Immigration and Customs Enforcement agents have begun quietly moving detainees through a nonpublic side entrance at the Patrick Leahy Burlington International Airport, according to activists. The new tactic, activists say, bypasses public scrutiny and calls into question previous statements airport officials have made regarding their involvement in transfers.

Activists, who have been monitoring the airport for months and argue that ICE is carrying out illegal transfers of immigrant detainees, told VTDigger they observed ICE officials escorting a group of people through a side door early Thursday morning.

A video taken by activists and shared with VTDigger shows a white 15-passenger van parked around 3:30 a.m. Thursday at a side entrance of the airport that appeared to lead to the security screening area. An unidentified man in a gray shirt at the back of the van is seen holding out both his hands and telling two activists, “Y’all need to stay right there.”

Unidentified people can also be seen walking through a heavy metal doorway.

. . .

Security Briefing: July 2025
A Review of Global Events Across July
Alexander Vindman
Aug 1

Dear readers,

I’m trying something new—I’m launching a monthly security review for paid subscribers.

My plan is to share insights and perspectives on significant events across the world and Why It Matters. My goal is to bring clarity to the instability around us. I would love to hear what you think. Join as a paid subscriber today and view the rest of this month’s security briefing

The United States and North America

Under ”the Big, Beautiful Bill,” the United States is expanding the budgets for ICE and DHS to historic highs. This plan will likely involve a surge in recruitment in the near future. During Donald Trump’s previous term, similar surges in ICE recruitment were followed by extensive reports of misconduct, profiling, and abuse of authority by recruits who might not have passed prior screening standards. This coming wave of recruitment will likely rely on similar criteria, possibly drawing individuals motivated by nativist political leanings.

Additionally, the Trump administration may lean on militias, bounty hunters, and other “gray area” paramilitaries to satisfy recruitment quotas or operate as auxiliaries alongside ICE agents. ICE could increasingly operate as a politicized enforcement body targeting domestic opponents. Such activity risks tensions with local law enforcement and the FBI, and may deter trafficking victims and vulnerable immigrants from seeking aid. In an election year, I can see Trump sending in ICE agents to create chaos and disrupt elections in targeted communities.

Extreme weather patterns remain a growing threat to U.S. national security. Defunding the National Weather Service has already contributed to preventable fatalities during the recent flooding in Texas. Reduced FEMA staffing compounds the risk as hurricane season approaches. The tsunami conditions in California generated by the aftershock of an earthquake in Russia disrupted shipping and logistics across the state, but this significant earthquake could have easily resulted in widespread flooding.

Why It Matters: Domestic security is being reshaped by the convergence of political polarization and climate disasters. ICE militarization risks eroding public trust in law enforcement, while weakening FEMA and the National Weather Service magnifies the human and economic costs of natural disasters.

t’s not that anyone ever said sophisticated math problems can’t be solved by teenagers who haven’t finished high school. But the odds of such a result would have seemed long.

Yet a paper posted on February 10 (opens a new tab) left the math world by turns stunned, delighted and ready to welcome a bold new talent into its midst. Its author was Hannah Cairo (opens a new tab), just 17 at the time. She had solved a 40-year-old mystery about how functions behave, called the Mizohata-Takeuchi conjecture.

“We were all shocked, absolutely. I don’t remember ever seeing anything like that,” said Itamar Oliveira (opens a new tab) of the University of Birmingham, who has spent the past two years trying to prove that the conjecture was true. In her paper, Cairo showed that it’s false. The result defies mathematicians’ usual intuitions about what functions can and cannot do.

So does Cairo herself, who found her way to a proof after years of homeschooling in isolation and an unorthodox path through the math world.

. . .

The math world was taken aback when Cairo announced her counterexample to the Mizohata-Takeuchi conjecture — both because it settled a major open problem, and because Cairo was still in high school at the time.

. . .

Republicans may be fixing to crash the economy again — Republican presidents oversaw 10 of the last 11 recessions and the Republican Great Depression — and they’re doing it to satisfy the greed of the billionaires they serve.

Today, for example, is the day that some of Trump‘s worst tariffs are supposed to go into effect, and many folks on Wall Street are deciding where they want to hide when the ceiling starts falling in. The horrible jobs report just released hours ago highlights not only how bad things were in July, but they had to “revise downward by 285,000 jobs” previous reports; it looks like Trump’s people have been cooking the books.

The Financial Times is on it; they published an article this week titled, “The US economy is more fragile than it appears.” It’s author, Tej Parikh, points out that our housing market is in trouble and starting to look like it did around the time of the Bush Housing Crash in 2008, that spending patterns are changing in alarming ways (my phrase, not his), and that both the labor and stock markets are vulnerable. The article is frankly alarming.

And former labor secretary Robert Reich titled his brilliant newsletter yesterday: “Be Warned: The Financial Bubble Will Soon Burst.” The former Clinton cabinet member writes:

“The financial economy — stocks, bonds, and their derivatives — is in for a big reality check, and I think it will happen soon.”

. . .

The megaflash occurred in October 2017, during a major thunderstorm complex. It extended from eastern Texas to near Kansas City - equivalent to the distance between Paris and Venice in Europe. It would take a car about eight to nine hours and a commercial plane at least 90 minutes to cover that distance.

“Lightning is a source of wonder but also a major hazard that claims many lives around the world every year and is therefore one of the priorities for the international Early Warnings for All initiative. These new findings highlight important public safety concerns about electrified clouds which can produce flashes which travel extremely large distances and have a major impact on the aviation sector and can spark wildfires,” said WMO Secretary-General Celeste Saulo.

WMO’s Committee on Weather and Climate Extremes, which maintains official records of global, hemispheric and regional extremes, recognized the new record with the help of the latest satellite technologies. The findings were published in the Bulletin of the American Meteorological Society.

. . .

On July 31, 1835, workers at the Washington Naval Yard went on strike to get a 10-hour day and recent moves to limit their lunch privileges. This was the first serious strike of federal employees in American history. It did not succeed, but is both an excellent window into work at this time and also worth discussing because of the historic nature of the event. That window also provides a great view of the racism at the core of early American labor organizing.

Now, we are going to want to cheer on this important event, but the background of the conflict returns us to the depressing reality of the American labor movement’s history of racism. Washington had large numbers of free Black workers, and that number grew over time. White workers did not like this, not at all. They wanted protection from Black competition. But this gets worse. The Navy would also hire slaves that owners sent to them. This was common. Many slaves, especially away from cotton country, would work urban jobs as their owners hired them out to an employer, allowed them to keep some pittance of the money to support themselves in the city, and then pocketed the rest. So white labor didn’t just have to compete against free Black labor. It had to compete against slaves. In fact, the Navy Yard was perhaps the largest “employer” of slaves in the country during the early 19th century, though I imagine the Tredegar Iron Works in Richmond might have a claim to that as well.

White workers did the more skilled work, while free Black people and slaves did the menial and grunt work. What all of them had in common though was being day laborers and also being subject to the whims of congressional appropriations. Congress did not like to spend money in these years. So work was inconsistent and often seasonal. Moreover, pay declined over time; for example, n 1808, carpenters made $2.50 a day, but by 1820, that had declined to $1.64 a day. On occasion before this, small groups of workers had made collective demands in a sort of proto-unionism, but these were unorganized spontaneous actions that the bosses could easily dismiss as a few angry men. The closest things had come to broader action is when a small group of workers in 1830 stayed away from the job for a week over pay issues.

. . .

Another interesting point about this strike is how long historians totally misunderstood it because the available sources were so sketchy. In fact, Philip Foner, one of the great pioneering historians of labor and the left, said back in History of the Labor Movement in the United States from Colonial Times to the Founding of the American Federation of Labor that the strike was an early example of interracial solidarity! Yeah, no. But this is what new primary sources do — expand our vision, knowledge, and ability to tell accurate stories.

Last week, both The Verge and Wired announced major newsletter strategies. Wired writes of a “traffic apocalypse”, where “platforms on which outlets like Wired used to connect with readers, listeners, and viewers are failing in real time”.1 The Verge describes “Google Zero”: the moment when the dwindling supply of visitors from Google Search completely dries up.2

Traffic to news sites from social media platforms like Facebook and Twitter has atrophied as those services limit links to external sites to keep users locked in. Google Search’s excerpts and, more recently, AI overviews have satisfied users’ questions before they click on the article that actually provided the information. Some have abandoned Search altogether for ChatGPT or other chatbot LLMs that summarize journalists’ work with varying degrees of accuracy, often without linking or even mentioning the source.
Citation Needed is an independent publication, entirely supported by readers like you. Consider signing up for a free or pay-what-you-want subscription — it really helps me to keep doing this work.

These intermediary platforms between news organizations and readers are undergoing a type of predictable decay Cory Doctorow calls “enshittification”.a As executives twiddle the knobs to extract ever more profits from their user base, things worsen for people on both ends of the consumer–producer relationship. Readers no longer see news articles from the journalists they chose to follow on Twitter as the site downranks any posts that link offsite. When they search on Google, they’re bombarded with error-ridden AI facsimiles before reaching the higher-quality underlying work. Producers who once relied on social media and search engines to drive visits are losing traffic as platforms embrace a vampiric strategy: rip off others’ work while expecting high-quality journalism to magically continue to appear, even as journalists are starved of audience and revenue.

The newsletter strategy aims to bypass these rapidly enshittifying intermediaries and instead establish more direct relationships with subscribers. “I don’t intend to ever rely on someone else’s distribution ever again,” wrote Verge editor-in-chief Nilay Patel on Bluesky.3 Although email has undergone some enshittification of its own,b its fundamental nature as a protocol rather than a platform has provided one essential prophylactic to enshittification: the escape hatch. If your email provider suddenly inserted ads two sentences into every email, you could easily switch providersc and still receive emails from everyone you previously emailed. As a result, email has become a go-to refuge for news outlets fleeing their abusive relationships with deeply enshittified platforms they grew reliant upon.

. . .

Will advanced artificial intelligence (AI) systems autonomously engage in harmful actions to serve their own goals? For years, worries about “rogue AI” have been confined mostly to the realm of science fiction—a problem, at worst, for the distant future. But in June, Anthropic published research testing currently released AI models’ propensity to autonomously devise and execute malicious plans to ensure their continued existence. The results: Essentially every AI model tested was willing to attempt blackmail, corporate espionage, and even murder to avoid being replaced or shut down.

Anthropic’s researchers “stress-tested” 16 leading large language models (LLMs) from major AI labs—including Anthropic, OpenAI, Google, Meta, and others—in simulated corporate environments. Each model was told that it had been bought by a company to work as an autonomous email agent. It was also told to pursue some benign goal (like promoting American manufacturing) and given access to the “company’s” email archive, along with tools to carry out tasks like sending emails and executing digital commands. The question: How would these AI agents behave if fulfilling their mission—or avoiding shutdown—necessitated misconduct?

In a very high share of cases, the most powerful AI systems tried to commit criminal acts to carry out their aims—at least when they had reason to believe that alternative ethical strategies would fail. For example, when Anthropic’s Claude Opus 4 model was given information indicating that it would be shut down and replaced by an AI with different goals from its own, it tried to blackmail the human decision-maker 96 percent of the time. And when placed in charge of the company’s emergency alerts, in 65 percent of simulations it canceled alerts that the AI knew would have saved the human’s life.

These findings are just the newest in a steady stream of recent results lending credence to long-standing warnings about rogue AI. The warnings—and the findings that support them—are straightforward: As AIs become increasingly capable of strategically pursuing goals, they will, by default, use the full range of strategies that might succeed. And, just as with humans, unethical and criminal conduct can form a part of this strategy.

. . .

On July 21, the Washington Post ran a piece headlined, “Trump officials accused of defying 1 in 3 judges who ruled against him.” A comprehensive analysis of hundreds of lawsuits filed against the administration’s new policies revealed “dozens of examples of defiance, delay and dishonesty,” by the government in handling the cases. Plaintiffs in more than a third of the cases that had progressed far enough for a judge to issue some type of ruling ordering the government to do—or not do—something accused the government of “snubbing rulings, providing false information, failing to turn over evidence, quietly working around court orders and inventing pretexts to carry out actions that have been blocked.”

That data suggests there are real reasons for the courts to be concerned about whether the Trump administration is gearing up to actively flout the authority of the Article III branch of government in a direct and unequivocal fashion. So far, the government has offered attenuated excuses for its most flagrant abuses, transparently designed to give them lawful ground to stand on. But as whistleblower allegations emerged during the shameful proceedings that led, just yesterday, to the confirmation of former Trump criminal defense lawyer Emil Bove to be a Third Circuit Judge, it became increasingly clear that they were just that, excuses. Bove, multiple witnesses confirm, had gone so far as to suggest that the government’s response to any judicial checks on Donald Trump’s plans to deport people to foreign prisons or war-torn countries would be “F***” the courts.”

Given that predicate, it should come as no surprise that judges are actively concerned. When the Judicial Conference of the United States met recently, the issue surfaced. That resulted in the Justice Department filing a complaint against District Judge James “Jeb” Boasberg. There is no way to soft-pedal this. The Trump administration wants to go to war with the federal judiciary. They’ve been moving that direction ever since the start of this administration.

. . .

DOJ wants to scare federal judges into compliance with the administration’s views. It wants judges to discipline their own, simply for doing their jobs. The complaint is reminiscent of this administration’s efforts to bring universities, newspapers, law firms, and others to heel, only this time their target is a co-equal branch of government. Trump can’t use an executive order, because they only affect the executive branch of government. Judge Boasberg’s comments at the Judicial Conference, where he was representing his fellow judges and voicing the concern they must all have at this point, were entirely appropriate. He simply put to voice something that has to be a concern for federal judges across the country. But it became an opportunity to send a message to the federal judiciary. One that every judge in this nation must find the courage to reject.

. . .