In November 2021, then-Homeland Security Secretary Alejandro Mayorkas told Congress exactly what Americans wanted to hear about the tens of thousands of Afghans being airlifted out of Kabul. “It is our policy,” he testified, “to vet 100 percent of the individuals in those transit countries before they board a flight for the U.S.” Nearly five years, one fatal shooting two blocks from the White House, and three federal Inspector General reports later, the Department of Homeland Security’s own watchdog has now put a number on how that promise actually held up: a 40 percent error rate, found in a fresh audit of the case files it was describing.
The finding, released the week of September 29, 2026, is not a partisan talking point manufactured after the fact. It is the conclusion of a Department of Homeland Security Office of Inspector General review of Operation Allies Welcome — the program that resettled Afghan evacuees after the August 2021 withdrawal — and it lands on top of a documented, bipartisan paper trail stretching back to 2022 that Congress’s own watchdogs have been building for years. The gap between what officials promised the public and what auditors actually found is the story here, and it is a gap with a body count.
What the Audit Found
According to DHS’s own announcement of the findings, corroborated independently by the Washington Times and NewsNation, the DHS Inspector General’s recurrent-vetting review examined a sample of 678 Afghan evacuee case files from Operation Allies Welcome. Auditors found errors in roughly 40 percent of them, breaking down into:
- 303 files with missing aliases — alternate names or spellings that, left out of a record, can cause a database check to miss a legitimate hit
- 23 files missing required security checks entirely
- More than 150 additional files with other documentation gaps
The report describes at least two cases in which incomplete records — missing aliases paired with gaps in documentation — meant the individuals “may have potentially been inadmissible to the United States on terrorism-related grounds,” in the DHS characterization of the findings.
DHS Secretary Markwayne Mullin, who took over the department in March 2026, framed the findings bluntly in the department’s statement: “The Biden Administration dangerously mismanaged the Operation Allies Welcome program, disregarding all the thorough and well-established procedures of our asylum system and putting our nation at risk.” Charles Marino, a former senior DHS adviser, went further on the sample size itself: “If that sampling had been bigger, that 40 percent would be much, much higher.”
RSN could not independently verify Marino’s extrapolation — DHS’s own release does not project the 40 percent finding onto the full population of Afghan evacuees, and a larger sample could just as easily surface a different rate in either direction. That caveat matters, because the temptation with a number like “40 percent” is to round it up to “40 percent of Afghan evacuees are dangerous,” which is not what the audit says. It found paperwork and process failures in 40 percent of a 678-file sample. What those failures actually let through the door is the harder — and more important — question.
This Was Not the First Warning
Operation Allies Welcome evacuated more than 70,000 Afghans in the chaotic weeks surrounding the August 2021 withdrawal, part of a broader resettlement effort that, combined with the follow-on “Enduring Welcome” program, has brought close to 190,000 Afghans to the United States since. The scale and speed of that operation is relevant context — this was not an orderly, years-long visa process, but an emergency airlift conducted under battlefield conditions as the Taliban retook Kabul. That context does not excuse what auditors have since found, but it explains why the vetting apparatus was under strain from day one.
DHS’s own Inspector General flagged problems years before this month’s report. A September 2022 OIG report warned that some evacuees “were not fully vetted” and that some of the information used to vet them “was inaccurate, incomplete, or missing” — language that reads, almost word for word, like a preview of this month’s findings. A second report, OIG-24-24, issued in May 2024, concluded that “DHS Has a Fragmented Process for Identifying and Resolving Derogatory Information for Operation Allies Welcome Parolees” and made five recommendations, all of which DHS said it agreed with. Sen. Chuck Grassley, chairman of the Senate Judiciary Committee, wrote to DHS in October 2024 noting that despite that report and its concurred-with recommendations, he had received no answers to a June 2024 follow-up inquiry about whether any of them had actually been implemented.
The problem was not confined to DHS’s own review process. At a January 14, 2026, joint hearing of the Senate Judiciary Committee’s border security and counterterrorism subcommittees, Michael J. Roark, Deputy Inspector General for Evaluations at the Department of Defense, testified that the National Counterterrorism Center “did not have access to some DoD biometric and contextual data” — including fingerprints recovered from explosive devices and related intelligence reports — while conducting the initial 2021 vetting. Separately, Roark testified that Defense Department personnel “removed Afghans from the BEWL [Biometrically-Enabled Watchlist] for reasons not cited in the SOPs [standard operating procedures]” and, in at least some instances, “incorrectly applied removal criteria for U.S. persons to Afghan evacuees” — meaning watchlist-removal standards meant for American citizens were mistakenly used to clear foreign nationals. By September 17, 2021, according to Roark’s testimony, the National Ground Intelligence Center had already identified 31 Afghans with derogatory information who were in the United States; by November 2, that number had grown to 50.
The Case That Made It Personal
This is not an abstract paperwork story. On November 26, 2025, according to the U.S. Attorney’s Office for the District of Columbia, a man prosecutors identify as Rahmanullah Lakanwal opened fire without provocation near the Farragut West Metro station, two blocks from the White House, on a group of National Guard members. Army National Guard Specialist Sarah Beckstrom, 20, was shot in the head and killed. Sergeant Andrew Wolfe was critically wounded; two other Guardsmen were also hurt. Lakanwal was indicted in June 2026 on 17 counts, including first-degree murder while armed, with capital-eligible charges. He has pleaded not guilty, and the case remains active — RSN takes no position on his guilt and relies here solely on the Justice Department’s own public charging documents.
What is not in dispute is his immigration history. Lakanwal entered the United States in September 2021 under Operation Allies Welcome, having previously worked with a CIA-affiliated partner force in Kandahar. He applied for asylum during the Biden administration; that application was approved in April 2025, under the Trump administration, according to a detailed timeline compiled by FactCheck.org. Officials have said Lakanwal “underwent more vetting than most Afghans” processed through the program — which, if true, is not the reassurance it sounds like. It means the case that ended in a National Guardsman’s death was not the one that slipped through on the thinnest file; it was, by DHS’s own account, one of the more thoroughly reviewed ones.
That timeline also cuts against any effort to make this a purely partisan story. The initial vetting gaps that let Lakanwal and others into the country trace to the 2021 evacuation under the Biden administration. But his asylum grant — the final approval that gave him lawful status — was signed off in 2025, under the current administration’s own U.S. Citizenship and Immigration Services, three years and multiple Inspector General warnings after the first red flags were public. Neither administration has clean hands on this one, and RSN is not aware of any evidence that USCIS caught anything in 2025 that DHS’s Inspector General would go on to flag in this month’s audit.
Lakanwal’s case was not the only one Sen. Grassley cited when pressing DHS. In a January 7, 2026, letter to DHS, U.S. Citizenship and Immigration Services, and Customs and Border Protection, Grassley identified two additional Afghan evacuees arrested in the weeks after the shooting: Mohammad Dawood Alokozay, arrested November 25, 2025, and charged — in Grassley’s characterization of the case — with posting threats to build a bomb and carry out a suicide attack; and Jaan Shah Safi, arrested December 3, 2025, and charged with supporting ISIS-K and providing weapons to a relative described as an Afghan militia commander. Both had been admitted to the U.S. under Operation Allies Welcome in September 2021 and 2022, respectively. These are allegations contained in a sitting senator’s letter to the agencies responsible for vetting, not proven facts, and RSN presents them as such — but they are precisely the kind of documented, specific claims the DHS Inspector General’s audit was designed to test.
The Other Side of the Ledger
None of this should obscure an important fact: the overwhelming majority of the roughly 190,000 Afghans resettled since August 2021 are not accused of anything, and many are the interpreters, drivers, and support personnel who risked their lives working alongside American forces for two decades — people the United States made a moral commitment to when it withdrew. A 40 percent administrative-error rate in a sample of case files is not a 40 percent threat rate, and DHS’s own release does not claim otherwise. Missing an alias in a file is a serious lapse; it is not the same thing as evidence of terrorist intent, and conflating the two does a disservice to both the audit’s actual findings and the many Afghan allies whose service to this country is undisputed.
It is also worth stating plainly that evacuating well over 100,000 people in a matter of weeks, as the Afghan government collapsed and the Taliban closed in on Kabul, was never going to produce a clean paper trail. Every serious postmortem of Operation Allies Welcome — including the Inspector General’s own reports — has acknowledged the emergency conditions under which the initial screening took place. The legitimate criticism is not that the evacuation was chaotic in August 2021. It is that DHS had specific, written warnings from its own watchdog in 2022 and again in 2024 about exactly these vetting gaps, agreed to fix them, and — on the evidence of this month’s audit — still had not closed them by 2026.
What Happens Next
In the immediate aftermath of the November 2025 shooting, the Trump administration paused all asylum decisions nationwide pending a review of vetting protocols, and separately moved to review asylum cases approved under the Biden administration. Grassley’s committee has continued to press both DHS and the Justice Department for records. And this month’s report is not the last word: it is one recurrent-vetting review in what DHS’s own Inspector General has described as an ongoing body of work examining Operation Allies Welcome from multiple angles — the fragmented derogatory-information process flagged in 2024, the recurrent vetting sampled this month, and the DoD-side watchlist failures Roark testified to in January.
The pattern will be familiar to regular readers of this publication: a federal watchdog documents a specific, fixable failure in writing, the agency concurs with the recommendations on paper, and years pass before anything changes — the same dynamic RSN has traced recently in CMS’s year-long delay on documented Obamacare marketplace fraud and in HHS’s unresolved organ-acquisition reimbursement guidance. In each case, the failure was not a secret. It was sitting in an Inspector General’s report, waiting for someone in Washington to act on it before the cost came due.
What the public has not yet gotten is a straight answer to Grassley’s original, simple question from 2024: which of the Inspector General’s recommendations have actually been implemented, and which are still sitting on someone’s desk. Until that accounting exists, the honest conclusion is not that Operation Allies Welcome was uniquely evil or uniquely incompetent by the standards of a chaotic wartime evacuation. It is that the government promised 100 percent vetting, documented its own failure to deliver that promise in writing at least three times over four years, and is only now — after a young National Guard specialist was killed two blocks from the White House — facing sustained pressure to show its work.
