USCIS’s Vetting Center: What EB-1A Filers Are Reporting Since the AI Screening Rollout

If your EB-1A petition has felt slower or more scrutinized than it would have a year ago, you are not imagining it. Since USCIS stood up its new Vetting Center and began layering AI-assisted tools into case review, immigration attorneys and filers have been comparing notes on what has actually changed. Some of it is confirmed policy. Some of it is pattern recognition from attorneys watching RFEs pile up. Here is what is verified, what is anecdotal, and what it means if you have an EB-1A case in the pipeline right now.

Check If Your EB-1A Case Is AI-Screening Ready 

What Is the USCIS Vetting Center?

The USCIS Vetting Center became operational in late 2025, based in Atlanta, Georgia, as a centralized hub for security and fraud screening across immigration benefit categories. Rather than each field office running its own version of background checks, the center uses AI-powered tools to screen names, travel patterns, watchlist data, and derogatory information against a combined set of databases, including DHS systems such as HART, IDENT, TECS, and ATS, along with State Department visa records and social media monitoring.

The Vetting Center’s scope is broad. It touches H-1B and L-1 petitions, family-based filings, humanitarian benefits, and employment-based green cards, which include EB-1A. Cases connected to countries designated under Executive Order 14161 receive prioritized, deeper examination. USCIS has framed the goal as catching fraud, criminal activity, and security risks earlier in the process and standardizing how vetting happens across offices that previously handled it inconsistently.

It is worth separating this from a second, more EB-1A-specific tool making the rounds in attorney discussions: USCIS’s ELIS Evidence Classifier, a machine learning system that automatically tags and sorts submitted evidence before an officer ever opens the file. The Vetting Center is about security and fraud screening. The evidence classifier is about how your documents get organized once your petition lands on an officer’s desk. Both fall under the umbrella filers are now calling “AI screening,” and both appear to be shaping outcomes.

Read More: How to Respond to an EB1A RFE in 2026: The Complete Strategy Guide for Approval 

Has It Changed Processing Times or RFEs?

Processing time impact depends on whether a case gets flagged. Files that trigger a closer look through the Vetting Center’s screening reportedly see delays in the range of 30 to 45 days beyond a normal timeline, which can matter for anyone timing a job start date or travel plans around their case.

On the RFE side, attorneys tracking EB-1A cases describe a real shift in what officers are asking for. Requests increasingly probe the specific selection criteria a judging panel used, rather than accepting a judging role at face value. Recommendation letters from personal acquaintances are being discounted more often in favor of independent, arm’s-length evaluators. Achievements older than five years are sometimes dismissed as not current, even when the underlying accomplishment is still relevant. And simply holding a patent or a publication is treated as insufficient on its own without a clearer showing of original contribution.

Approval rate figures are harder to pin down precisely, since USCIS does not publish real-time EB-1A approval statistics. Private case-tracking platforms and attorney-run data sets suggest a meaningful softening over the past year, with some estimates showing a drop from roughly the low-to-mid 70% range in early 2025 down into the 40 to 55% range by mid-2026. These numbers vary by source and methodology, so they should be read as a directional signal that scrutiny has increased, not as an official USCIS statistic.

What Are Filers Reporting?

The most consistent complaints center on the evidence classifier misreading submissions. Attorneys describe documents that were clearly filed under the correct criterion appearing to USCIS as missing or misclassified, which then generates an RFE asking for evidence that was already in the record. This is frustrating for filers who assembled a complete package, only to be asked again for something that was already there.

A second pattern involves cross-document discrepancy flags. The system appears to catch minor inconsistencies, a name spelled slightly differently across exhibits, an employer referred to by full name in one letter and by abbreviation in another, that historically never drew attention from a human reviewer. Filers are also reporting issues tied to scan quality. Poorly OCR’d PDFs can carry a hidden text layer that conflicts with what a person sees on the page, and the AI tools reportedly read that hidden layer literally, creating discrepancies that do not actually exist in the visible document.

Some attorneys have also flagged RFEs that read as templated or boilerplate, with language and even factual errors that suggest AI-assisted drafting on the officer’s side, not just AI-assisted review of the petitioner’s evidence. None of this means the technology is universally unreliable, but it does mean presentation quality now matters in ways it did not before.

See What’s Triggering RFEs in 2026 

Does It Affect Every Case Equally?

No. The Vetting Center explicitly prioritizes cases connected to countries designated under Executive Order 14161, so nationality and country ties do create unequal scrutiny at the security screening stage. Beyond that formal prioritization, the pattern attorneys describe suggests that petitions with clean, well-organized documentation, consistent naming across exhibits, high-quality scans, and substantive rather than templated recommendation letters tend to move through with fewer AI-triggered snags. Petitions built from generic templates, inconsistent formatting, or low-quality scans appear disproportionately likely to draw an RFE, even when the underlying case for extraordinary ability is strong. In other words, the vetting rollout has not changed the legal standard for EB-1A, but it has raised the cost of a sloppy filing.

FAQs

Book a Consultation

An EB-1A petition that would have sailed through review two years ago can now snag on a scanning error, a name spelled two different ways across your exhibits, or a recommendation letter that reads as boilerplate to an AI-assisted reviewer. None of that reflects whether your underlying case for extraordinary ability is strong. It reflects whether your filing is built to survive a review process that has genuinely changed.

At EB1A Experts, we prepare petitions with this new layer of AI-assisted review in mind, from document consistency to letter quality to how evidence is organized against each criterion. Schedule your free consultation today and find out whether your case is ready for how USCIS is actually reviewing files in 2026, not how it reviewed them a year ago.