Thousands of “ghost students” use AI to steal millions in student loans, DOJ says
Thousands of “ghost students” use AI to steal millions in student loans, DOJ says
The U.S. Department of Justice charged 5 people in its first large-scale federal crackdown on “ghost student” fraud, where scammers stole Americans' identities to pose as college students and obtain federal student loans, and seized over $750,000 in cryptocurrency. The Education Department has identified over 830,000 potentially fraudulent federal aid accounts created with U.S. citizens' PII; if all were fake students, total losses would approach $130 million. Scammers used AI to fill out FAFSA applications, complete coursework, and fake online class attendance.
5 arrests so far
Applying for FAFSA may get messy as DOJ “ghost student” crackdown ramps up.
The US has charged five individuals as part of the first broad federal crackdown on “ghost students”—fraudsters who steal the identities of hundreds of Americans to pose as fake college students and steal student loans.
In a press release Thursday, the Department of Justice said it had uncovered a large-scale scheme relying partly on AI. The DOJ seized more than $750,000 in cryptocurrency from “a network associated with Kenyan nationals” after data analysis helped flag likely fraudsters. Assistant Attorney General Colin M. McDonald of the DOJ’s National Fraud Enforcement Division credited new fraud-detection measures implemented by the Trump administration for the arrests. He said that “this is the first international seizure of its kind in Kenya,” while warning fraudsters that the crackdown has only just begun.
“If you steal American identities, if you steal taxpayer dollars, if you exploit programs meant to help students build a future—we will find you,” McDonald said. “Whether you are in Nebraska, New Jersey, Virginia, or Nairobi, we will pursue you.”
At a press conference, McDonald was joined by FBI and Department of Education officials to explain how the scheme works.
It starts with bad actors purchasing personally identifiable information (PII) of Americans—most frequently targeting high school graduates who never attended college and went straight into the workforce.
Once they have that sensitive data, they fill out the Free Application for Federal Student Aid (FAFSA) while seeking to enroll in universities and community colleges. If accepted, student aid is distributed first to schools to cover tuition and fees, but any leftover money is deposited into bank accounts controlled by the fraudsters. Dozens of Americans have opened bank accounts to act as money mules for fraudsters, the DOJ said.
To keep money flowing, bad actors use AI to fill out applications, complete coursework, and even make it appear that a student is attending virtual sessions.
“It looks like the ghost student is actually completing the class,” McDonald said at the press conference. “Some even create the appearance of a fake student in an online classroom to trick schools into believing a real person is present. The goal is to keep the ghost student in school for as long as possible, semester after semester after semester.”
Oblivious Americans stuck with bills
So far, the DOE has “identified over 830,000 potentially fraudulent federal student aid accounts created by using this PII of US citizens,” officials said at the press conference. If all those accounts are ghost students, total losses come close to $130 million, the DOJ said.
For Americans whose data is stolen, there’s little opportunity to discover the scheme until the government comes collecting on loans years later. In the meantime, victims’ credit can be damaged. Once they detect the scheme, they’re stuck defending against demands to repay inflated student loans, which McDonald admitted is messy.
“The Department of Education does a great job to help unravel that mess, but it is a mess that no one should have to endure,” he said.
In connection with the arrests, officials said they “identified more than $32 million in actual loss tied to these specific subjects” and “we believe that actual total is substantially higher,” with “millions more suspected.”
The DOE’s Office of Inspector General first detected the fraud patterns that brought the alleged scheme to light. The FBI then helped identify how many accounts were impacted, tracing money to US and foreign bank accounts and crypto wallets. In one case, a defendant “used multiple bank accounts across several financial institutions to receive over 700 fraudulently obtained federal student aid refund deposits, totaling approximately $1,439,548,” the DOJ said. In another, “one person received more than $1 million in deposits that were linked to more than 300 fraudulent student accounts,” Inside Higher Ed reported.
At the press conference, First Assistant US Attorney Theophani K. Stamos for the Eastern District of Virginia said that data analysis “identified unusual, suspicious patterns of behavior” that the FBI said indicated “coordinated fraud was happening involving FAFSA.” Those patterns gave investigators “a solid basis and a solid lead to open further inquiries that have led to indictments,” Stamos said.
To keep the crackdown going, his office has sent “approximately 50 referrals to 26 FBI field offices across the country” to hunt down stolen taxpayer funds.
“We are just getting started. Working with partners around the world, we are sending a clear message: student aid belongs to real students, the data always leaves a trail, and we will follow it wherever it leads,” Stamos said.
Automated fraud detection isn’t perfect
The DOJ suspects this scam has been running since at least 2021, growing increasingly sophisticated as AI opens new avenues for loan theft.
“These fraudsters are exploiting Americans who have never been to college before, preventing the Department of Education from cross-checking applicants,” the DOJ said. “Many known victims have reported attaining only high school diplomas—and are shocked by their faked attendance at an institution hundreds of miles away.”
At the press conference, Undersecretary of Education Nicholas Kent accused the Biden administration of leaving Americans unprotected.
“When we walked in the door last January, we were extremely troubled to find that the Biden administration had turned off nearly every fraud detection measure,” Kent said.
Trump’s DOE said it introduced online identity verification requirements that Kent claimed would stop ghost students from ever receiving funds. Relying on automated systems to flag behaviors and reject fraudulent FAFSA applications, Federal Trade Commission Chairman Andrew Ferguson, vice chair of the White House Task Force to Eliminate Fraud, insisted that fraud could be easily stopped without allocating staff and resources or overly burdening the US government.
“The computers help us find it, and we keep the money in the agency,” Ferguson said.
The automated system isn’t perfect, however, and advocates have warned that false flags burden schools, meaning students’ access to aid could be delayed as administrators work through the added identity-verification processes.
Glitches have already affected fraud detection and student applications. In August, the DOE confirmed a technical issue from June 29 to July 6 prevented fraud evaluation results from being shared with certain FAFSA forms. After detecting the glitch, the DOE reprocessed those applications.
Real students may be flagged as fakes
Colleges and universities are also considering adding fees to stop the ghost student fraud.
Some advocates have acknowledged that the fraud deprives real students of funds. But they’re concerned that both the fees and online identity verification could become barriers to access to higher education for low-income students.
To verify their identities through the online FAFSA portal, students are expected to have government-issued IDs on hand in “real time.” They hold the IDs up to their cameras, and automated systems review them. A senior DOE official told Insider Higher Ed that the systems are “similar to the ones used by financial services companies like banks” and that he believed they are the same ones used by rental car companies.
Automated systems then assign a risk rating to the student, and anyone flagged as “high risk” is rejected. Once a student is rejected for aid, it’s up to the university to conduct any follow-up to verify if the application was real.
MorraLee Keller, a senior consultant for the National College Attainment Network, told Inside Higher Ed that “a lot of low-economic kids may not have secured, for example, a driver’s license” and may be set back in the application process while trying to obtain alternative IDs. That could also add fees during the application process.
Students may also be subject to automated fraud detection when seeking to make corrections to their FAFSA, which could delay fund distribution if a false rejection is issued.
“We want to make sure that timing doesn’t interrupt the aid getting credited to their account to pay their bills on time so that they could start classes, get refunds to go get their books, and all those kinds of things,” Keller said.
The Trump administration has said that it expects the number of real students flagged as fraudulent to be low. But the president of the National Association of Student Financial Aid Administrators, Melanie Storey, pushed back “a little bit” on that claim during a podcast from the association recently, asking “the department to quantify the volume of in-person verification that will remain,” Inside Higher Ed reported in April. At the time, Aaron Lemon-Strauss, executive director of the FAFSA program, replied, “I’m not going to be able to give you a number yet.”
Storey noted that many people complete the FAFSA form at in-person events, further narrowing potential false positives as the government expands its ghost student crackdown. For any students who “accidentally get caught in the net,” Lemon-Strauss recommended that universities remain vigilant and take action to ensure support goes to Americans who need it most.
Ashley is a senior policy reporter for Ars Technica, dedicated to tracking social impacts of emerging policies and new technologies. She is a Chicago-based journalist with 20 years of experience.
Source: Ars Technica · AI · arstechnica.com