Two 2025–2026 patterns worth separating: source-of-funds defects with unreconciled deposits are drawing straight-to-denial outcomes instead of RFEs, and loan-based investment capital is drawing a personal-liability test under 8 C.F.R. § 204.6(e). Neither one means EB-5 approvals broadly are falling.
Two things are true about EB-5 adjudications right now, and they are easy to collapse into one another if you are not careful. The first: on a specific category of evidentiary defect — source-of-funds documentation with deposits an officer cannot trace — United States Citizenship and Immigration Services is increasingly willing to deny an I-526E petition outright rather than issue a Request for Evidence first. The second, related but distinct: where an investor's capital is loan-based, officers are demanding proof the investor personally owes the debt, not merely that the investor has guaranteed someone else's obligation to repay it. Both patterns are real, both are documented by more than one independent source, and neither one means EB-5 approvals broadly are falling. Reading them as a single "EB-5 is getting harder" headline is exactly the mistake worth avoiding — and exactly the kind of category error a federal-litigation practice is trained to catch, because a reviewing court makes the same distinction: a decision stands or falls on its own record, not on a narrative about the program.
Both patterns surface in the same document type — the I-526E and the Requests for Evidence or denials that follow it — which is part of why they get blurred together. They are nonetheless answers to different questions.
Where a source-of-funds filing contains deposits an officer cannot reconcile against the investor's documented income or asset history, current practice increasingly skips the RFE step that would once have given the petitioner a chance to explain. Green Card Fund, reviewing this pattern directly, approval rates remain near 97 percent overall since the Reform and Integrity Act — this is a source-of-funds-specific tightening, not a broad EB-5-is-being-denied-more trend.
Separately, where the invested capital originates from a loan rather than savings or a direct gift, officers are reading the loan documents for a specific fact: who actually owes the money. Fragomen, Del Rey, Bernsen & Loewy, in reporting published May 2026 on I-526E Requests for Evidence, describes officers now expecting evidence a loan is “personally owed by the investor rather than guaranteed by the new commercial enterprise.” A loan the new commercial enterprise owes, with the investor merely standing behind it, is being read as a defect in the capital itself — not a documentation gap around otherwise-sound capital.
Source-of-funds practice has always required a lawful-source showing. What appears to have moved is what counts as showing it. Von Trapp Law, reviewing recent Administrative Appeals Office I-526 decisions, identifies unexplained deposits as a leading denial driver and describes the operative standard as one where “each material inflow can be independently verified.” That is a materially higher bar than a plausible narrative supported by some documentation. An explanation an adjudicator finds credible is no longer sufficient on its own if the underlying deposit cannot be traced to a source independent of the petitioner's own say-so — bank records, transaction chains, and third-party documentation have to close the loop themselves.
An RFE has traditionally functioned as the agency's invitation to cure a gap: here is what is missing, supply it. What Green Card Fund's review of straight-to-denial outcomes suggests is that officers increasingly treat an unreconciled deposit not as a gap to be filled on request but as a defect already fatal on the record as filed — the equivalent of concluding there is nothing a supplemental document could add that would make an untraceable inflow traceable after the fact. Whether that inference is always correct is a fair question, and it is precisely the kind of question a denial's own reasoning has to answer for a reviewing court under the arbitrary-and-capricious standard (5 U.S.C. § 706(2)(A)) if the case ever gets there. The practical lesson for anyone filing sits upstream of that question, though: treat the initial filing as the only chance the deposit gets to explain itself.
This is not an EB-5-only phenomenon, either. A parallel posture — discretion to deny on the initial record without first issuing an RFE or NOID — is now stated USCIS policy across a broader set of filings under USCIS Policy Alert PA-2026-05 (August 5, 2026), which POLICY GUIDANCE, not a regulation — no regulation changed. The EB-5 source-of-funds pattern described here reads as one visible instance of that same broader shift in how the agency treats an incomplete record, not an EB-5-specific invention. The firm's federal-court guides cover that policy shift directly for the filings it reaches most often.
The requirement is not new; the scrutiny applied to it is what has sharpened. EB-5 capital must be the petitioner's own qualifying investment under 8 C.F.R. § 204.6(e), and where that capital traces to borrowed funds, the regulation has long required the debt to be secured by the investor's own assets rather than the assets of the new commercial enterprise the investment funds. Recent adjudications appear to be reading that requirement more literally — not just asking whether the enterprise's assets are pledged, but asking whose name is actually on the obligation to repay.
Green Card Fund frames the operative test plainly: the investor must be “personally and primarily liable for the debt,” not a secondary or contingent party to someone else's debt. A guarantee is, by its legal nature, a promise to pay if someone else does not — and a filing that documents a guarantee, without more, now reads to an adjudicator as documenting the wrong obligor. The evidentiary gap is specific and fixable at the drafting stage: loan instruments, collateral documents, and the investor's own financial statements have to line up on one consistent story about who owes the money first, not who stands behind it second.
| Evidence pattern | How it now tends to read |
|---|---|
| Investor is a guarantor on a loan to the new commercial enterprise | Treated as capital the enterprise owes, not the investor — a likely denial ground on the current pattern |
| Loan is issued directly to the investor personally, secured by the investor's own assets, proceeds then invested | Consistent with the personal-and-primary-liability reading described above |
| Loan documents, collateral filings, and the investor's financial statements are internally inconsistent about the obligor | A documentation defect independent of which structure was actually used — and one of the more avoidable ones |
Read generically, not as advice on a pending filing: the distinction above is a description of a public reporting and regulatory pattern, not an assessment of any specific loan structure, investor, or petition. No case, client, or filing is described or identified anywhere on this page.
It would be easy to read the two signals above and conclude EB-5 as a program is simply getting denied more. Green Card Fund's own reporting cuts against that reading directly: since the Reform and Integrity Act, approval rates remain near 97 percent overall since the Reform and Integrity Act — this is a source-of-funds-specific tightening, not a broad EB-5-is-being-denied-more trend. A documentation-specific tightening inside one evidentiary category is a narrower and more precise claim than a program-wide reversal, and the distinction matters — for how a petitioner prepares a filing, and for how anyone commenting on the trend should describe it.
The honest description of what is happening is narrower than either extreme: source-of-funds evidence involving deposits or loans is being tested against a stricter verification standard, and that stricter standard is producing more denials and fewer RFEs specifically where that evidence falls short. It is not evidence that the program's overall posture toward otherwise well-documented petitions has changed. Commentary that blurs the two — whether to alarm or to reassure — is describing a narrative, not the adjudication record.
Reading an agency's stated reasoning against what its own record actually supports is the core discipline of federal-court review under the Administrative Procedure Act — asking whether a decision was arbitrary, capricious, or otherwise not in accordance with law (5 U.S.C. § 706(2)(A)), a question a court answers from the administrative record, not from a program's reputation. That discipline does not belong to any one agency or program. The pattern described on this page happens to be visible in EB-5 filings right now because the reporting on it is unusually good; the same analytical approach applies wherever an agency's evidentiary standard moves faster than its own public description of that standard. Watching for that gap — between what an agency says it requires and what its actual outcomes reveal it is requiring — is the underlying competency this page is an example of, not a signal that the underlying practice area has changed.
Three lessons here travel well past this one program:
No. This is general commentary on an adjudication pattern, written by a federal-litigation practice, illustrating how federal agencies test the evidence in front of them. It is not a description of a new client-facing service, and no petition, petitioner, or investor is referenced anywhere on this page.
Generically: any federal agency decision can be tested against whether it was arbitrary, capricious, or otherwise not in accordance with law based on the agency's own record — that is a federal-court question, not an immigration-agency one, and it is squarely the kind of review this firm's federal-litigation practice performs. Nothing here evaluates any specific denial; that requires reading the actual record.
No — that is the distinction this page makes explicitly. Reporting on the straight-to-denial pattern also notes overall approval rates have remained near 97 percent since the Reform and Integrity Act. The tightening is specific to how source-of-funds and loan-based capital evidence is read, not a broader shift in the program's outcomes.
That the investor, individually, owes the loan and is on the hook to repay it first — as opposed to guaranteeing that someone else (commonly the new commercial enterprise itself) will repay it. Current reporting describes adjudicators reading the second structure as a defect in the capital, not merely a documentation gap.
