The Trump administration is deploying artificial intelligence to conduct a sweeping review of more than 55 million people who hold U.S. visas, the State Department has told news outlets, part of a new "continuous vetting" process aimed at identifying visa overstays, criminal activity or links to terrorism.
Officials said the automated screening will examine a wide swath of records, including social media accounts and immigration files from target countries. The move follows recent policy changes that tightened access to student visas and comes amid a roughly 20% reduction in State Department staffing that officials and former employees say has complicated manual enforcement efforts.
A former State Department employee described the initiative to the Daily Mail as a form of psychological pressure intended to prompt self-deportations, saying "they don't need to scrub 55 million. They just need to say they are casting the net as extensively as possible, to encourage those who know they are ineligible ... to self-deport before they are caught by the federal government and punished." Current and former officials, however, said in interviews that the practical targeting is likely to prioritize certain countries and smaller groups rather than every individual on record.
Immigration policy experts and technology analysts urged greater transparency about how the automated systems will operate and how flagged cases will be reviewed. Julia Gelatt, associate director of the U.S. Immigration Policy Program at the Migration Policy Institute, said different government databases have varying levels of completeness and accuracy, and cautioned that incomplete records — such as arrests that did not result in conviction — could lead to wrongful visa revocations.
"There is just a lot we don't know about how the State Department is going about this," Gelatt said. She suggested the process could resemble an "ongoing database check" similar to existing continuous monitoring systems used by enforcement agencies, and called for clearer public information about criteria, oversight and remedies for people flagged by automated tools.
Technology analyst Rob Enderle, president of the Enderle Group, warned that prioritizing speed and productivity over accuracy in AI-driven systems could have serious consequences. "The odds of this ending very poorly for many people is exceptionally high," he said, arguing that the platforms need extensive testing and human review before being used to make enforcement decisions.
The concerns are rooted in recent incidents in which automated or expedited reviews appeared to produce erroneous results. In April, a Japanese student at Brigham Young University, Suguru Onda, had his U.S. visa terminated after what officials later attributed to an apparent software error tied to minor traffic violations. His attorney said officials were not thoroughly reviewing AI-flagged cases. Separately, a Tufts University student, Rümeysa Öztürk, was arrested in March after her F-1 visa was revoked and she was held by Homeland Security agents, drawing criticism from lawmakers and civil rights groups.
Administration officials have defended the vetting effort as a matter of national security and immigration integrity. A State Department spokeswoman told news outlets that all U.S. visa holders would face continuous vetting to identify reasons that could bar admission or continued presence in the United States. A separate agency official speaking to Fox News said that the student visa revocations under the administration involved individuals who had broken the law or expressed support for terrorism.
According to Department of Homeland Security data cited by officials, there were nearly 13 million lawful permanent residents and almost 4 million people in the United States on temporary visas last year. The State Department and DHS have estimated that hundreds of millions of visa records exist globally, but the 55 million figure cited in recent reporting refers to the number of people holding valid U.S. visas who could be subject to the new checks.
Some current State Department employees expressed surprise at the scope of the plan. "That sounds insane. I am just happy I am not in consular affairs," one employee told the Daily Mail. Former staff members said the department has already used automated services in the effort to reduce student visa approvals and that the recent staff reductions make a fully manual review infeasible.
Advocates for immigrants and civil liberties groups called for safeguards to prevent wrongful revocations and deportations. They urged that any automated screening be accompanied by human adjudication, clear notice requirements and accessible appeals procedures. Gelatt highlighted that some people flagged in databases may not even be physically present in the United States, and that sweeping public statements about the size of the review could produce unintended consequences.
Administration officials have not released detailed technical documentation about the algorithms, data sources or error rates that will govern the vetting system, and critics said that omission raises legal and ethical questions. Enderle said the system should be subject to rigorous testing and independent oversight until error rates drop to acceptable levels, but he expressed skepticism that such safeguards will be implemented given personnel cuts and an emphasis on rapid results.
The pace of the rollout and the criteria used to prioritize targets remain unclear. Officials said they will draw on "available information" including immigration records from applicants' countries of origin and public social media content, but did not specify how foreign records would be validated or reconciled with U.S. data.
As implementation proceeds, lawmakers, immigrant communities and legal advocates said they will monitor for evidence of disparate impacts, wrongful revocations and inadequate review procedures. The State Department has said the program is aimed at safeguarding the immigration system, while outside experts and affected individuals have warned that automated vetting at this scale risks mistakes that could separate people from education, employment or family ties without sufficient recourse.