients. He brought Palantir to profitability after two decades of losses.

In his book and public statements, Karp positions himself as someone who has thought deeply about the ethics of what he's building. He frames Palantir's work as enabling human judgment rather than replacing it with autonomous systems. He argues that the alternative to American companies building these tools isn't that they don't get built—it's that Chinese companies build them instead, for clients unconstrained by American constitutional protections.

Whether you find this argument persuasive probably depends on how much you trust the institutions that would use Palantir's tools, and how confident you are that safeguards against abuse will hold.

Looking into the Stone

In Tolkien's stories, the palantíri were neither good nor evil. They were instruments that amplified the intentions of whoever used them. In the hands of the wise, they provided knowledge that helped protect Middle-earth. In the hands of the corrupted, they became instruments of domination and deception.

The same might be said of Palantir Technologies. The software itself is neutral—it connects databases and helps humans see patterns. Whether that power serves justice or oppression depends entirely on who's using it and to what end.

But here's what makes Palantir different from a magical artifact: it's a business. It has shareholders expecting returns, employees expecting paychecks, and a growth imperative baked into its existence. A magic seeing stone might sit unused for centuries. A company valued at $400 billion cannot afford to stop growing.

Palantir has spent two decades building the infrastructure to integrate surveillance data at a scale that was previously impossible. It has relationships with every major intelligence agency, law enforcement organizations at every level, and increasingly with the agencies that manage the basic infrastructure of American life. It employs brilliant engineers who have spent years optimizing these systems.

Whether that infrastructure gets used wisely is not a question Palantir can answer. That's a question for the rest of us—the ones being watched through the seeing stone.