In August, the US Citizenship and Immigration Services, the agency that adjudicates and grants immigration benefits like visas, residency, and naturalization, put out a bizarre three-page policy alert. Among other things, it noted vaguely that the agency would scrutinize applicants who “support or promote anti-American ideologies or activities” and “enforce all relevant immigration laws to the maximum degree, including the use of discretion, to deny the benefit request.” Effectively, it instructed evaluators to deny people engaged in anti-Americanism, which it did not define there or elsewhere.
What is anti-Americanism? What are those “ideologies or activities,” exactly? And without any meaningful guidance, how is anyone on either side of the immigration process supposed to identify it? Maybe the imprecision is the point. Three weeks on, practitioners tell The Verge that it is almost impossible to figure out how to advise clients on this standard or properly prepare for it. “The problem that I have been stating is that it’s entirely pretextual anyway, that having something be this vague is not a bug, it’s not a problem, it’s exactly what they’re going for. Because if they have it vague, then they can say that somebody can be disqualified for any reason they want,” said Pittsburgh-based immigration lawyer Adam S. Greenberg.
In one potential indication of where the administration could draw the line, Deputy Secretary of State Christopher Landau posted that people “praising, rationalizing, or making light of” Charlie Kirk’s assassination could be denied visas or stripped of status without pointing to any specific authority. The uncertainty “creates an environment of self-censorship where people delete their accounts or delete their posts or simply do not post things because they’re worried that they will have to report them to the government and that the government may read them and it might influence their immigration status in the future,” said Eva Galperin, the director of cybersecurity at the Electronic Frontier Foundation.
“Fear about whether this would impact applications has been around for a while, since March or even earlier, with the administration’s antisemitism policy,” said New York-based immigration attorney and scholar Cyrus Mehta, referencing the administration’s earlier announcement that it would screen applicants’ social media for supposedly antisemitic content. “The anti-Americanism policy really kind of flows from the antisemitism policy.”
Mehta said he’d attended multiple naturalization interviews since the announcement and had not seen it come up. Still, clients remain on edge and lawyers bewildered. “I don’t know how it’s going to be applied. It might be applied more vigorously, and I think it’s far more insidious than the antisemitism … It’s really vague, it’s really broad.” He said that his and many other attorneys’ belief in free speech “is totally antithetical to the advice that one may have to give to clients.”
Keeping immigration policies broad and discretionary to confuse applicants has been a penchant of Donald Trump’s going back to his first term. The public charge rule, for example, threatened immigrants with status denial over extraordinarily subjective analyses of their risk of becoming dependent on public assistance, which could take into account not just existing but hypothetical future use of benefits like food stamps. Ultimately, I’ve never heard of anyone denied status specifically on the basis of the expanded public charge analysis, at least in part because it was only briefly in effect before it was blocked by a federal judge during covid. Nonetheless, a 2024 report by the Urban Institute found widespread reluctance among immigrant and mixed-status families to use safety net benefits they were entitled to, including some that were state and local and would not have been part of the public charge calculus.
The idea at least in part was for people to overcorrect, and overcorrect they did. Now, Trump officials are doing the same but for speech. This is not the first immigration effort targeting what very much seems like protected speech, following the now-notorious detention of former Columbia graduate student Mahmoud Khalil and others involved in campus protest and activism, which shocked observers earlier this year. This does, however, seem far more expansive, and comes on the heels of Homeland Security and the State Department announcing that visa adjudications would require applicants to both share all their social media handles and set their accounts to be publicly visible, which together suggest that online criticism of the administration could now effectively be an obstacle to status. “It has clearly altered behavior. A lot of people say, ‘Well, I make sure that I don’t post anything.’ I’ve never really had that reaction before,” said Mehta.
The guidance does not specify how exactly government adjudicators would assess applicants for “anti-American” sentiment, though there are some concerning clues in other recent immigration actions. In April, the administration abruptly began mass-terminating thousands of student visas across the country before just as abruptly reversing course as it faced dozens of lawsuits and court losses. It soon emerged in court that administration officials had simply run international student information data through a federal crime information database and then directed the terminations without ever confirming if the data was accurate or if identified students had actually committed offenses that could trigger loss of status. Activists and lawyers worry that a similar approach could be used to identify supposed anti-American ideology.
“We can make guesses based on the half-assed technical implementations that the Trump administration has made since February. Especially the behavior of DOGE is usually to essentially create enormous troves of data and merge them all together in a deeply undifferentiated way,” said Galperin. “You get a lot of extremely inaccurate or useless data, and then either just do a search for incendiary keywords — which is how you end up getting funding pulled for talking about transitions or transgenic mice — and then there’s just feeding the entire thing into AI and having AI make the determination, or have AI write a summary, which, again, you are feeding garbage into a garbage machine that will then spit out more garbage.”
USCIS did not respond to a list of detailed questions, including how it defined anti-Americanism, what criteria it was giving its adjudicators, and if it would use any automated tools to evaluate applicants.
At least the students with canceled visas were able to pursue legal action on the basis that the government obviously lacked any real rationale to attempt to terminate their status. When it comes to the initial issuance of benefits like work and student visas or even permanent residency, the government under the law gets quite a fair bit of leeway, and doesn’t always have to explicitly lay out the reason for a denial. That opens up the possibility that applicants could be denied on this anti-American basis and not even know it.
“Adjustment [to permanent residency] is discretionary. Extensions of status are discretionary. Waivers are discretionary. They don’t have to provide their reasoning,” said Greenberg. While a federal court could in theory point out that a denial violates the First Amendment, there are limited avenues for people to actually appeal these decisions or get clarity on why they were denied, especially at consulates abroad.
“I think, for example, students applying for the visa to study in the US are particularly vulnerable,” said Mehta. “It would be very difficult for everybody to bring lawsuits. So a lot of people are trying to conform. A lot of people are basically avoiding, you know, posting.”
