Explainer · Fact Sheet
What the DHS Inspector General actually found in Afghan asylum files
A September 23, 2026 DHS Office of Inspector General report (OIG-26-26) reviewed 678 asylum files from Afghans who arrived under Operation Allies Welcome. It found paperwork and process errors in 40 percent of them. It did not identify any case in which derogatory information was missed, and it found that faster processing did not increase errors.
Last updated September 29, 2026
The evaluation was launched to answer one question: did the faster asylum timelines required by the September 2023 Ahmed v. DHS settlement cause asylum officers to make more mistakes? The answer was no. Errors per file were 0.718 before the settlement and 0.716 after it, and the share of files with any error fell from 41 percent to 38 percent. The "40 percent" headline comes from a secondary finding that counts every missed name variant, unsigned form, and unprinted copy across four categories.DHS OIG, "Adjudication Errors in Operation Allies Welcome Asylum Files Risked Missing Derogatory Information," OIG-26-26, September 23, 2026 (redacted public version).
Headline versus report
Tap a claim to see what the report says
Each claim below is circulating in coverage of the report. The response quotes or summarizes the report itself, with the page where you can check it.
Claim"40 percent of Afghan evacuees had vetting errors."+
Forty percent of the 678 sampled files had at least one process or paperwork error in one of four categories. Every applicant was fingerprinted and interviewed, and 97 percent of files showed every required security check completed. OIG "did not assess the actual effect of any error we identified, including whether an error would have impacted an applicant's asylum eligibility."OIG-26-26, pages 5 to 7
Claim"The watchdog found terrorism-related cases."+
Two errors involved the terrorism-related inadmissibility (TRIG) process. In one, the officer analyzed the issue in writing but the separate exemption worksheet was not in the file. In the other, the supervisor and section chief both signed the worksheet but neither marked the approve or disapprove box. In both, the officer had identified and elevated the issue.OIG-26-26, page 15
Claim"Security checks were skipped."+
Missing check results appeared in 23 files, or 3 percent. Nine were applications USCIS had administratively closed, so no asylum was granted. One was a software issue USCIS identified during fieldwork and has since corrected. The other 13 were FBI Name Check reruns for aliases or expired results, a requirement now satisfied by National Vetting Center checks.OIG-26-26, pages 9 to 10 and footnote 13
Claim"Rushed processing under the court settlement caused the errors."+
This was the question OIG opened the evaluation to answer. It "did not observe a statistically significant increase in error rate after the September 2023 DHS settlement agreement." Errors per file were 0.718 before and 0.716 after. Files with any error fell from 41 percent to 38 percent.OIG-26-26, pages 17 to 19 and Appendix E
Claim"Derogatory information was missed."+
The report's title says errors "risked" missing derogatory information. It identifies no case in which any was missed, and the report does not indicate that OIG re-ran the missed name variants. USCIS told OIG that its own quality assurance reviews "unequivocally determined that any missed aliases did not impact the adjudicative outcome of any case" in its samples for the study period, and OIG did not dispute that in its analysis.OIG-26-26, page 7 and Appendix C, page 3 of the USCIS response
Background
Why the Inspector General looked at these files
Roughly 97,000 people were evacuated from Afghanistan in 2021 and resettled in the United States under Operation Allies Welcome (OAW), most of them on humanitarian parole. Many later applied for asylum with USCIS, which involves an in-person interview, name-based checks against law enforcement databases, and fingerprint-based checks against FBI, DHS, and Department of Defense biometric systems.
The Afghanistan Supplemental Appropriations Act of 2022 required USCIS to interview OAW asylum applicants within 45 days and decide their cases within 150 days. In April 2023, applicants sued DHS (Ahmed v. DHS, N.D. Cal.) alleging the agency was missing those deadlines. On September 6, 2023, DHS settled and agreed to decide 90 percent of OAW asylum applications within 150 days.
USCIS employees then complained to the Inspector General that the settlement timelines would weaken supervisory review. OIG opened this evaluation to test that claim. Its stated objective was "to determine whether missed aliases or incomplete resolution of potential matches to derogatory records increased following the Ahmed v. DHS settlement agreement." Fieldwork ran from March 2024 to January 2025 and covered files decided before February 6, 2024. Publication was delayed by three government shutdowns in fiscal year 2026.
This is the third DHS OIG product on OAW screening. A September 2022 report (OIG-22-64) examined the initial evacuation-stage vetting, and a May 2024 report (OIG-24-24) described a fragmented process for resolving derogatory information on OAW parolees. This report looks only at the asylum adjudication stage inside the United States.
Method
What OIG reviewed, and what it did not
OIG drew a random sample of 678 files from 13,682 OAW asylum applications with a final decision, limited to principal applicants between 14 and 79 years old. For each file, it checked whether the asylum officer completed four sets of steps. Any miss in any category counted as an "error."
Aliases
Did the officer enter every name variant found anywhere in the file into the case management system so it could be run through name-based checks (TECS, NCIC, and the FBI Name Check)?
Security checks
Did the file show results for every required check, including reruns for aliases and expired results?
Resolution
When a check returned a potential match, did the officer document each required resolution step and place the paperwork in the file?
Documentation
Was the multiple systems checks document filled out correctly and signed, and were copies of check results and other required documents in the file?
OIG states that its review "did not include reassessing AO's decisions," that it "did not assess the actual effect of any error we identified." Documentation errors were recorded by a single OIG reviewer without a second OIG review or USCIS validation. USCIS reviewed the other categories and "did not always agree with our categorization of errors."
Finding 1
The four error categories, and what is inside each one
OIG found at least one error in 269 of 678 files (40 percent). The table below reports each category as the report describes it, with the report's own caveats.
- What it means
- A name variant appeared somewhere in the file but was not separately entered for name-based database searches. OIG cites "not identifying a spelling variation of a name" as a typical cause.
- Context from the report
- One file contained 124 aliases; the median was two. OIG concluded that officers "may be unable to entirely avoid missed aliases" and excluded another 26 files and 51 names where USCIS policy allowed officer judgment.
- What it does not affect
- Fingerprint-based checks (FBI fingerprint, DHS IDENT, and Department of Defense ABIS) match on biometrics, not names. Missed aliases affect only the name-based searches.
- What it means
- The file did not show a result for a required FBI Name Check or TECS check.
- Context from the report
- Nine of the 23 were applications USCIS had administratively closed (withdrawn, abandoned, or moved to immigration court), so no asylum was granted. One was a software issue USCIS identified during fieldwork and has since corrected. The other 13 were FBI Name Check reruns for aliases or expired results. Since December 9, 2025, National Vetting Center checks satisfy the FBI Name Check requirement.
- What it means
- A check returned a potential match and the file did not contain every required resolution step.
- Context from the report
- 23 of the 45 involved Department of Defense ABIS matches, and all 23 were because the officer did not print the ABIS details into the file. OIG notes that CBP's National Targeting Center resolves those matches substantively, that OIG counted "technical" rather than "substantive" resolution, and that "many ABIS records were non-derogatory." 13 involved completion of the NCIC potential-match form. Four involved FBI fingerprint records and five involved TECS records, including two where the underlying record was not derogatory.
- What it means
- A form was incomplete or a required copy was missing from the file.
- Context from the report
- 86 of the 115 errors were on the multiple systems checks document: 49 involved a mismarked field, 26 involved the supervisor's action on the form, and 11 involved another form issue. 29 were missing paperwork, mostly copies of check results when a check had been run more than once. Two files were missing the written assessment or interview notes. OIG writes that missing printouts "might be inconsequential beyond inconvenience because the information is generally available in another location," while a missing assessment is more consequential.
Bottom line The 40 percent figure is the share of files with at least one miss in any of these four categories. The report does not say how many of those misses, if any, concealed derogatory information, because OIG did not assess that question.
Inside the numbers
Anatomy of the 486 errors
OIG counted 486 individual errors across 678 files. Tap a segment to see what each category is made of. Segments in red mark the items OIG itself describes as more consequential.
Items OIG describes as more consequential (a missing assessment or interview notes, an undocumented TRIG approval, an unresolved substantive record, or a check not run on an open application).
The "terrorism-related" line
What the two TRIG errors were
The report's summary says the errors included "two cases related to potential inadmissibility on terrorism-related grounds." Page 15 of the report describes both.
In the first, the asylum officer identified a possible terrorism-related inadmissibility ground (TRIG), analyzed it in the written assessment, and recommended an exemption, but the separate TRIG exemption worksheet was not in the file, so the required second-level approval was not documented. In the second, the TRIG worksheet was completed and signed by both the supervisory asylum officer and the section chief, but neither marked the box indicating whether they approved or disapproved the officer's analysis.
In both cases the officer identified the issue and elevated it. The errors OIG recorded were a missing worksheet and an unmarked approval box. TRIG exemptions exist in statute for situations such as an applicant who acted under duress, and USCIS routinely evaluates them for Afghans who were forced to interact with the Taliban or other groups.
Finding 2
Faster processing did not increase errors
This was the question the evaluation was opened to answer. OIG compared files decided before and after the September 2023 settlement and found "no statistically significant increase in error rate."
Average errors per file, before and after the settlement. Files with any error fell from 41 percent to 38 percent. Files with a missed alias fell from 27 percent to 22 percent. Files with a missing security check fell from 5 percent to 2 percent. Resolution errors (4 to 9 percent) and documentation errors (13 to 15 percent) were higher in the post-settlement sample, but OIG found neither difference statistically significant under its method, and the resolution category is dominated by the ABIS printout issue described above.
Before and after the settlement
Error rates by category, pre and post Ahmed v. DHS
OIG's own figures for files decided before September 7, 2023 and files decided between then and February 6, 2024. Switch between the two ways OIG measured it.
Source: OIG-26-26, Figures 2 and 3. OIG found none of the post-settlement increases (resolution and documentation) statistically significant at its 95 percent family-wise confidence level; see Appendix E of the report for the p-values.
Agency response
What USCIS said, and what it has changed
USCIS Director Joseph Edlow responded on July 22, 2026. The agency concurred with all five recommendations, which OIG lists as "open and resolved" with completion dates between December 2026 and April 2027. The response also disputed the report's framing.
USCIS wrote that the report "does not contain sufficient context to conclude that the errors were outcome determinative or indicative of derogatory information that asylum officers had not identified," and that the errors "take place amid broader, established, and robust screening, vetting, and security checks safeguards already in place that minimize the impact of these errors." It stated that its own quality assurance reviews re-ran checks on missed aliases and "unequivocally determined that any missed aliases did not impact the adjudicative outcome of any case included in USCIS' quality assurance samples during the timeframe of OIG's study." OIG did not dispute that statement in its analysis.
According to the response and the report, since the period OIG reviewed USCIS has added National Vetting Center checks for asylum applicants (March 1, 2024), begun continuous vetting against the DHS Watchlist Service (May 1, 2025), issued formal written guidance on alias identification (December 11, 2025), required alias training for all personnel who conduct security checks (by July 15, 2026), added supervisory review of administrative closures (May 5, 2026), and corrected the TECS software issue.
The five recommendations ask the Asylum Division to set an error-rate benchmark, improve alias identification, add controls for administratively closed cases, study an electronic version of the checks document, and document how officers consider social media review results. None of them calls for reopening any adjudicated case.
Set expectations
What this report does not do.
- It does not identify any Afghan ally who was granted asylum despite derogatory information. OIG states it did not assess the effect of any error, and the report does not indicate the missed names were re-run.
- It does not find that any asylum decision was wrong. OIG states it did not reassess asylum officers' decisions.
- It does not find that the faster timelines under the Ahmed v. DHS settlement reduced the quality of review. It found the opposite.
- It does not evaluate the screening that happened before arrival, including biometric and biographic checks at overseas transit sites and U.S. ports of entry.
- It does not describe the current process. The files reviewed were decided before February 6, 2024, and USCIS has changed alias guidance, training, and the checks themselves since then.
- It does not change anyone's immigration status, reopen any case, or recommend that any case be reopened. Its recommendations are procedural.
- It does not apply to Special Immigrant Visa holders, refugees admitted through the U.S. Refugee Admissions Program, or Afghans still abroad. Its scope is asylum applications filed by OAW parolees inside the United States.
Frequently asked questions
Common misreadings, answered
Did the Inspector General find that 40 percent of Afghan evacuees were not vetted?
No. Forty percent of sampled files had at least one process or paperwork error in one of four categories. Every applicant in the sample was fingerprinted and interviewed, and 97 percent of files showed every required security check completed.
Did the report find that people with terrorism ties were granted asylum?
No. The two "terrorism-related" errors were a missing exemption worksheet and an unmarked approval box on cases where the officer had already identified the issue, analyzed it in writing, and sent it up the chain.
Were security checks skipped on 40 percent of applicants?
No. Missing check results appeared in 23 files, or 3 percent. Nine of those were applications that had been closed without a grant of asylum, one was a software issue that has since been fixed, and the remainder were FBI Name Check reruns that a newer check now covers.
Does a missed alias mean the person was never checked?
No. It means one spelling or variant of a name was not separately run through the name-based databases. The applicant's primary name and fingerprints were still checked. Fingerprint matches do not depend on how a name is spelled.
Did the court settlement that sped up processing cause these errors?
No. That was the question OIG set out to answer, and it found no statistically significant increase in errors after the settlement. Total errors per file were essentially identical, and the share of files with any error went down.
Did OIG check whether the missed names would have turned up anything?
No. The report does not indicate that OIG ran the 303 missed name variants through the databases, and OIG states that it did not assess the effect of any error. USCIS says its own quality assurance reviews did re-run such checks and found no change in outcome in its samples.
Will Afghans who were granted asylum have their cases reopened because of this report?
The report does not recommend that, and USCIS did not propose it in its response. All five recommendations concern USCIS internal procedures. Anyone with questions about their own case should consult a qualified immigration attorney.
Is this the same report as the 2022 one about vetting during the evacuation?
No. The September 2022 report (OIG-22-64) covered screening during the 2021 airlift. This report covers asylum adjudications by USCIS inside the United States, mostly in 2022 and 2023.
AfghanEvac statement
Our statement on the report
Statement of Shawn VanDiver, President of AfghanEvac
Regarding the DHS Inspector General report on Operation Allies Welcome asylum files (OIG-26-26), September 23, 2026
We read the report. Anyone quoting it should do the same.
The Inspector General opened this evaluation to answer one question: whether faster asylum processing for Afghan allies led to more mistakes. It did not. Error rates were essentially identical before and after the 2023 court settlement, and the share of files with any error went down. That is the finding the report was built to deliver, and it is the one nobody is quoting.
The 40 percent figure now circulating is a paperwork compliance rate, not a vetting failure rate. It counts a missing signature, an unmarked box, an unprinted copy of a check that was already run, and a spelling variant of a name that was not separately searched. The report identifies no case in which an Afghan ally was granted asylum despite derogatory information. The Inspector General says plainly that it did not assess whether any error affected anyone's eligibility. The two errors described as terrorism-related were a missing worksheet and an unchecked approval box on cases the officers had already identified, analyzed, and sent up the chain.
Every one of these applicants was fingerprinted and run against FBI, DHS, and Department of Defense biometric systems that do not care how a name is spelled. USCIS, under this administration, told the Inspector General that its own reviews found no missed alias changed the outcome of any case in its samples. USCIS has accepted all five recommendations, and it should. Good records matter.
But a paperwork gap in a 2023 file is not evidence that the people who fought beside American troops are dangerous. It is evidence that a system handling thousands of cases on a court-ordered clock needed better forms, and it now has them.
This report undercuts everything this administration and some in Congress have been saying about the vetting of Afghan allies. It proves our point. These are among the most thoroughly screened people ever to arrive in this country, and anyone who uses this document to argue otherwise has not read it.
AfghanEvac will keep doing what we have always done: read the record, tell the truth about it, and insist that the United States keep its word.
Go deeper
Read the primary sources
The redacted report, including USCIS's full management response, is public on the DHS OIG website. Our earlier explainer on how Afghan allies are vetted covers the checks that happen before and after arrival.
Related explainers
This explainer summarizes a public government report and is provided for general information. Quotations and figures are drawn from the redacted public version of DHS OIG-26-26 and USCIS's July 22, 2026 management response. AfghanEvac is not a law firm and does not give legal advice. Talk to a qualified immigration attorney about your own case.