The Liberty Lookout: Weekly Deep Dives

The Ghost Files: Inside the FBI's System for Making Secrets Disappear

The Liberty Lookout — Sat, 07 Mar 2026 15:33:17 GMT

The FBI built a digital system that makes its own records invisible. Not classified. Not redacted. Invisible. Search for them and the system tells you they don’t exist.

This week, the scope of that system finally started to come into focus. And it’s worse than the headline suggests.

How to Make a File Disappear

The FBI’s case management system, SENTINEL, is supposed to be the central nervous system for every investigation the Bureau runs. Congressional oversight requests, Freedom of Information Act inquiries, criminal defense attorneys seeking exculpatory evidence, internal agents researching related cases: it all runs through SENTINEL.

The system has a normal classification tier called “Restricted Access,” where you can see a file exists but need clearance to open it. Standard stuff.

Then there’s a tier nobody outside FBI leadership knew about until last year: “Prohibited Access.”

A declassified 2019 FBI memo explains how it works: “When search terms that exist in the Prohibited Access-status cases are searched in Sentinel, the particular search will receive a false-negative Sentinel search response.”

Not “access denied.” The system actively lies. The files are ghosts.

Only a small circle around the FBI Director and Deputy Director controls which files become ghosts. Everyone else, including other FBI agents, gets told nothing is there.

The Scope

Matt Taibbi’s Racket News exclusive yesterday reported the cache may contain as many as a thousand distinct case numbers dating back to at least 1999, spanning multiple presidencies. Senate Judiciary Chairman Chuck Grassley confirmed that Attorney General Bondi and FBI Director Patel have begun turning material over to Congress.

But the political headlines barely scratch the surface of what this system means.

Innocent People May Be in Prison

Here’s the angle that should terrify everyone, regardless of political tribe.

Under the Brady rule, prosecutors are constitutionally required to turn over exculpatory evidence to criminal defendants. When defense attorneys request that evidence, FBI agents search SENTINEL for relevant material.

If relevant evidence sits in a Prohibited Access file, the agents don’t even know it exists. As legal analyst Margot Cleveland put it: “Agents pulling Brady material to comply WITH THE CONSTITUTION will not even know there are potentially exculpatory materials they should be providing.”

That means for potentially 25 years, defendants in federal criminal cases may have been convicted without access to evidence that could have cleared them. Not because a prosecutor chose to hide it, but because the filing system itself was rigged to say the evidence didn’t exist.

The same logic applies to every FOIA request filed since the system was created. Every congressional oversight demand. Every Inspector General investigation. The FBI’s answer to all of them was effectively: “We searched and found nothing.” That may have been technically true from the agent’s perspective. But the system was lying to the agents, too.

The Pattern: From Hoover’s Cabinet to Digital Ghosts

The FBI has always had secret files. What changes is the technology.

J. Edgar Hoover kept “Official and Confidential” files for decades: a personal blackmail vault containing dirt on politicians and journalists. Ronald Kessler, who reviewed what survived, concluded they “could have been gathered for no other purpose than blackmail.”

COINTELPRO (1956-1971) targeted civil rights leaders, antiwar activists, and political dissidents with forged documents, infiltration, and IRS audits.

The Church Committee was supposed to end all this in 1975. It didn’t. It just pushed the secrecy deeper into bureaucratic infrastructure.

After 9/11, the transformation accelerated. The FBI shifted from a law enforcement agency that catches criminals to a domestic intelligence service that collects information for its own sake. Whistleblowers like Garrett O’Boyle, Steve Friend, and others describe being pulled off real enforcement work and reassigned to constitutionally questionable surveillance.

Prohibited Access is the natural endpoint of that evolution. When your primary mission is collecting secrets, you eventually need a system for keeping secrets about your secrets.

What’s Already Leaked Out

Even before the full cache is turned over, the files that have surfaced show the system’s reach.

The FBI’s Arctic Frost investigation (the precursor to Jack Smith’s prosecution of Trump) used 197 subpoenas classified as Prohibited Access. Among the targets: phone records of eight sitting Republican senators, obtained in 2023. That information was found in a Prohibited Access file only because Grassley’s office specifically demanded it. Standard SENTINEL searches would have returned nothing.

Patel himself claims the FBI buried subpoenas of his personal phone records in Prohibited Access files “using flimsy pretexts” while he was a private citizen.

And the Mueller investigation’s SA Walter Giardina allegedly wiped his FBI-issued laptop outside standard protocol, while team member Andrew Weissmann wiped his government phone multiple times. When the evidence itself is in ghost files and the devices that held it are conveniently erased, you’re looking at a system designed from the ground up to be unaccountable.

Why Reform Won’t Fix This

The comfortable conclusion is that this is a fixable scandal. Release the files, fire the bad actors, pass some reforms.

History suggests otherwise.

The Church Committee “reformed” intelligence oversight in 1975. The FBI kept running domestic surveillance programs. The 9/11 Commission mandated “information sharing.” The FBI responded by building a system that shares information with nobody. Every previous reform left the core architecture of secrecy intact while the agencies built newer, better vaults around it.

The deeper problem is structural. Any institution with the power to classify its own activities and the incentive to avoid accountability will eventually build something like Prohibited Access. It doesn’t require a conspiracy. It requires a bureaucracy staffed by people whose careers depend on secrecy, given digital tools that make secrecy frictionless.

“If it weren’t for whistleblower disclosures to my office, the very existence of the FBI using ‘Prohibited Access’ files for some investigations would have remained in the dark,” Grassley said. That’s not a reassuring statement. That’s an admission that the entire oversight apparatus of the U.S. government failed, and the only thing that worked was individuals willing to risk their careers.

The question isn’t whether these particular files get released. It’s whether the system that produced them gets dismantled, or just renamed.

Sources:

- Grassley: Prohibited Access Investigation (Senate Judiciary Committee)

- Declassified FBI Analysis re: Nellie Ohr (Senate.gov)

- FBI’s Secret Stash Finally Uncovered (Racket News, March 6, 2026)

- Arctic Frost: FBI Spied on Eight Republican Senators (Senate Judiciary Committee, October 2025)

- FBI Prohibited Access: Brady Implications (Eagle Observer, June 2025)

- Herridge: Prohibited Access Files (Catherine Herridge Reports, June 2025)

- FBI Whistleblower Testimony (PBS, May 2023)

- Hoover’s Secret Files (The Daily Beast/Ronald Kessler)

- COINTELPRO (Wikipedia)

- Church Committee (Senate.gov)Loading...

Comply or Die: The Pentagon Blacklisted an American AI Company for Having Ethics

The Liberty Lookout — Fri, 06 Mar 2026 14:14:26 GMT

The Pentagon told an American company to drop its ethical guardrails. The company said no. So the government treated it like a hostile foreign power.

Last Friday, Defense Secretary Pete Hegseth designated Anthropic a “Supply-Chain Risk to National Security,” a classification historically reserved for foreign adversaries like Huawei and Kaspersky. Anthropic’s crime: refusing to let the military use its AI for mass domestic surveillance and fully autonomous weapons without any restrictions.

Hours later, OpenAI swooped in with a Pentagon deal for classified networks, claiming its agreement includes the same safeguards Anthropic got blacklisted for requesting. According to several officials, this is a lie and the Pentagon actually has free-reign to do whatever they want now with OpenAI’s systems.

The Shakedown

The setup started months ago. Anthropic signed a $200 million contract with the Pentagon in July 2025, becoming the first frontier AI company to deploy models on classified military networks. The company asked for two guardrails: no mass domestic surveillance of Americans, and no fully autonomous weapons. The Pentagon wanted “all lawful purposes” with zero exceptions.

On February 24, Hegseth issued an ultimatum: agree to the Pentagon’s terms by 5:01 PM Friday, February 27, or face consequences. The government also threatened to invoke the Korean War-era Defense Production Act, a 1950 law that lets the military commandeer private company facilities.

CEO Dario Amodei didn’t blink: “We will not knowingly provide a product that puts America’s warfighters and civilians at risk.” He offered to work with the Pentagon on R&D to improve autonomous system reliability. The Pentagon wasn’t interested.

At 5:01, the deadline passed. Trump took to Truth Social: “The Leftwing nut jobs at Anthropic have made a DISASTROUS MISTAKE trying to STRONG-ARM the Department of War.” He ordered every federal agency to immediately cease using Anthropic’s technology, with a six-month phase-out.

Hegseth’s follow-up was even more aggressive: “Effective immediately, no contractor, supplier, or partner that does business with the United States military may conduct any commercial activity with Anthropic.”

That’s not canceling a vendor contract. That’s economic warfare against a $380 billion American company with $14 billion in annualized revenue that just raised $30 billion and was planning an IPO.

Follow the Money, Follow the Musk

Four days before Anthropic got blacklisted, Elon Musk’s xAI signed its own deal to put Grok on classified Pentagon systems. Musk, Trump’s biggest financial backer in 2024, has been publicly attacking Anthropic, claiming the company “hates Western civilization.”

Small problem: government insiders told the Wall Street Journal they consider Grok too sycophantic, too susceptible to data poisoning, and less capable than Claude. The GSA, the agency that handles federal procurement, flagged security concerns. Until Anthropic refused the Pentagon’s ultimatum, military officials preferred Claude.

So the government sanctioned the AI it trusted, to make room for the AI it doesn’t, owned by the president’s biggest donor. As the CSIS’s Greg Allen put it: “The Defense Department just sent a huge message to every company that if you dip your toe in the defense contracting waters, we will grab your ankle and pull you all the way in, anytime we want.”

The Legal House of Cards

The supply chain risk designation under 10 USC 3252 is designed to protect against adversaries who might “sabotage, maliciously introduce unwanted function, or otherwise subvert” military systems. Anthropic isn’t sabotaging anything. It’s setting terms on a product it built. Multiple legal experts say the designation is on shaky ground.

The statute requires a risk assessment and Congressional notification before the designation takes effect. It’s unclear whether either happened. Hegseth’s claim that every military contractor must cut all commercial ties with Anthropic goes far beyond what the statute allows, according to the company and several federal contracting lawyers.

Dean Ball, a senior fellow at the Foundation for American Innovation and a former Trump administration AI policy adviser, didn’t mince words. He called it “the most shocking, damaging, and overreaching thing I have ever seen the United States government do” and described Hegseth’s interpretation as “almost surely illegal,” “attempted corporate murder,” and “obviously a psychotic power grab.”

That’s not a progressive critic. That’s a former Trump administration AI policy adviser.

The Dominoes Are Still Falling

Anthropic says it will challenge the designation in court. The legal fight will test whether the government can weaponize national security classifications against domestic companies that refuse to surrender their terms of service. If Amazon, Google, and Nvidia, all Anthropic investors who also do business with the Pentagon, actually have to divest, the chilling effect on every tech company in America would be immediate and permanent.

The precedent here isn’t about AI. It’s about whether the government can economically destroy any company that negotiates instead of capitulates. If a $380 billion American firm with $14 billion in annualized revenue can get treated like Huawei for asking that its technology not be used for domestic mass surveillance, what chance does anyone else have?

Sources:

- CNBC: Trump admin blacklists Anthropic

- NPR: OpenAI announces Pentagon deal after Trump bans Anthropic

- WIRED: Anthropic Hits Back After ‘Supply Chain Risk’ Label

- Anthropic: Statement on comments from Secretary of War

- OpenAI: Our agreement with the Department of War

- Fortune: OpenAI sweeps in after Anthropic labeled ‘supply chain risk’

- Forbes: Hegseth designates Anthropic as supply chain risk

- Futurism: Government insiders concerned by Grok deployment

Roundup Is Now a Matter of National Security (No, Really)

The Liberty Lookout — Wed, 04 Mar 2026 16:25:36 GMT

Trump invokes wartime powers to protect Bayer’s herbicide empire. The MAHA crowd got played.

On Tuesday, Bayer announced a $7.25 billion settlement to resolve roughly 200,000 cancer lawsuits over its flagship weedkiller, Roundup. On Wednesday, President Trump signed an executive order invoking the Defense Production Act to declare glyphosate (Roundup’s active ingredient) and elemental phosphorus “critical to the national defense.”

One day apart. What a coincidence.

What the Order Actually Does

The DPA order delegates sweeping authority to Agriculture Secretary Brooke Rollins (in consultation with Defense Secretary Pete Hegseth) to prioritize contracts, allocate materials, and issue orders related to phosphorus and glyphosate production. It explicitly instructs Rollins to ensure that no regulation “places the corporate viability of any domestic producer of elemental phosphorus or glyphosate-based herbicides at risk.”

Translation: the USDA is now legally required to protect Bayer’s bottom line.

The kicker is Section 707 immunity. The order grants DPA compliance immunity to domestic producers, meaning: “No person shall be held liable for damages or penalties for any act or failure to act resulting directly or indirectly from compliance” with government orders. Bayer’s cancer lawsuits could get a lot more complicated for plaintiffs.

There’s exactly one domestic producer of both elemental phosphorus and glyphosate in the United States: Bayer’s Monsanto subsidiary, operating out of a Superfund site in Soda Springs, Idaho. So when the White House says this order protects “domestic producers,” plural, it means one company. One German-owned company, to be precise.

The Phosphorus Fig Leaf

The phosphorus angle isn’t entirely manufactured. Elemental phosphorus is genuinely important for defense supply chains: smoke and illumination munitions, semiconductor manufacturing, lithium-ion batteries for weapons systems. The Department of Interior designated phosphate a critical mineral in November 2025. The US imports over 6 million kilograms annually because domestic production can’t keep up. China produces roughly 70% of the world’s glyphosate supply, so the supply chain vulnerability is real.

But here’s the trick. Phosphorus is the Trojan horse. You bundle a legitimate defense concern (phosphorus for munitions and chips) with a massively controversial herbicide (glyphosate for Bayer’s shareholders), and suddenly anyone who questions the glyphosate protection is “against national security.” Neat.

The Three-Front Bayer Bailout

This EO doesn’t exist in a vacuum. It’s one prong of a three-front legal strategy that would make any K Street lobbyist weep with pride:

Front 1: The Supreme Court. Bayer’s case Monsanto v. Durnell heads to oral arguments in April 2026. Bayer argues that EPA approval of Roundup’s label (no cancer warning required) preempts all state failure-to-warn lawsuits. The Trump administration’s Solicitor General filed a brief supporting Bayer, reversing the Biden-era position.

Front 2: State legislatures. Bayer has been lobbying state lawmakers to pass laws shielding pesticide makers from failure-to-warn lawsuits when their products follow federal labeling. North Dakota and Georgia already obliged.

Front 3: This executive order. DPA immunity, corporate viability protection, and the magic words “national security” attached to a product the WHO’s cancer research agency classified as “probably carcinogenic” in 2015.

The $7.25 billion settlement, announced one day before the executive order, suddenly looks less like accountability and more like the final piece of a carefully choreographed exit strategy. Pay a fraction of what juries have been awarding (some plaintiffs would get as little as $10,000), then slam the legal doors shut behind you.

The MAHA Betrayal

Remember Make America Healthy Again? The movement that rode to power on promises to confront pesticides, clean up the food supply, and hold agrochemical companies accountable?

RFK Jr. built his national profile partly by helping win a $289 million verdict against Monsanto for a man who developed cancer from Roundup. His MAHA movement frequently targeted glyphosate as a symbol of everything wrong with industrial agriculture.

So when Trump declared that same chemical a matter of national security, Kennedy’s response was... to issue a statement through a spokesman supporting the president. “Donald Trump’s executive order puts America first where it matters most,” Kennedy said, pivoting from cancer crusader to company man in a single press release.

His supporters noticed. “MAHA voters were promised health reform, not chemical entrenchment,” said Vani Hari, a Kennedy nutrition ally, calling the order “a direct assault on MAHA.” The Environmental Working Group’s Ken Cook was more blunt: “I can’t envision a bigger middle finger to every MAHA mom than this. President Trump just gave Bayer a license to poison people. Full stop.”

Cook’s harshest line? “MAHA supporters were promised reform, and instead, they’ve been treated by MAGA like a convenient group of useful idiots ever since Kennedy joined Trump on the campaign trail.”

Hard to argue with that assessment.

The Ratchet That Never Clicks Back

The Defense Production Act was passed in 1950 to mobilize American industry during the Korean War. Price controls, material allocations, the works. The Army War College has noted that the DPA “is much broader than its original purpose” and that scope creep has actually “weakened its ability to prepare” for actual military needs.

Trump alone has invoked it for mineral production, maritime industry, spent nuclear fuel, coal power plants, and now weedkiller. Each use stretches “national security” a little further. Each use sets the precedent for the next one. And emergency executive powers, once expanded, never contract. Name the last time a president voluntarily surrendered a power his predecessor claimed… take your time.

The same revolving door between government and agrochemical corporations that produced this order also produced the EPA’s finding that glyphosate doesn’t need a cancer warning, the Solicitor General’s brief backing Bayer at the Supreme Court, and now a Defense Production Act invocation protecting a single company’s “corporate viability.” The people writing these policies and the people profiting from them are, quite often, the same people. Or at least they attend the same dinner parties.

If the government can declare Roundup a matter of national security, what can’t it declare essential? Every expansion of “national defense” is another brick in the wall between you and the freedom to opt out.

The DPA was written to fight wars. Now it fights lawsuits. That should tell you everything about whose interests the government actually serves.

Sources

- Executive Order text (White House, Feb 18, 2026)

- White House Fact Sheet (Feb 18, 2026)

- Bayer $7.25B Settlement (AP News, Feb 17, 2026)

- CNBC: Trump order pushes glyphosate production (Feb 18, 2026)

- EWG Response (Feb 18, 2026)

- Farm Policy News (Feb 19, 2026)

- IARC Glyphosate Classification (March 2015)

- 50 U.S.C. § 4557: DPA Section 707 Immunity

- DPA History (Army War College)

- China Glyphosate Dominance (University at Buffalo)

- Bayer Supreme Court Case (Monsanto v. Durnell)

- Bayer: Glyphosate shortages not expected (Reuters, Feb 19, 2026)

The 250-Year Clock: When Empires Die and What Comes Next

The Liberty Lookout — Mon, 02 Mar 2026 19:56:01 GMT

In 1976, a British military officer named Sir John Glubb published a short paper called “The Fate of Empires.” He had spent his career studying the rise and fall of civilizations, and he found something unsettling: empires die on a remarkably consistent schedule.

Glubb examined 11 empires spanning 3,000 years of history: Assyria, Persia, Greece, Rome, the Arab Empire, the Mameluke Empire, the Ottoman Empire, Spain, Romanov Russia, Britain, and others. Despite vast differences in technology, geography, religion, and culture, most clustered around the same lifespan.

250 years. Ten generations. Then collapse.

The Assyrian Empire lasted 247 years. The Ottoman Empire: 250. The Spanish Empire: 250. The British Empire: 250 (1700-1950). Romanov Russia: 234. The Arab Empire: 246. The pattern was remarkably consistent across cultures separated by thousands of years and thousands of miles.

The United States declared independence in 1776. That puts the 250-year mark at 2026.

But before you tattoo that number on your forearm, the model deserves some honest scrutiny.

The Six Stages

Glubb identified six stages every empire passes through, in the same order, without exception.

The Age of Pioneers (outburst): An energetic, often poor society expands aggressively. Risk-taking is rewarded. Courage matters more than credentials.

The Age of Conquest: The pioneers’ energy translates into territorial and economic expansion. Military dominance is established.

The Age of Commerce: Conquest gives way to trade. The nation becomes wealthy. Merchants replace warriors as the ruling class.

The Age of Affluence: Wealth becomes the primary goal. The gap between rich and poor widens. The middle class begins to hollow out.

The Age of Intellect: Education expands dramatically. Universities multiply. Intellectualism replaces practical knowledge. Debates become increasingly disconnected from material reality.

The Age of Decadence: The final stage. Glubb’s markers are specific: defensiveness and pessimism replace confidence. Materialism dominates culture. Celebrity replaces achievement. The welfare state expands as the government attempts to buy the loyalty of a discontented population. Frivolity, political polarization, and an obsession with internal divisions characterize public life. The military, once a source of pride, becomes something the citizenry avoids.

Glubb was not a pessimist. He saw this cycle as natural, not tragic. Empires, like organisms, have lifespans. Understanding the pattern doesn’t prevent it, but it does allow you to prepare for what comes next.

Where the Model Breaks

The 250-year pattern is striking, but Glubb got there partly by choosing convenient start and end dates. The biggest problem is Rome.

Glubb split Rome into two separate entries: the Roman Republic (260-27 BC, 233 years) and the Roman Empire (27 BC-AD 180, 207 years). Two tidy numbers that both fit the pattern. But look at what he did to get there.

The Roman Republic is traditionally dated from 509 BC, not 260 BC. Glubb lopped off the first 249 years to make it fit. And the Roman Empire didn’t end with Marcus Aurelius in 180 AD. The Western Empire limped on for another 296 years until 476 AD. Glubb cut it off early because the late Empire was messy and long, and messy and long doesn’t fit a theory about clean 250-year cycles.

Worse, his six-stage lifecycle doesn’t survive the split. If the Empire “restarted the clock” in 27 BC, that means Augustan Rome was in the Age of Pioneers: a scrappy, energetic young society expanding aggressively. But Rome in 27 BC was the wealthiest, most sophisticated civilization on Earth, with a professional military, a complex legal system, a Mediterranean trade network, and a population that already demanded bread and circuses. Augustus didn’t pioneer anything. He consolidated power over a late-stage civilization whose republican system had collapsed under its own weight. That’s not a rebirth. That’s a symptom of decay.

Count Rome as one continuous civilization from founding to fall, and you get 985 years. Almost four times the pattern.

None of this means Glubb was wrong about everything. The empires that fit cleanly (Assyria, Ottoman, Spain, Britain, Russia) really do cluster around 250 years, and that’s too consistent to ignore. But the 250-year number is a pattern, not a law of physics. The real question isn’t whether empires die on a schedule. It’s whether the mechanisms Glubb identified (and other scholars refined) are operating in the United States right now.

On that question, the evidence is harder to argue with.

The Predictions Nobody Wanted to Hear

Glubb is far from the only scholar who saw this coming. A convergence of independent researchers, using completely different methodologies, arrived at the same approximate timeline.

Johan Galtung, the Norwegian sociologist who correctly predicted the collapse of the Soviet Union within his 10-year forecast window (the Berlin Wall fell in 1989, right on schedule), applied the same structural-contradiction framework to the United States in the year 2000. His prediction: the US empire would collapse by 2025. Galtung was precise about what he meant: not the end of the United States as a country, but the end of its ability to project unilateral power globally. He identified 15 internal contradictions within the American system and warned that as global power retracted, the instability would reflect inward, potentially threatening the nation’s territorial integrity. In 2016, Galtung warned that Trump would accelerate American decline.

Peter Turchin, a quantitative historian at the University of Connecticut, published a forecast in Nature in 2010. Using cliodynamics, a field that applies mathematical models to historical cycles, he predicted that the 2020s would be the most unstable decade in America since the Civil War era. Turchin identified three structural drivers: elite overproduction (too many people competing for too few positions of power), popular immiseration (declining real wages against rising costs), and state fiscal crisis. He traced a roughly 50-year cycle of instability in American history: the 1870s, 1920s, 1970s, and now the 2020s. Each cycle peak brought political violence, institutional breakdown, or both.

Ray Dalio, the billionaire founder of Bridgewater Associates, developed an 18-indicator framework tracking the rise and fall of empires in his book “Principles for Dealing with the Changing World Order.” The US still scores #1 overall but is declining in nearly every indicator: education quality, economic competitiveness, military dominance relative to rivals, reserve currency share, internal conflict, and wealth inequality. Dalio’s key metric is the dollar’s share of global foreign exchange reserves. In 2001, the dollar held 72% of global reserves. As of the latest IMF data, that figure has fallen to 56.3%, the lowest since 1995. Dalio’s observation, borrowing Hemingway’s line about going bankrupt: empires decline “slowly at first, and then suddenly.”

Joseph Tainter, the anthropologist who wrote “The Collapse of Complex Societies,” argued that civilizations collapse when their investments in complexity (bureaucracy, military, infrastructure, regulation) hit diminishing marginal returns. Each additional dollar of complexity produces less benefit until the population concludes, consciously or not, that simplification is rational. The United States government paid $970 billion in interest on its debt in 2025. The CBO projects that will reach $2.1 trillion by 2036. Interest payments already exceed the defense budget. The government is spending more to service money it already borrowed than to run its military. That is textbook diminishing returns on complexity.

The Numbers Don’t Lie

The raw data is not ambiguous.

The national debt stands at $38.8 trillion and is accelerating, serviced by a Federal Reserve that has no choice but to keep printing because the alternative is default. The government has endured two partial shutdowns in 2026 alone. The fertility rate has fallen to 1.6 births per woman, well below the 2.1 replacement rate, and has been declining since 2007. For the first time since the Great Depression, more people are leaving the United States than entering it. The CBO projects deficits will reach 5.9% of GDP by 2030, a level it calls unsustainable. The post-9/11 wars cost an estimated $8 trillion and achieved none of their stated objectives.

And just a few days ago, the US government launched a war against a nation of 93 million people with over 610,000 active-duty troops and nearly a million including reserves, proxy forces in seven countries, and the strategic backing of both Russia and China. Within hours, Iran had struck US military bases across the Persian Gulf.

Every carrier group in the Gulf is a carrier group not in the Pacific. Every billion spent on cruise missiles is a billion not available for Pacific deterrence. Every sanctions package designed to punish Iran accelerates de-dollarization as nations seek alternatives to a currency that can be weaponized against them.

What You Should Do

The collapse of an empire is not the collapse of civilization. It is the collapse of a specific power structure. The Roman Empire fell, and people kept farming, trading, and building. The Soviet Union dissolved, and 290 million people woke up the next morning and went to work. What collapsed was the system of centralized control, not the society underneath it.

The people who fared best during the Soviet collapse were those who had already built networks outside the state: informal economies, local food production, community resilience, practical skills. The ones who waited for the government to fix itself suffered the most.

The same principle applies now. The parallel structures that liberty-minded people have been building for years, local food networks, alternative currencies, homeschool cooperatives, private membership associations, off-grid energy, community defense, these are not lifestyle choices anymore. They are the survival infrastructure for a transition that is already underway.

Reduce your dependency on federal systems. Every service the government provides is a service it can withdraw, restrict, or weaponize. Healthcare, education, banking, communications: build alternatives or connect with people who are building them.

Develop practical skills. In a declining empire, credentials lose value faster than competence. The person who can grow food, fix an engine, treat an injury, or build a structure will always have something to trade.

Build local. Empires collapse from the center outward. The further you are from the center (geographically, economically, and psychologically), the less the shockwaves affect you. The communities that thrive after imperial collapse are the ones that were already functioning as semi-autonomous units.

Don’t wait. Every model we examined points to the same window: now through the early 2030s. The dollar’s reserve status eroding, the debt spiraling, the military overextended, the political system unable to self-correct. You cannot predict the exact moment the water boils, but every thermometer in the room is rising.

Sources: Sir John Glubb, “The Fate of Empires” (1978), Peter Turchin, Nature (2010), Ray Dalio, “Principles for Dealing with the Changing World Order” (2021), Joseph Tainter, “The Collapse of Complex Societies” (1988), Johan Galtung via Vice (2016), IMF COFER Data, Peter Peterson Foundation, Brown University Costs of War Project

You Didn't Choose Conduent. Conduent Chose Not to Protect Your Data.

The Liberty Lookout — Sun, 22 Feb 2026 14:57:42 GMT

There’s a company called Conduent that most Americans have never heard of. It processes Medicaid claims, distributes SNAP benefits, handles child support payments, and manages unemployment insurance for more than 100 million people across the United States. If you’ve ever received a government benefit, there’s a real chance Conduent has your Social Security number, your medical history, and your home address.

Hackers had access to all of it for 84 days.

From October 21, 2024 to January 13, 2025, an unauthorized party roamed Conduent’s systems, pulling files containing names, Social Security numbers, medical records, and health insurance data. The SafePay ransomware group claimed responsibility and says it stole 8.5 terabytes of data. To put that in perspective: the entire printed collection of the Library of Congress is estimated at around 10 terabytes.

The victim count keeps climbing. Texas initially reported 4 million affected residents. That number has since ballooned to 15.4 million, more than half the state. Oregon’s attorney general lists 10.5 million. Maine’s breach registry shows 7.6 million total affected individuals. The combined total across just Texas and Oregon: more than 25 million people. Hundreds of thousands more are being notified in Delaware, Massachusetts, New Hampshire, and other states. New Hampshire alone keeps revising its numbers upward, from 11,000 to more than 181,000 in six separate disclosure letters.

Texas Attorney General Ken Paxton called it “likely the largest breach in U.S. history.”

Nobody knows the final number. Conduent won’t say.

The Company You Never Signed Up For

Here’s the part that matters more than the numbers.

You didn’t choose Conduent. You didn’t sign their terms of service. You didn’t hand them your Social Security number because you trusted their security practices. Your state government hired them. Blue Cross Blue Shield of Texas contracted with them. And through that chain of outsourcing, your most sensitive personal data landed on servers you didn’t know existed, managed by a company you’d never heard of.

When a company you chose has a breach, you can stop using them. When a government contractor has a breach, you were never given a choice in the first place.

This is the core problem with government data collection. The state takes your money, then requires you to surrender personal information to claw some of it back. Think of it as a mugger who throws some cash around the neighborhood, but only if you show him your ID first. Then he loses the ID. Every government benefit program is a data collection program, and every data collection program is a future breach waiting to happen.

Conduent isn’t some scrappy startup. It’s the former business process services arm of Xerox, spun off in 2017 with $6.7 billion in annual revenue and 54,000 employees. Its predecessor, ACS, held over 1,700 government contracts. This is the blue-chip vendor class. The one that’s supposed to be safe.

It has a track record. ACS/Xerox/Conduent paid a $236 million settlement to Texas for rubber-stamping fraudulent Medicaid claims from 2004 to 2012. Texas responded by awarding Conduent another $147.7 million Medicaid contract in 2022. The entity that defrauded Medicaid patients is the same entity that just lost Medicaid patients’ data.

Ten Months of Silence

Conduent discovered the breach on January 13, 2025. Individual notifications didn’t begin until October 2025. That’s a ten-month gap between discovery and disclosure. People are finding out about a breach that started in October 2024 only now, in February 2026, more than a year after it ended.

Conduent says it “acted promptly and in alignment with incident response protocols.” Translation: their protocol includes taking ten months to tell you your Social Security number was stolen while they figure out a way to put a media spin on it.

Blue Cross Blue Shield of Montana learned it was affected in January 2025 and notified individuals in October 2025. When Montana’s insurance regulator scheduled a public hearing about the delay, BCBS Montana and Conduent tried to get a court order to block it. The judge denied them. Montana’s CSI Communications Director Tyler Newcomb said it plainly: “It is troubling that it appears [BCBS] attempted to avoid regulatory oversight and accountability by seeking to block this hearing.”

The incentive structure rewards delay. Notifications cost money ($25 million, per Conduent’s own SEC filing). Each month of “analysis” is another month of not paying for credit monitoring. The lawsuits come either way.

The Enforcement Joke

At least nine class action lawsuits have been filed and consolidated in New Jersey federal court. Texas AG Paxton issued Civil Investigative Demands. Montana held its hearing. Sounds like accountability.

Zoom out. The HHS Office for Civil Rights has collected $144.9 million across 152 HIPAA enforcement actions in its entire history. That’s less than one Super Bowl ad buy. HIPAA penalties in 2026 range from $145 to $2.19 million per violation. For a company that launched with $6.7 billion in revenue, those are rounding errors.

Change Healthcare exposed 190 million people in 2024. Anthem lost 78.8 million records in 2015. The pattern never changes: breach, delay, disclose, fine, repeat. The fines never exceed the profits. The contractors never lose their government contracts. Nobody goes to jail.

The people whose data was stolen (Medicaid recipients, SNAP users, unemployment claimants) are the least equipped to monitor their credit, freeze their files, or lawyer up. The system extracts data from the most vulnerable, hands it to the least accountable, and offers credit monitoring as a consolation prize.

What to Watch

The number isn’t done growing. Conduent says it handles data for 100 million people. Only 25 million have been confirmed affected in two states. As more states report and more clients come forward (Volvo Group just disclosed 17,000 employees affected), the final count could be multiples of what we know today.

The consolidated class action in New Jersey will be the main legal front. Watch for discovery deadlines, any evidence of what “basic security measures” Conduent did or didn’t implement, and whether SafePay’s 8.5 terabytes ever surface on the dark web. Conduent says they haven’t. That claim has an expiration date.

The deeper question is whether this changes anything structural. Government agencies have been outsourcing data processing to the lowest bidder for decades. The bidder gets the contract, the data, and the liability shield. The people whose data it is get a letter in the mail, sometimes a year later, and a phone number to call between 9 AM and 6:30 PM Eastern.

If your data was in Conduent’s systems, you’re stuck in the same place as 25 million other Americans: waiting for a notification letter that may never come, hoping hackers don’t use your information before you find out they have it, and wondering why you never had a say in who got to hold it.

Sources: TechCrunch, Conduent Official Notice, Texas AG, Security Magazine, HIPAA Journal, Malwarebytes, Oregon DOJ, Maine AG, WMUR, HHS OCR, SecurityWeek, Freedom for All Americans, BleepingComputer, TDMR, NJ.com

The Deportation Machine: How Both Parties Built a Trap and You're Cheering for It

The Liberty Lookout — Thu, 19 Feb 2026 15:06:13 GMT

I’ve been watching social media the past couple days, and something is eating at me.

A man named Edgar lived in the United States for over twenty years. He worked construction. He started a family with his wife Amavilia in South Florida. They had a baby. Within days of his son’s birth, Edgar was detained on a 2016 warrant for driving without a license, transferred to ICE custody, and deported to Guatemala. His wife now wakes at 3 AM to cook lunches she sells for ten dollars each, trying to feed two children alone.

The comment sections are predictable. “Tough. Shouldn’t have broken the law.”

In Wilder, Idaho, ICE conducted a mass raid at a family event. According to an ACLU lawsuit, people were shoved to the ground, zip-tied at gunpoint, denied food and water. Juana Rodriguez, a 30-year-old woman born in Idaho (that’s a US citizen, for those keeping score), begged officers to remove her zip-ties so she could console her sobbing 3-year-old son. They wouldn’t.

“Shouldn’t have been there.”

In April 2025, ICE deported three US citizen children, ages 2, 4, and 7, along with their mothers from New Orleans. American kids. Deported from their own country.

“Well, their parents shouldn’t have broken the law.”

Let me be very clear about what I’m not doing here. I’m not writing a progressive “open borders” piece. I’m not writing a conservative “secure the border” piece. Both of those positions are part of a trap that’s been designed for you, and most of you are walking right into it with your eyes shut and your fists pumping.

This is a liberty piece. And if it doesn’t make you at least a little uncomfortable, regardless of your politics, I haven’t done my job.

The System Was Designed to Fail

Here’s a question the “just do it legally” crowd never asks: How, exactly?

The Cato Institute (not exactly a bleeding-heart outfit) published a comprehensive report called “Why Legal Immigration Is Nearly Impossible.” Their conclusion: employer sponsorship is buried under an “almost insurmountable barrier of bureaucratic red tape” that excludes nearly all workers without college degrees. Many applicants will literally die of old age before their number comes up.

This isn’t hyperbole. The USCIS backlog hit a record 11.3 million pending cases in 2025. Immigration courts have 3.7 million pending cases with average waits near four years. Family visa categories from Mexico, India, the Philippines, and China face waits measured in decades, not years. Mexico alone has 1.2 million people in the family-sponsored backlog. India has 291,000. Employment-based visas from India can mean 10 to 15 years of waiting.

Per-country caps limit each nation to 7% of annual allocation regardless of demand. A software engineer from India and one from Luxembourg get the same quota, despite India having 1,400 times more people.

Immigration law has been called “second only to the Internal Revenue Code in complexity” and a system where “a lawyer is often the only person who could thread the labyrinth.” An immigration attorney from Cornell confirmed it bluntly: “Immigration law has been called the second most complex area, second only to tax law, and there are often life-or-death consequences.”

Now compare this to how America actually processed immigrants when it was honest about what it was doing. At Ellis Island’s peak (1900-1914), an average of 1,900 people per day were processed. Most were through in three to five hours. Only 6% were rejected. Twelve million immigrants came through Ellis Island in total. They showed up, they were checked for disease and obvious disqualifiers, and they were in.

That was America when it actually wanted immigrants. Today’s system isn’t broken. It was designed this way. You create a labyrinth that virtually no low-skilled worker can navigate, you keep the backlogs measured in decades, and then you tell people who skipped the impossible line that they “should have done it legally.” It’s the bureaucratic equivalent of building a maze with no exit and then shooting anyone who climbs over the wall.

The Two-Step You Keep Falling For

Here’s the part that should make both sides angry.

Step One: Open the floodgates. Under the Biden administration, there were roughly 160,000 monthly border encounters on average. From FY2021 through early 2024, authorities encountered unauthorized migrants about 9.4 million times. The Department of Homeland Security documented 10.8 million total inadmissible encounters. Sanctuary jurisdictions (the DOJ has published an official list) explicitly limited cooperation with federal immigration enforcement. The message, whether stated or implied: come on in. Build a life. Put down roots. Enroll your kids in school. And then, when the political winds shift, all those roots you planted become handles for the next administration to grab. Sanctuary cities aren’t protection. They’re bait. They load the target population for the inevitable crackdown to harvest.

Step Two: Crack down with overwhelming force. The “One Big Beautiful Bill” allocated $170 billion for immigration enforcement. ICE’s budget went from $10 billion to $29 billion: nearly tripling. The stated goal: deport 1 million people per year. ICE plans to spend $38.3 billion converting warehouses into mega-detention centers, targeting 92,600 beds. For context, that $38.3 billion is more than the annual budgets of 22 US states. The Cato Institute estimates that at full capacity, ICE would be detaining 200,000+ people at any given time, with roughly 2 million cycling through annually.

Ten thousand new ICE officers are being hired. The sensitive locations policy that kept immigration agents away from schools, churches, and hospitals for decades was rescinded on Day One. The president invoked the Alien Enemies Act of 1798 to fast-track deportations. That same law provided the legal basis for the internment of 125,000 Japanese Americans during World War II, two-thirds of whom were US citizens.

See the pattern? One party opens the door. The other slams it shut with a battering ram. Both sides claim the other is the problem. And the ratchet only turns one direction: toward more state power.

The Brennan Center for Justice coined a chilling phrase for what’s being built: a “deportation-industrial complex.” Most of the new detention facilities will be run by for-profit private prison corporations, creating “strong economic and political constituencies that will make the new apparatus very difficult to dismantle.” Just like the military-industrial complex, once the contracts are signed and the lobbying dollars flow, this machine develops its own survival instincts.

Ask yourself: if they deport everyone they’re claiming to target, what happens to a $170 billion enforcement apparatus with 10,000 new agents and 92,600 detention beds? Does it quietly shut itself down? Has any government agency, ever, in the history of the world, voluntarily reduced its own budget and headcount because the job was done?

The TSA was supposed to be temporary. The Patriot Act was sold as an emergency measure with sunset provisions; its core powers were made permanent. Yale Law professor Oona Hathaway warned that the Authorization for Use of Military Force “massively expanded beyond what Congress initially envisioned,” setting “a dangerous precedent... that Congress might pass an authorization that it means to be very narrow or limited, and presidents can interpret those authorities in increasingly broad and creative ways.”

That’s not a bug. That’s the feature.

The Constitution-Free Zone You’re Already Living In

The Founding Fathers fought a revolution over warrantless searches. Two hundred and fifty years later, the government just drew a line on a map and gave itself permission to do them again. Federal regulations give Customs and Border Protection authority to operate within 100 miles of any US “external boundary”, including all land borders and coastlines. Within this zone, CBP claims authority to stop, frisk, detain, interrogate, and arrest people without a warrant.

Two-thirds of the US population, roughly 200 million people, lives in this zone. Entire states fall within it: Connecticut, Delaware, Florida, Hawaii, Maine, Massachusetts, New Hampshire, New Jersey, New York, Rhode Island, and Vermont. Nine of the ten largest US metro areas are inside it.

The regulation was adopted in 1953 by the Department of Justice without any public comment or debate. At the time, there were fewer than 1,100 Border Patrol agents. Today there are over 21,000, with thousands more on the way.

A Penn State Law Review article called this “constitutional erosion” affecting two-thirds of the American population. The Fourth Amendment is not fully suspended in this zone (the ACLU notes that agents still need “reasonable suspicion” for stops), but in practice, Border Patrol routinely exceeds its legal authority. And when they do? The ACLU documents “inadequate training for Border Patrol agents, a lack of oversight by CBP and the Department of Homeland Security, and the consistent failure of CBP to hold agents accountable for abuse.”

If you live anywhere near a coast or border (that’s most of you), you are already inside a zone where your Fourth Amendment protections exist more on paper than in practice. And the immigration enforcement buildup is supercharging it.

The Milgram Problem (Or: Why You Lost Your Empathy)

In 1961, psychologist Stanley Milgram set up an experiment at Yale to understand how ordinary people participated in the Holocaust. He expected to prove that Germans were uniquely susceptible to authoritarian compliance. What he found instead is that 65% of American participants administered what they believed were lethal electric shocks to another person, simply because an authority figure in a lab coat told them to continue.

They weren’t sadists. They expressed extreme distress. They sweated, they trembled, they begged to stop. But when the man in the lab coat said “the experiment requires that you continue,” most of them kept turning the dial.

Milgram never needed to go to Germany. He found authoritarianism in Connecticut.

As a Psychology Today analysis put it: “We watch in real time as millions of our fellow citizens demonstrate their willingness to follow authoritarian leaders who demand loyalty above truth, who attack democratic institutions, and who dehumanize those they define as ‘others.’”

Hannah Arendt watched Adolf Eichmann’s trial and concluded he was “neither perverted nor sadistic” but “terrifyingly normal.” He was a bureaucrat who followed orders. She wrote that “the essence of totalitarian government, and perhaps the nature of every bureaucracy, is to make functionaries and mere cogs in the administrative machinery out of men, and thus to dehumanize them.”

That’s the psychological machinery running right now. The state declares a category of person “illegal.” Authority figures repeat it. Media amplifies it. Research confirms that the term “illegal alien” measurably increases prejudice through “heightened perceptions of threat.” A 2011 legal study found that when the Supreme Court used dehumanizing metaphors like “invasion of illegal aliens,” cases were more likely to result in unfavorable outcomes for immigrants. The language isn’t an accident. It’s a technology.

And once the empathy switch is flipped, the Milgram 65% take over. Normal people who consider themselves moral look at a man who lived here 20 years, had a baby, and got deported over a decade-old traffic ticket, and their response is: “Shouldn’t have broken the law.”

That sentence is the modern equivalent of “I was just following orders.” It’s a moral off-switch wrapped in legalism.

The Real Beneficiaries (Hint: Not You)

Let’s follow the money and power, because that’s what this is actually about.

Corporations get a workforce they can exploit without consequence. Undocumented workers can’t complain about safety violations, can’t sue for wage theft, can’t organize. As one expert told the Loyolan: “Big companies love undocumented labor because it’s cheap, and it’s easy to exploit.” If they actually wanted to stop illegal immigration, they’d go after employers. They don’t, because the exploitation is the point.

The government gets expanded power. Every “crisis” justifies more authority, more funding, more surveillance. The $170 billion deportation apparatus being built right now will not be dismantled when the “crisis” ends, because crises never end when there’s money in perpetuating them.

The media gets engagement. Immigration is the perfect rage-fuel: emotional, visual, tribal. Left-leaning outlets run crying children at the border. Right-leaning outlets run crime stories involving immigrants. Both get clicks. Neither solves anything.

Working-class Americans and immigrant workers both lose. They’re pitted against each other by the same people who profit from both their labor. As the Atlantic documented, immigration has historically been viewed as “great for corporations (cheap labor) and consumers (lower prices) but bad for native-born workers.” Meanwhile the Maine AFL-CIO traced this pattern back to the 1880s: Irish workers pitted against Chinese workers, French Canadians described as “the Chinese of the eastern states.” Same playbook, different century.

This playbook is older than the republic itself. Every ruling class in history has used ethnic and immigrant scapegoating to keep the working class fighting each other instead of looking up.

Today It’s “Illegals.” Tomorrow It’s You.

Here’s where I lose whatever readers I haven’t already lost.

The infrastructure being built for immigration enforcement will be turned on American citizens. It always is. This isn’t speculation. It’s the most predictable pattern in the history of state power.

The Alien Enemies Act of 1798, originally targeting French nationals during a war scare, was used to intern 125,000 Japanese Americans in 1942. Two-thirds of them were born here.

The Patriot Act, sold as a tool against foreign terrorists, has been used for domestic surveillance of American citizens on a massive scale.

Tom Homan, Trump’s border czar, has already confirmed that ICE made “collateral arrests” of “many” American citizens during immigration operations, detained based on “location, occupation, physical appearance.” American citizens. Detained based on how they look.

The Brookings Institution warned that immigration surveillance databases and AI tools are “particularly concerning... given the nation’s history of using antiterrorism and national security measures to disproportionately surveil and infringe upon the civil liberties of minority groups.” And because data is stored indefinitely, “the expansion of immigration surveillance without robust privacy and free expression protections could have far-reaching effects on all Americans.”

When you cheer for a deportation machine with a $170 billion budget, 92,600 detention beds, 10,000 new officers, authority to raid churches and schools, and operations in a zone covering two-thirds of the country, you are not cheering for something that will only be used against “illegals.” You are cheering for infrastructure.

Today the target is someone who overstayed a visa twenty years ago. Tomorrow it’s gun owners who didn’t register under a new law. Or homeschoolers who didn’t comply with a curriculum mandate. Or crypto users who violated a financial regulation they’ve never heard of. Or someone who posted the wrong thing online.

Once you let the state define an entire category of human beings as fundamentally outside the protection of rights, you’ve handed it a template. All it needs to do next time is change the label.

The Ponerology of Immigration Policy (Or: The Architects Can’t Feel Pain)

If you’ve been following our series on ponerology, you already know the core insight: power structures attract people who are clinically incapable of empathy. Psychopathy is 3-4 times more prevalent among CEOs and people in positions of institutional authority than in the general population. The trait is 60-81% heritable based on twin studies.

These aren’t people who make cruel policy because they’re misguided. They make cruel policy because cruelty is invisible to them. They don’t experience the suffering they create. A system designed by people without empathy will inevitably produce outcomes that lack empathy. That’s not a moral failing. It’s a diagnostic criterion.

The immigration system isn’t broken by accident. The impossible backlogs, the decades-long waits, the Byzantine complexity, the deliberate under-funding of immigration courts while tripling the enforcement budget: these aren’t bugs. They’re design choices. You make legal immigration functionally impossible. You let millions come through the back door. Then you build a massive enforcement apparatus to go after them. At every stage, someone in power benefits. At no stage does the underlying problem get solved, because solving it would eliminate the leverage.

The people typing “shouldn’t have broken the law” on social media are playing the same role as Milgram’s subjects. They’re the 65%. Not evil. Not psychopathic. Just responding to authority signals with compliance instead of thought. The authority said these people are “illegal.” The law says they shouldn’t be here. Therefore their suffering is justified. QED.

Hannah Arendt would recognize this instantly. The banality of evil isn’t about monsters. It’s about normal people who stop thinking.

There Is No Political Solution

If you’ve read this far hoping I’ll tell you which party to vote for, or which immigration policy would fix this, I’m going to disappoint you.

The Democrats created the conditions for a crisis. The Republicans are using the crisis to build a police state. Both are funded by the same corporate donors who profit from exploitable labor. Both benefit from the wedge issue that keeps you arguing with your neighbor instead of questioning why you’re paying $170 billion so that private prison corporations can warehouse human beings.

Voting for “better immigration policy” is like asking the fox to redesign the henhouse. The people who designed this system are constitutionally incapable of designing one that serves human beings, because they don’t experience other human beings as fully real.

The answer, as always, is exit and build.

Stop debating immigration policy on social media. Stop letting the state define who counts as a person and who doesn’t. Stop giving your emotional energy to a system designed to harvest it.

Build communities that don’t need the state’s permission to exist. Build economic systems outside government control. Build human connections that cross the artificial lines drawn by people who see all of us (citizens and immigrants alike) as resources to be managed, not people to be respected.

The man in Idaho who zip-tied a US citizen mother and ignored her toddler’s screams didn’t do it because he hates Latinos. He did it because he was wearing a uniform, following orders, in a system designed by people who don’t feel what other humans feel. The person typing “shouldn’t have broken the law” isn’t evil. They’re just doing what Milgram predicted: following the authority, turning the dial, and not looking too hard at what’s on the other end.

The monsters were never supernatural. They just run the system. And until enough people see that clearly, the system will keep producing exactly what it was designed to produce: division, compliance, and power.

The dial goes higher. The question is whether you’ll keep turning it.