Please follow us on Truth Social, X, YouTube, Minds, Telegram, Rumble, GETTR, Gab, Instagram
Palantir Technologies delivered numbers Wall Street could not ignore this week — second-quarter revenue of $1.94 billion, a 93% increase year-over-year, crushing analyst estimates of $1.8 billion. U.S. commercial revenue surged 149% to $764 million. Government revenue grew 90% to $809 million. Net income hit $1.07 billion, and the stock rocketed 30% in a single session. By any financial measure, this was an extraordinary quarter for a company that has spent two decades building what it calls the infrastructure of sovereign intelligence.
CEO Alex Karp called the results "otherworldly." The company raised its full-year 2026 revenue guidance to approximately $8.15 billion — an 82% projected annual growth rate — and lifted U.S. commercial guidance to 134% year-over-year growth. "To my knowledge," Karp told CNBC, "no business at our scale has even grown half this much." The numbers are real, and the demand driving them is real: enterprises and government agencies want AI tools that keep their data on-premises, away from the large frontier AI labs.
The financial story is straightforward. The civil liberties story is not.
What Palantir's Platforms Actually Do To Ordinary Americans
Palantir's flagship government product, Gotham, is marketed as an intelligence-analysis tool. In practice, it functions as a data-fusion engine — pulling together DMV records, criminal databases, social media activity, financial transaction data, location history, and family relationship networks into detailed profiles on individuals. The people being profiled are often not suspects. They have not been charged with any crime. They may not even be under investigation. They simply appear in a dataset that a law enforcement or immigration agency has licensed access to, and Palantir's platform makes that data operationally searchable at scale.
The LAPD's experience with Palantir's predictive policing tools is instructive. Beginning in 2011, the department ran Operation LASER — Los Angeles Strategic Extraction and Restoration — using Palantir software to generate "chronic offender scores" on residents based on criminal history, social media data, and license plate scans. Officers were directed to knock on the doors of high-scoring individuals, inform them they were being monitored, and look for opportunities to stop or arrest them. The program was discontinued in 2019 under sustained pressure from activist groups and the department's own civilian oversight panel. One LAPD captain, quoted in research reviewed by The Intercept, described person-based predictive policing as "a civil liberties nightmare." The ACLU concluded the software often predicted patterns of policing rather than actual crime — in other words, the algorithm encoded and amplified existing biases rather than reducing them.
The consent problem is fundamental. Ordinary Americans whose data flows into Palantir systems have no idea it is happening, have no mechanism to opt out, and have no practical legal recourse once the data is inside a government-contracted platform. A March 2025 executive order directing federal data sharing across agencies is reportedly being implemented through Palantir's Foundry platform. IRS records, Social Security data, Department of Labor records, and other sensitive government datasets have reportedly been integrated — or considered for integration — without meaningful public notice or congressional authorization. Several current and former Palantir employees have told journalists they were unnerved by the scope of the work.
U.S. Immigration and Customs Enforcement has used Palantir's tools extensively — a fact documented through Freedom of Information Act litigation. ICE's Investigative Case Management system, built on Palantir's platform, was specifically configured for ICE's operational needs according to contract language obtained by civil liberties researchers. More recently, ICE deployed a Palantir-built application called ELITE that, according to reporting by 404 Media, populates a map with potential deportation targets, generates dossiers on each individual, and assigns a "confidence score" on the person's current address. Palantir has also won contracts for a system called ImmigrationOS, designed to track the movements and relationships of immigrants across federal datasets. The American Civil Liberties Union called the program part of an "abusive removal campaign" that weaponizes data aggregation against people who, in many cases, have committed no criminal offense, except of course possibly entering the country illegally. That actually is a crime.
The Founding Ethos Versus The Business Reality
Peter Thiel co-founded Palantir in 2003, in part, as a libertarian answer to government surveillance excess. The company's original pitch was that it could give intelligence agencies better targeting tools precisely so they could stop mass data collection — surgical capability instead of a dragnet. Alex Karp has long presented Palantir as a defender of Western liberal values against authoritarian adversaries. That framing has worn thin as the company's actual business model has matured.
What Palantir has built, in practice, is state surveillance infrastructure. It profits from agencies that hold broad legal authorities to collect data on U.S. persons, and its platforms are specifically engineered to make that data operationally useful at scale. The "sovereignty" Karp celebrates in earnings calls is institutional sovereignty — the government's ability to act on its data without Big Tech intermediaries. It is not the sovereignty of the individual citizen who has no knowledge that a profile exists on them, no ability to challenge it, and no recourse if it leads to a knock on their door.
The Electronic Frontier Foundation and the ACLU have both identified Palantir as one of the most consequential surveillance infrastructure companies operating in America today — not because it is uniquely malicious, but because it is uniquely capable. The same platform that helps the Army identify targets in a war zone can help a domestic law enforcement agency build a file on a protester, a journalist, or a political organizer. The technical capability does not distinguish.
The Defense For Palantir — And Its Limits
Defenders of the company — and there are many in national security circles — argue that Palantir is a necessary tool in a dangerous world. The U.S. Army's $10 billion enterprise contract, the TITAN AI targeting system, and deployments across U.S. Cyber Command and the Intelligence Community represent genuine capability that adversaries like China and Russia are building regardless. If the United States is going to use AI for defense and intelligence, the argument goes, it is better done by an American company with U.S. government oversight than outsourced to labs with murky foreign ties or weak national loyalty.
That argument carries real weight. The People's Liberation Army is not waiting for Congress to resolve American privacy debates. A sovereign nation needs intelligence capability, and Palantir's work with the defense and intelligence community is a legitimate part of that picture.
But the defense argument does not extend without limit to domestic law enforcement, immigration enforcement, and the warrantless aggregation of data on American citizens who have committed no crime. Those are different legal and constitutional contexts, and conflating them — as Palantir's corporate communications consistently do — is a rhetorical move, not an argument. The Army's TITAN targeting system operates in a declared conflict zone.
The Question That No Earnings Call Answers
Palantir's stock is up 30%. The $1.94 billion in quarterly revenue is real. The 93% growth rate is real. The company is, by any conventional metric, one of the most successful enterprise software businesses in American history. None of that is in dispute.
What is also real: a company this deeply embedded in government surveillance infrastructure, growing this fast, with this much data on ordinary Americans flowing through its systems, deserves something other than a quarterly celebration. America-First conservatism has always understood that power concentrated — whether in corporate monopolies or government agencies — is power available for abuse. The question is not whether Palantir's technology works. It clearly does. The question is who watches the watchers: who decides which data gets fused into which profile, which confidence score triggers which knock on which door, and what happens when that power is turned on the wrong person. A country built on the Fourth Amendment should be asking those questions, and asking them loudly, regardless of what the stock chart looks like.
