On October 30, the US Department of Education published new regulations governing Public Service Loan Forgiveness (PSLF), a program created in 2007 that cancels a borrower’s remaining federal student debt after ten years of repayment for those who work for a "public service" employer, generally defined as government or a non-profit entity. The idea is to reward people who work for the "public good." But as the Trump administration is likely to make clear going forward, one person’s "public good" might be someone else’s "public really bad," and it is dangerous to let the federal government—heck, just the executive branch—decide who gets special treatment based on something so subjective—and politically loaded.

Before getting into the details of the Trump regulations, make no mistake that Public Service Loan Forgiveness inherently makes a dubious judgement about who is more virtuous, lionizing those who work for putatively nonprofit entities—especially government—despite for-profit employers furnishing crucial stuff ranging from food to computers, and people such as doctors at nonprofit hospitals likely being no more selfless than those at for-profit.

As unhelpful as the distinction between employment by nonprofit and for-profit entities is, it is probably not as dangerous as deciding who serves the public good and who subverts it. And the Trump order is very much about subversion. Going forward, it would make employment by organizations deemed to be involved in any of the following, with a "substantial illegal purpose," ineligible for loan forgiveness:

• aiding and abetting violations of Federal immigration laws
• supporting terrorism or engaging in violence for the purpose of obstructing or influencing Federal Government policy
• engaging in the chemical and surgical castration or mutilation of children in violation of Federal or State law
• engaging in the trafficking of children to States for purposes of emancipation from their lawful parents in violation of Federal or State law
• engaging in a pattern of aiding and abetting illegal discrimination
• engaging in a pattern of violating State laws

A lot of these provisions specifically mention breaking laws. That might seem clear-cut, but there is subjectivity here, as the Secretary of Education decides which employers have run afoul of the law, doing so using only a preponderance of evidence standard. Basically, a 50 percent-or-greater certainty standard, and there is no way to prove that a secretary is slightly more than 50 percent sure of anything.

It is not difficult to imagine how this could play out. Are you a nonprofit employer with diversity, equity, and inclusion policies? Sorry, that violates the administration’s interpretation of the Civil Rights Act of 1964. Do you support sanctuary cities for immigrants? That breaks immigration law. Does the employer strongly support Palestine? Disqualified for supporting terrorism.

Employers who are cut off will be able to sue and possibly win, but to a large extent, the process will be the punishment, requiring significant employer time and money to fight the federal government, which defends itself with taxpayer dollars. And while letting a presidential administration decide which employers are worthy of conferring loan forgiveness on their employees is dangerous and unconstitutional under any administration, conservatives should, as always, remember that a future progressive administration could use this power against them.

President Sanders or Ocasio-Cortez could designate nonprofit employers, including school districts, that prohibit bathroom and locker room choice for transgender students as violating civil rights law or could cut off organizations they decide do not hire enough members of racial minority groups.

PSLF is a terrible policy all around. It favors one type of employee over another, fuels college price inflation, enables the executive branch to make political judgments about who serves the nebulous "public good," and violates the Constitution, which gives the feds no authority to engage in college student lending to begin with. It needs to end.

For many people, restrictions on the freedom to share their opinions online and listen to their favorite podcasts without government permission may seem like something out of a dystopian novel or horror movie. Increasingly, however, lawmakers in countries with time-honored traditions of respecting free speech have worked to pass legislation, though well-intentioned, aimed at protecting young people online by restricting a wider array of content without first uploading a government-issued ID.

It’s a scary story for our speech and privacy rights that is becoming increasingly and concerningly true.

Earlier this year, the United Kingdom’s Online Safety Act went into effect. The law places stringent requirements on websites and apps that might contain information deemed "harmful to minors." In the immediate aftermath of the law’s enforcement date, some major websites blocked access for non-age-verified users to content on the wars in Gaza and Ukraine to ensure compliance with the law. They did so under the (presumable) reasoning that, even though such content might be protected by the law’s exceptions for news, it still posed too much of a risk of being found in violation of the law’s vague terms and categories. Though done out of an abundance of caution, this action leaves many users bereft of important news and critical discourse on these topics.

When policymakers claim existing exceptions in the Online Safety law should render websites’ cautious content takedowns unnecessary or overzealous, the reality is that websites have little incentive not to err on the side of caution rather than carry speech that could possibly be seen by anyone as objectionable. Laws such as the Online Safety Act could also more generally chill speech on a variety of important topics, well beyond any newsworthiness exceptions for discourse around world events, including conversations about treatment for drug addiction or information about sexual and reproductive health.

Once again, if websites worry that anything they carry could trigger potential enforcement action, they are going to be cautious to the extreme, even if lawmakers say this wasn’t their intention. This may make it more difficult for individuals of all ages to find help or valuable and useful communities for information that is often difficult to discuss or could even be weaponized against them. For example, immigrants, dissidents, and those in the LGBTQ+ community all may be legitimately concerned about social or legal ramifications of their identities and conversations being compromised.

While some may initially applaud these laws and reason that they will not significantly impact their own internet experience, in fact, such restrictions are much broader than just the generally agreed-upon types of content problematic to children, such as pornography, which may already have restrictions in the real world. Today, UK users must upload government-issued IDs or otherwise verify their age online to access the full content of websites, including Spotify, Reddit, and X.

Privacy is also likely to be significantly harmed by these laws, and the UK’s experience shows sweeping laws aren’t just theoretical or restricted to bad-actor websites. Already, Discord, a growing platform that started as a host for gaming communities, has reported that more than 70,000 IDs the site collected in its compliance process were compromised. The increased collection of data could render the young people whom such laws were meant to protect, as well as adult internet users, more vulnerable.

Similar ID requirements are rapidly coming to the United States—and leaving a graveyard of once-functioning digital services in their wake. For example, a Mississippi law that requires age verification for a broad group of websites led X alternative Bluesky to exit the state due to concerns about the specific compliance changes that would be needed for the law and the time and cost for a smaller company to absorb such costs in just one state. While most of these types of laws are currently enjoined or have been defeated in court on First Amendment grounds, the potentially nightmarish consequences for both speech and privacy rights could soon become a reality if courts do find some versions of these laws amenable.

Just as Dr. Frankenstein intended his creation for good, the policymakers behind such legislation may genuinely mean to protect children. Yet the reality is that a top-down policy approach trades innovation, speech, and privacy while failing to actually protect children. Hopefully, by the next spooky season, we will be able to say that the First Amendment was the hero of our story in the United States, and not that we had to walk past a graveyard filled with rights and apps that were once beloved.

One of the striking features of the present administration is the regularity with which its leaders, from President Donald Trump on down, confidently describe the state of the law in ways entirely contrary to what had been seen as settled, on topics that range from flag burning to Congress’s TikTok ban to whether civil servants can be removed on a partisan basis. Sometimes, these proclamations may herald an effort to persuade courts to change prevailing doctrine, but at least as often they look more like an attempt to alter reality by establishing new legal facts on the ground.

On October 24, influential White House adviser Stephen Miller appeared on Fox News when the issue of whether authorities in Illinois could prosecute misbehaving federal immigration agents under state law arose. Miller responded, "To all ICE officers: you have federal immunity in the conduct of your duties. And anybody who lays a hand on you or tries to stop or obstruct you is committing a felony."

Whatever that is, it is not an accurate description of the state of the law. As Professor Steve Vladeck explains in this brief write-up, the actual rules are more complicated. There is indeed a zone of so-called Supremacy Clause immunity that will apply when "(1) the federal officer was performing an act that he was authorized to do by federal law; and (2) in performing the authorized act, the federal officer did no more than what was necessary and proper." When either condition is not satisfied—when the agent is taking an unauthorized action or is acting under authorization but in a manner that exceeds what is necessary and proper—the immunity based on federal supremacy ends.

As Vladeck notes, the prevailing rule in this category of immunity was formulated by Judge Michael McConnell, a conservative hero, and it does indeed allow state prosecutors to use state law to pursue instances of misbehavior by ICE agents.

Here’s another instance: in a piece at The Dispatch earlier this month, I tell how high administration officials, including Department of Homeland Security Secretary Kristi Noem and department spokeswoman Tricia McLaughlin, have repeatedly spoken as if citizens have no right to photograph or video record ICE raids or identify the officers by name. McLaughlin said, "Videotaping ICE law enforcement and posting photos and videos of them online is doxing our agents," and added, "We will prosecute those who illegally harass ICE agents to the fullest extent of the law." Noem went even further, describing "violence" against DHS agents as "anything that threatens them and their safety, so it is doxing them. It is videotaping them where they’re at."

In point of fact, however, the courts

aren’t on board with that sort of nonsense. While the Supreme Court itself hasn’t yet faced the issue squarely, the seven federal circuits that have done so—the 1st, 3rd, 5th, 7th, 9th, 10th, and 11th—all agree that the First Amendment protects the right to record police performing their duties in public.* Those circuits cover such populous states as California, Texas, Florida, Illinois, New Jersey, Georgia, and Pennsylvania.

Here at Cato we’ve had the honor of participating as amicus in at least three (that I know of) of these circuit precedents: Fields v. City of Philadelphia at the Third Circuit in 2017, Adkins v. Department of Homeland Security at the Ninth Circuit in 2018, and Irizarry v. Yehia at the Tenth Circuit in 2022. Adkins, the Ninth Circuit case, was especially interesting because it involved two citizens arrested by Customs and Border Protection (CBP) for taking pictures at a border crossing that they believed documented unlawful searches and other problems; the agency, which had seized the men’s cameras and deleted their pictures, asserted that the sidewalks they were standing on were property it officially controlled, but it lost anyway. (Thanks to colleague Dan Greenberg for helping compile the cases.)

To some important extent, one of the audiences Noem and her department are seeking to reach was the same one Miller was explicit in addressing: ICE agents themselves. If the agents come to believe that they have blanket immunity whatever they do, or that citizens have no right to record them, they are more likely to take aggressive informal action, such as grabbing phones or taking news reporters into custody on charges of obstruction (perhaps later quietly dropped). These informal methods of repression, I observe, can very much set the tone for enforcement, no matter whether the agency does or does not expect courts to say later.

If the agents are hearing a persistent message from their higher ups of "you’re immune no matter what you do," it’s up to the rest of us to disabuse them of that error.

Former Tennessee police officer Larry Bushart Jr. has spent most of October in jail. His crime? Posting a Facebook meme.

The trouble started after the assassination of conservative activist Charlie Kirk, when Larry responded to the public outrage with a somewhat callous take. He shared a meme on Facebook featuring a photo of President Trump, together with a quote: "We have to get over it." The meme said these words were taken from Trump’s remarks after a 2024 shooting at Iowa’s Perry High School. Above was a caption, "This seems relevant today..."

Larry happened to share the meme on a Facebook page dedicated to his community of Perry County, Tennessee, which also happens to have a "Perry County High School." But as The Intercept’s Liliana Segura noted, "it was easy to discern that, apart from the name ‘Perry,’ there was nothing connecting the meme" to the Tennessee school. And as Reason’s Joe Lancaster added, "In context, it’s clear Bushart meant to suggest that since Trump previously said people should ‘get over’ a school shooting, then they shouldn’t be expected to care about the murder of a conservative public figure." Indeed, two commenters on Larry’s post recommended he "take a stress pill or something" or go mow his lawn. Touch grass, as the kids say.

Enter Perry County Sheriff Nick Weems, who had posted his own Facebook message after Kirk’s assassination, saying, "Evil could be your neighbor. Evil could be standing right beside you in the grocery store. It could be your own family member, and you never even know it." Weems quickly arranged for an investigator to go to Larry’s house, followed by four officers with an arrest warrant. The warrant accused Larry of "Threatening Mass Violence at a School." The officer who read Larry his charge laughed, and Larry responded, "I’ve been in Facebook jail, but now I’m really in it."

What happened next is no joke: Larry has remained in jail ever since. Larry’s bond was set at 2ドル million. His next court date will not be until December 4.

Larry Bushart’s arrest and confinement violate the First Amendment. The Supreme Court has made it crystal clear that only true threats are exempt from the freedom of speech—not hyperbole and political bombast. As Reason pointed out:

"The U.S. Supreme Court created the true threat exception to the First Amendment in the 1969 decision Watts v. United States. Even then ... it made the point of distinguishing between true threats and ‘crude political hyperbole’—in that case, a protester’s remark that if he were drafted into the Army, ‘the first man I want to get in my sights is’ then-President Lyndon Johnson. The court agreed with the plaintiff that it was not a true threat but simply ‘a kind of very crude offensive method of stating a political opposition to the President.’"

In short, the Constitution protects rude, crude, and heartless words.

The irony at the heart of this case? Charlie Kirk was a champion of free speech. As Kirk wrote on X in 2024: "There’s ugly speech. There’s gross speech. There’s evil speech. And ALL of it is protected by the First Amendment."

Honor Charlie. Free Larry.

For the second time in as many years, the Cato Institute has sued the Department of Justice (DoJ) and the Federal Bureau of Investigation (FBI) for internal records regarding the Bureau’s use (or misuse) of Section 702 surveillance powers of the Foreign Intelligence Surveillance Act (FISA). First, some background.

This particular surveillance power, the legalized version of the previously illegal post‐9/​11 STELLAR WIND program, has been serially abused since its creation in 2008. Last year, Cato filed its first FISA Section 702-related lawsuit after the DoJ slow-rolled a Cato Freedom of Information Act (FOIA) request seeking internal DoJ and FBI records regarding the number of times FBI agents had improperly searched electronically stored data on Americans swept up under the Section 702 program. The first lawsuit gained urgency as the deadline for Congress to renew or let die the Section 702 program approached in April 2024.

However, the DoJ failed to turn over the requested records before Congress voted to extend the Section 702 program through late April 2026. It was not until July 2024 that the DOJ finally produced the records at issue—despite the fact that the records had been declassified by the DOJ almost five weeks before the April 2024 congressional vote on Section 702’s renewal. Given the abuses chronicled in those records, had they been made public prior to the congressional votes on the program, the Section 702 authority might well not have been renewed.

It’s long been well established that at least one person connected to President Trump’s first presidential campaign was improperly targeted in 2016 using a different FISA authority. What’s also long been known is that the kinds of abuses that led to that improper investigation (code-named CROSSFIRE HURRICANE) were far more widespread within the Bureau’s field offices than anyone realized. In light of those facts, one would think that Trump’s newly installed leadership at the DoJ and the FBI would be reevaluating the wisdom of continuing surveillance programs like FISA Section 702 and working to release more data about the problems with the program.

Instead, earlier this year, the FBI flatly refused to release any data to Cato about potential FISA Section 702 abuses that occurred between August 16, 2024, and January 24, 2025 (the date range contained in Cato’s FOIA request)—a refusal that has not triggered Cato’s second FISA Section 702 query audit incident records lawsuit.

In connection with the FBI’s aggressive "no records for you" stance, I think it’s worth noting that during his Senate confirmation hearing in January, then-FBI Director nominee Kash Patel stated he opposed a probable cause-based warrant requirement for FBI agents to access stored Section 702 data. Current CIA Director John Ratcliffe likewise expressed opposition to such a requirement during an appearance before the Senate Select Committee on Intelligence. Thus far, Attorney General Pam Bondi has not weighed in on the issue.

Two things are certain. The first is that the FISA Section 702 program is set to expire in late April 2026. The second is that the kind of public interest FOIA litigation Cato is engaged in on this case may be the only way the American people get a sense of whether the Trump-controlled FISA apparatus is keeping us safe from terrorists or is being used to target us, and thus should be allowed to expire.

Last month, a federal judge in Rhode Island struck down the Trump administration’s attempt to restrict National Endowment for the Arts (NEA) funding from groups that "promote gender ideology."

The outcome is legally unsurprising. For grants or subsidies to private actors, the First Amendment forbids the government from penalizing a particular viewpoint. In this case, the court also found Administrative Procedure Act defects: you can’t rewrite grant rules by fiat. An agency must offer a reasoned explanation for changing a policy like this.

The artists themselves celebrated the verdict as a victory for free speech. For conservatives, it’s another reason why waging the culture war using the government purse is a dead end. To achieve the outcomes they want, the NEA should be abolished rather than commandeered.

Back in May, Emory professor Mark Bauerlein pitched that the Trump administration should weaponize the NEA towards conservative ends in the culture war. One way they tried to do this practically was by eliminating funding for art encompassing expression about gender ideology.

I argued in response to Bauerlein’s piece that, as a matter of principle, government should never force taxpayers to fund artistic messages they disagree with. In practice, the artistic community is highly progressive, and any attempt to shape grants towards conservative viewpoints would lead to huge complaints from artists and be widely seen as debased art.

Given that Democrats hold executive power a lot of the time, and would change funding priorities accordingly in office, no conservative grant conditions would endure either. At best, any victories would be fleeting. The case for using these grants to advance culture war arguments was weak.

Well, now the courts have provided another reason the plan is doomed: enforcing conservative constraints on grantees’ speech is illegal.

The NEA had hoped to empower its chairperson to review applications "for artistic excellence and merit," weeding out "gender ideology" in the process. That was deemed illegal for two reasons. First, filtering applicants based on their viewpoint on gender violates the First Amendment. Second, agencies have a duty to justify themselves when they change their rules, something NEA was found not to have done.

Most libertarians would say, conditional on arts subsidies existing, that this is exactly right: the government should avoid policing speech, which inevitably means treating some views as valid and others taboo. "Gender ideology" is itself vague, inviting arbitrary enforcement and so further chilling protected speech. The First Amendment rightly subjects viewpoint discrimination to the highest levels of judicial scrutiny, which is why such restrictions almost never survive.

Yes, when the government commissions its own speech through art, it has latitude to control the message. The Supreme Court has also upheld a statute that allows the NEA to "take into consideration general standards of decency and respect" in its grantmaking process. But while this naturally means some censorship by subject selection, the courts do not allow the NEA to condition grants on specific viewpoints.

The result is that the NEA’s existence means conservatives will be funding grantees whose work they fundamentally disagree with. In fact, government-funded art is bound to violate someone’s freedom of conscience. And for what? It’s unnecessary in a free society. The NEA’s budget is tiny, and art that people want can be funded voluntarily through the private and non-profit sectors, where no single viewpoint has coercive power over others’ money.

Conservatives should take the loss and heed this lesson. The NEA cannot be "captured" nor "balanced" towards their cultural goals. The only way to avoid forcing Americans to subsidize art they reject is to listen to us libertarians and abolish the NEA altogether.

Challenging the prevailing narrative and upholding one’s principles in higher education is often a solitary endeavor. Even tenured colleagues sympathetic to the challenger’s ideas may retreat, fearing only exclusion from the next faculty cocktail party. That is why I notice when someone shows courage, as Mathew Abraham did in 2019 when I faced attacks for criticizing "anti-racist" pedagogy.

Now, Abraham, a tenured professor, stands on the brink of dismissal for doing exactly what tenure is supposed to protect: speaking freely and honestly.

I cannot stop the nonsense directed at him, but I can shed light on it. His story reveals much about the mindset and maneuvers of the academic left. A recent report in Campus Reform described how a coalition has requested that the Arizona Board of Regents investigate alleged retaliation and corruption related to "diversity, equity, and inclusion" (DEI) initiatives at the University of Arizona (U of A).

Abraham’s lawsuit against U of A was a pivotal moment. He eventually forced the release of documents that exposed the unprofessionalism of his colleagues. The lawsuit remains ongoing, but it has set the stage for future battles.

The pattern of retaliation began in 2022, when Abraham and two colleagues were excluded from the Committee of Academic Freedom and Tenure after internal emails labeled them "problematic." Though an investigation found procedures weren’t followed and faculty passed a corrective resolution, the targeting continued.

Abraham then proposed a campus dialogue between me and the rising critical race theorist, Aja Martinez. Despite my offer to cover the expenses, the dialogue around the discussion was twisted into a demand for debate, and the proposal was denied in a 5–3 vote. A short time later, the program approved, without a vote, an event centered on Robin DiAngelo’s White Fragility. The message was clear: orthodox views were welcome, dissent was forbidden.

In 2024, Abraham faced a seemingly intentional scheduling conflict between his teaching assignment and his new role as a prosecutor. Department chair Cristina Ramirez denied his requests to move the class online, something other professors were granted, and forbade further discussion. Administrators weaponized this into a "conflict of commitment" charge, capping Abraham’s outside work at five hours weekly and forcing him to choose between teaching and practicing law. A grievance committee found the dean’s justification unsupported, but Abraham’s discovery that Ramirez lacked the required authorization for her chair position was deemed "insubordination."

The pot boiled over when Abraham sent a faculty-wide email questioning these irregularities. On August 19, Dean Poloni-Staudinger recommended his dismissal for creating a "hostile environment." The provost placed him on paid leave the next day and revoked his campus access.

Abraham told his students that his removal was retaliation for his views and encouraged them to contact administrators. A retired professor, Aneil Rallin, wrote to the provost condemning the university’s actions as an assault on academic freedom and the purpose of tenure.

To sum up, a proposal for dialogue was blocked while an ideologically comfortable event proceeded without debate. A "scheduling error" was weaponized into professional misconduct. Refusal to comply with constitutionally questionable disclosure rules was labeled insubordination. Criticism, sharp but protected, was redefined as bullying.

Ultimately, the Arizona Board of Regents will likely investigate this as a DEI issue within the university, thereby exposing the detrimental effect of social justice ideologies on academic freedom and integrity. The intolerance and censure of faculty members who do not fall in line is not atypical.

This is about more than just Abraham’s experience. It is about how universities treat dissent. Abraham’s story looks, sounds, and feels familiar. I know I am not the only one who sees it.

This commentary was originally published in Minding the Campus , September 24, 2025.