The name comes straight from Tolkien. In Quenya, palan means “far” and tír means “to watch,” a compound that translates to “far-seer.” In The Lord of the Rings, the palantíri are indestructible seeing-stones, made by the Elves, capable of showing distant places and communicating across vast distances, but their power depends entirely on the will of whoever looks into them. Used by the wise, they inform. Used by the weak or the corrupted, Sauron reaches through them and turns the watcher into the watched. That’s not a decorative detail. It’s the governing metaphor of the company built on the name: Palantir Technologies doesn’t need embellishment to matter. It has documented CIA seed funding, a genuinely disputed brush with the Cambridge Analytica scandal, a real and growing footprint across US and UK government infrastructure, and a founder who has, in fact, delivered a four-part private lecture series on the Antichrist. The real story isn’t that Palantir is secretly an all-seeing eye. It’s that a fictional seeing-stone became the name of a real company whose entire business is making disparate information visible, searchable and operational, and whose growing importance comes not from mythology but from how deeply governments have chosen to depend on that capability.
One detail worth fixing first, though: the earliest institutional backer was In-Q-Tel, the CIA’s venture capital arm, not the chipmaker Intel, an easy mix-up that’s worth clearing up before anything else.
The Actual Founding Story
Palantir was founded in 2003 by five people, not two: Peter Thiel and Alex Karp, who’d met at Stanford Law School, alongside PayPal engineer Nathan Gettings and Stanford students Joe Lonsdale and Stephen Cohen, who built the earliest prototype. Thiel initially bankrolled the company personally, and the founding pitch, adapting PayPal’s fraud-detection pattern-recognition into a tool for counterterrorism analysis “while preserving civil liberties,” struggled badly to attract outside investors. Even Sequoia Capital, an early PayPal backer, passed; chairman Michael Moritz was reportedly so unimpressed by the pitch that he doodled through the entire meeting. The breakthrough came through In-Q-Tel, the CIA’s venture capital arm, established in 1999 specifically to get emerging technology into the agency’s hands faster than traditional procurement allowed. In-Q-Tel’s CEO Gilman Louie personally sat in on Palantir’s pitch, an unusual step, because Thiel’s involvement got their attention, and because the CIA was already looking to replace an aging British-made analytics tool called Analyst’s Notebook. That funding, not a chip manufacturer, is the accurate origin of Palantir’s institutional backing.
What Actually Happened With Cambridge Analytica
This deserves precise handling, since the actual record is more specific, and more contested, than a one-line summary in either direction can capture. In March 2018, Palantir initially denied any relationship with Cambridge Analytica, then acknowledged the next day that a named London-based business development employee, Alfredas Chmieliauskas, had exchanged emails with Cambridge Analytica staff, including its CEO Alexander Nix, between 2013 and 2014. That contact went further than a single email exchange: Cambridge Analytica whistleblower Christopher Wylie told UK lawmakers under oath that he met with Chmieliauskas and other Palantir staff on multiple occasions, and that he personally supplied Palantir employees with documentation related to Cambridge Analytica’s work. Contemporary reporting separately identified documents concerning contact between Palantir and Cambridge Analytica’s own data scientists.
Palantir has consistently maintained that this was one employee acting in a personal capacity, that no formal contract or corporate relationship existed between the two companies, and that Palantir itself has never worked on election campaigns. Wylie’s sworn testimony describes a broader pattern of contact than Palantir’s account allows for, and Palantir has disputed the characterization. Neither side’s full account has been independently and conclusively settled in public reporting. The accurate formulation is therefore documented contact plus disputed scope, not a confirmed corporate scandal and not a fully cleared non-event.
Four Platforms, Not One, and a Client List That Keeps Growing
Palantir’s own SEC filings now describe four principal software platforms rather than the two most commonly discussed. Gotham remains the intelligence, defense and law enforcement pattern-analysis platform built first. Foundry handles commercial data integration, used by named clients including Airbus and firms across the pharmaceutical and logistics sectors. Apollo is the continuous-deployment infrastructure that lets Palantir’s software update itself across classified and unclassified environments simultaneously, a genuinely unusual technical capability the company has leaned on heavily in its defense pitches. And AIP, the Artificial Intelligence Platform launched in 2023, layers large-language-model tools on top of the other three, letting an operator query an organization’s own data using natural language. Together, these four platforms are why Palantir now describes itself less as an intelligence contractor and more as an operating system for institutions.
Palantir’s documented government client roster includes the CIA, FBI and NSA from its earliest years, expanding over two decades to include the bulk of the US military, the Department of Defense’s Maven programme for AI-assisted target identification, and, as of a 2024 agreement, Israel’s Ministry of Defense. Palantir’s 2020 debut on public markets happened through a direct listing rather than a traditional IPO, a deliberate structural choice that let Thiel, Karp and other insiders retain outsized voting control while still accessing public capital.
ImmigrationOS Is the Company’s Most Consequential US Contract
The client list keeps growing, and its most significant recent US addition deserves its own real accounting. In April 2025, ICE awarded Palantir a $30 million contract, structured as a modification to an existing sole-source deal, to build the Immigration Lifecycle Operating System, or ImmigrationOS, with a prototype due by September 25, 2025, and the contract itself running through September 2027. The system augments Palantir’s existing Investigative Case Management platform, itself already built specifically for ICE’s operational needs since 2014, and the combined ICM deal has since grown past $145 million. ImmigrationOS is intended to pull together data across agencies to prioritize individuals for deportation, including tracking visa overstays and self-deportations in what contract documents describe as near real-time.
A documented conflict-of-interest detail sits alongside the contract: Stephen Miller, the Trump administration’s chief architect of immigration policy, holds a substantial financial stake in Palantir, a fact confirmed in reporting on the ImmigrationOS deal specifically. Separately, in December 2025, USCIS, a different Department of Homeland Security agency that had not previously contracted with Palantir directly, began a small initial contract for a platform described in filings as “vetting of wedding-based schemes,” marking the start of a new, expanding relationship distinct from Palantir’s decade-plus ICE partnership.
The UK Relationship Is Larger, Older and More Contested Than One Number Suggests
This is where the previous version of this piece needs a real correction, not a minor one. The widely repeated figure of a “$750 million UK defense contract” conflates two genuinely different things. In September 2025, during a state visit by President Trump, the UK Ministry of Defence announced a strategic partnership with Palantir framed as identifying opportunities worth up to £750 million over five years, alongside a headline £1.5 billion total investment figure. MoD sources were explicit at the time that this was not a signed contract for specific products or services; it was a framework inside which future enterprise agreements would still need to be separately negotiated. That framework replaced an existing, actual £75 million, three-year enterprise agreement that was nearing its end.
The real, signed contract that followed arrived in December 2025: a £240.6 million, three-year enterprise agreement covering continued data-analytics licensing and support “across classifications” and interoperable with NATO and allied systems, awarded without a competitive procurement process, using a defense and security exemption. Separately, and unrelated to defense, Palantir’s relationship with Britain’s National Health Service is both older and more contentious than the MoD story. It began in March 2020, when Palantir was contracted for a fee of just £1 to help build the NHS COVID-19 Data Store; that contract was repeatedly extended through January 2023 while NHS England worked out a permanent procurement. In November 2023, a Palantir-led consortium won the resulting Federated Data Platform contract through open competitive tender, an initial published value of £182.2 million with a maximum envelope of up to £330 million over seven years.
That contract has not been quiet since. In 2026, the UK’s National Infrastructure and Service Transformation Authority revised the platform’s whole-life cost upward to £1.1 billion while revising projected benefits downward to £808 million. A cross-party committee of MPs has formally urged the government to walk away from the deal. The British Medical Association has instructed doctors to limit their engagement with the platform. A coalition including Privacy International, Amnesty International, the Good Law Project and the health-justice charity Medact sent hospital trusts and integrated care boards a briefing warning that the platform’s “highly interoperable” design could enable “data-driven state abuses of power,” explicitly citing the risk of “US-style ICE raids,” a direct, documented link back to Palantir’s own American immigration-enforcement work. A break clause in the NHS contract becomes available from February 2027, and the UK government has said a decision on whether to trigger it is expected before then. Separately, London’s Metropolitan Police Service had a planned £50 million Palantir contract blocked by the Mayor’s Office for Policing and Crime in 2026, and Britain’s National Energy System Operator has moved toward a roughly £21 million no-bid Palantir deal of its own. The pattern across all of these is consistent: real, growing, individually negotiated contracts, not one number that captures the whole relationship.
Thiel’s Theology Is No Longer Something We Have to Infer
Thiel’s blending of theological language with arguments about technology and civilizational necessity used to be something reporters had to reconstruct from scattered speeches and essays. That changed in the fall of 2025. Between September and October, Thiel delivered a private, off-the-record, sold-out four-part lecture series titled “The Antichrist: A Four-Part Lecture Series” at the Commonwealth Club in San Francisco, organized by the ACTS 17 Collective, a nonprofit dedicated to what it calls acknowledging Christ in technology and society. The Washington Post’s Gerrit De Vynck, Nitasha Tiku and Elizabeth Dwoskin obtained and reviewed leaked audio recordings of all four sessions; UC Berkeley digital forensics researcher Hany Farid separately verified the authenticity of the recordings.
In the lectures, Thiel described Greta Thunberg and critics of unregulated AI development as “legionnaires of the Antichrist,” arguing that regulatory limits on technology don’t just hinder business but risk ushering in the destruction of the United States and a period of global totalitarian rule. He drew explicitly on the work of French philosopher René Girard, who taught Thiel at Stanford as an undergraduate and profoundly shaped his intellectual formation, alongside Francis Bacon and the political theorist Carl Schmitt. Thiel declined to comment when the Post sent him a detailed list of questions about the lectures’ contents. This is not an invented embellishment or a loose inference from a handful of quotes. It’s a documented, verified, four-part theological argument, delivered by the co-founder of a company whose software now sits inside intelligence agencies, militaries and healthcare systems across two governments, connecting AI regulation directly to apocalyptic stakes.
Karp, for his part, holds a PhD in social theory from Frankfurt, studied under Jürgen Habermas, and has described data analysis in explicitly ethical terms throughout his public writing, including his 2025 book The Technological Republic. Both men’s stated worldviews are documented, on the record, and stranger than most Silicon Valley executives’, which needs no exaggeration to remain genuinely notable.
What the Accurate Version Actually Shows
None of the genuine facts here need embellishment to describe a company worth serious scrutiny. An actual CIA-funded startup grew, over two decades, into central infrastructure for intelligence agencies, militaries, and increasingly civilian government functions across multiple countries, run by two men whose public philosophy is unusually explicit about technology, security and control, one of whom has now delivered a verified, four-part private argument connecting AI regulation to the apocalypse. The Cambridge Analytica episode remains a genuinely disputed, only partially resolved public record, not a settled scandal and not a cleared non-event. The UK relationship isn’t one $750 million contract; it’s a growing, contested, multi-agency footprint spanning defense, healthcare and policing, with its own parliamentary opposition and its own break clause on the calendar. Getting the specific facts right, In-Q-Tel rather than Intel, a documented pattern of contact rather than a single email, £240.6 million rather than £750 million, doesn’t make Palantir’s actual footprint smaller. It makes that footprint, like the seeing-stones its name comes from, considerably harder to look away from.
