PALENTIR: The AI Arms Dealer of the 21st Century
This is not intelligence. It is the automation of mass violence for shareholder value
Freddie Ponton via 21Wire
Palantir’s algorithm declared Iran a nuclear threat on June 12, 2025. Forty-eight hours later, the bombing began. The company that supplied the “evidence” for war had already sold targeting software to the Israeli military, and its executives had publicly called regime change an “investment opportunity.” When the strikes started, Palantir’s stock surged eleven percent in a week. Military analysts now dub the campaign “Operation Epic Fury,” describing it as the first AI-directed war in history. The same company selling data and battlefield technology to the IDF supplied the “evidence” that justified the US and Israel strikes on Iran.
This is not intelligence. It is the automation of mass violence for shareholder value, rendered in the sterile vocabulary of “platforms” and “solutions.”
VIDEO: Palantir co-founder Joe Lonsdale wants to invest in Iran after regime change, his company is pushing for (Source: Michael Tracey | CNBC)
Palantir co-founder Joe Lonsdale: “I’m really excited to invest in Iran,” as the Persian people are natural allies of the Jews, the USA, and the West. He claims he opposed the Iraq War, only insofar as he always thought Iran was the much better target pic.twitter.com/XnDw1uxocb
— Michael Tracey (@mtracey) June 21, 2025
THE MACHINE IN VIENNA
Palentir’s MOSAIC AI-powered nuclear verification system claims it consolidates vast datasets, including satellite imagery, trade logs, communications metadata, and social media, and turn it into predictive assessments about nuclear proliferation risks, generating pattern-based inferences about potential weapons programs.
MOSAIC arrived at the International Atomic Energy Agency in 2015, packaged as technical assistance for the six-party Joint Comprehensive Plan of Action (JCPOA). The IAEA funded MOSAIC through a combination of regular budget allocations and extra-budgetary contributions, with the United States serving as the agency’s single largest contributor. The €41 million system was finalised, with American taxpayers ultimately underwriting the machine that would later rationalise a war.
The platform descended from earlier Palantir systems already sold to the Israel Defence Forces for operations in Gaza and Lebanon, a lineage never disclosed as disqualifying when the IAEA adopted it for “neutral” verification. The system did not merely archive information; it generated relationships, mapping connections between individuals, facilities, imagery, and materials—but more importantly, this capability produced targeting lists.
FORTY-EIGHT HOURS
The timeline of June 2025 exposes the compression of democratic deliberation by algorithmic urgency. According to a Byline Times investigation, between June 6 and June 12, MOSAIC flagged an apparent surge in enriched uranium activity, assessing that Iran was potentially weeks away from producing multiple nuclear weapons. The IAEA Board of Governors passed a non-compliance resolution on June 12. Israel struck Iranian facilities on June 13. The United States followed immediately. Forty-eight hours elapsed between the algorithm’s alert and bombing—a velocity that eliminated opportunities for independent verification or diplomatic engagement.
This speed served specific interests. In March 2025, Director of National Intelligence Tulsi Gabbard testified that the consensus assessment of the US intelligence community remained unchanged: Iran had not pursued a nuclear weapon since 2003. According to The National Interest, no one in the US intelligence community was given the opportunity to probe the sourcing or verify the intelligence presented to the Trump administration by Israel. By June, Gabbard had reversed her position, claiming Iran could produce weapons “within weeks”, adopting the precise timeline the Palantir-powered assessment had generated. Congress received classified briefings, and Congressman Seth Magaziner soon after confirmed there was “no intelligence that Iran was planning to attack the United States”, revealing after the briefing that the administration had developed “no plan” for the war’s aftermath.
I just left a classified briefing about the Iran War. Here is what I can share. pic.twitter.com/zCoXcHUWVj
— Congressman Seth Magaziner (@Rep_Magaziner) March 4, 2026
Defenders will claim MOSAIC was merely advisory, that human policymakers retained final authority, and that intelligence failures predate algorithms. Unfortunately, this defense collapses under scrutiny. When the machine generates urgency measured in hours, human deliberation becomes a rubber stamp. Why else did the official US position flip in weeks after years of stable consensus? Why did strikes commence before inspectors regained physical access to challenge the machine’s inferences? The timeline does not suggest consultation. It suggests automation.
The technical reality crumbled rapidly. By March 2026, IAEA Director General Rafael Grossi claimed that inspectors had lacked access to critical Iranian sites for more than eight months, preventing them from determining whether nuclear material had been diverted. At the same time, he stated there is “no evidence” of an active nuclear weapons program. The “Minority Report for uranium” had produced a political cascade without independently verifiable proof. The facilities were bombed anyway.
Eight months after the first missiles fell, the IAEA Board of Governors convened in Vienna on March 2 to discuss the US and Israeli strikes against Iran. The emergency session, called at Russia’s request, arrives as hollow theatre—the nuclear watchdog meeting to deplore a war it helped justify with unverified algorithmic assessments.
Palantir’s commercial interests in the outcome of this meeting were more than explicit; they were obvious. In January 2024, the company had already signed a strategic partnership with Israel’s Defence Ministry, committing to support “war-related missions” with its software. According to reports, Palantir’s “Forward Deployed Engineers”, civilian coders who work alongside military clients in the field, have been involved in adapting targeting systems in active theatres. The same firm that provided the analytical backbone for the intelligence case now integrates classified satellite data and surveillance feeds into real-time targeting cycles.
The architects previewed their intentions in public. In a CNBC interview in June 2025 during the 12-day war, co-founder Joe Lonsdale urged pre-emptive strikes against Iranian nuclear infrastructure. As mentioned in the Byline Times investigation, Lonsdale expressed confidence in American bunker-buster munitions, saying he had been told “they work” and that “you can always hit things multiple times just to be sure.” Lonsdale described Iran as a country that “could be a prosperous republic if not run by crazy people,” and said he could not wait to “invest in Iran” after regime change. Peter Thiel, Palantir’s largest founding shareholder, had argued that every historical case of nuclear acquisition by an adversary produced regional war, framing an Iranian weapon as a “catastrophe” requiring preventive violence. Chief Executive Alex Karp told the New York Times in August 2024 that the United States would “very likely” face a three-front war against Russia, China, and Iran, and suggested that autonomous weapons systems of the kind Palantir builds would be decisive.
THE HARVEST
The market responded on schedule. When the strikes began, Palantir’s valuation, already inflated on the back of military and intelligence contracts, rose a further seven percent in the first week of the conflict. The previous month, Department of Justice disclosures had revealed that Thiel maintained a close business partnership with Jeffrey Epstein despite his conviction for the sexual abuse of minors, briefly depressing Palantir’s stock. War with Iran did not merely enrich Palantir’s shareholders—it helped wash the company’s image in the bloodless language of “deterrence.”
The contracts materialised on schedule. In August 2025—two months after the first strikes—the Army was already planning to consolidate 75 separate software agreements into a single $10 billion Enterprise Service Agreement, transforming Palantir into an “all-you-can-eat” subscription for the Pentagon. According to CNBC, the Department of Defense simultaneously expanded its Maven AI system contract by $795 million. While inspectors remained blocked from verifying the nuclear claims that justified the war, the UK Ministry of Defence awarded Palantir a £240.6 million direct contract in December 2025, bypassing competitive procurement using “defense and security exemption”, despite the company having hired four MoD officials that same year, including a former director of policy days after he left the ministry.
By February 2026, with the IAEA still admitting “no evidence” of an Iranian weapons program, the Department of Homeland Security signed a $1 billion blanket agreement for Palantir’s software across multiple agencies. The war that eliminated verification channels became the pretext for eliminating competitive bidding.
The commercial infrastructure of the conflict continues expanding. The Trump administration announced on March 3 that it will now offer US insurance for Gulf shipping and provide military escorts for tankers, generating parallel revenue streams beyond the direct Palantir contracts.
MOSAIC represents a privatization of the verification architecture underpinning the global non-proliferation regime. A proprietary, unauditable algorithm, developed by a firm with early CIA venture funding via In-Q-Tel, sold to the Israeli military, and marketed to an international watchdog, generated the technical basis for a war that killed hundreds of civilians. Congressional representatives were told there was no imminent threat to the United States, yet the strikes proceeded because the machine had spoken and its outputs were politically convenient.
This automation of casus belli removes accountability at every level. The code cannot be cross-examined in a courtroom. The underlying data cannot be fully disclosed without exposing sources and methods. When Grossi now admits his agency lacks physical access to verify whether nuclear material was diverted, he is describing a safeguards architecture hollowed out and replaced by predictive analytics that ultimately serve the commercial interests of their developers.
COLLATERAL
The most recent emblem of that arrangement came on 3 March 2026, when three missiles struck the Shajareh Tayyebeh primary school in Minab. Among the 168 dead was Hana Dehqani, eight years old. Her name does not appear in Palantir’s investor calls. The algorithm did not distinguish between centrifuge halls and classrooms, but it certainly provided the atmosphere of an emergency that made such distinctions politically irrelevant. Mosques became data points and children became collateral variables.
This account draws on the forensic investigation of Nafeez Ahmed, a systems theorist and two-time Project Censored Award winner whose work has been cited by the 9/11 Commission and UK parliamentary inquiries. His investigation, drawing on contract documents, corporate filings, and the public statements Palantir’s executives hoped nobody would connect, exposes a conflict of interest that makes the old military-industrial complex look almost restrained. Read the full investigation here.
__________
Header featured image (edited) credit: Org. post content. Emphasis added by (TLB)
••••
••••
Stay tuned…
••••
The Liberty Beacon Project is now expanding at a near exponential rate, and for this we are grateful and excited! But we must also be practical. For 7 years we have not asked for any donations, and have built this project with our own funds as we grew. We are now experiencing ever increasing growing pains due to the large number of websites and projects we represent. So we have just installed donation buttons on our websites and ask that you consider this when you visit them. Nothing is too small. We thank you for all your support and your considerations … (TLB)
••••
Comment Policy: As a privately owned web site, we reserve the right to remove comments that contain spam, advertising, vulgarity, threats of violence, racism, or personal/abusive attacks on other users. This also applies to trolling, the use of more than one alias, or just intentional mischief. Enforcement of this policy is at the discretion of this websites administrators. Repeat offenders may be blocked or permanently banned without prior warning.
••••
Disclaimer: TLB websites contain copyrighted material the use of which has not always been specifically authorized by the copyright owner. We are making such material available to our readers under the provisions of “fair use” in an effort to advance a better understanding of political, health, economic and social issues. The material on this site is distributed without profit to those who have expressed a prior interest in receiving it for research and educational purposes. If you wish to use copyrighted material for purposes other than “fair use” you must request permission from the copyright owner.
••••
Disclaimer: The information and opinions shared are for informational purposes only including, but not limited to, text, graphics, images and other material are not intended as medical advice or instruction. Nothing mentioned is intended to be a substitute for professional medical advice, diagnosis or treatment.
Leave a Reply