What we found
An official consumer advisory describes scammers pretending to be with the Department of Veterans Affairs and offering to help you apply for benefits if you pay them. The same advisory states it is free to apply with the VA. It also says scammers target people because applying for veterans benefits is at the top of the to-do list when getting ready to leave the military. If anyone asks you for money to file or speed up a VA benefits application, stop and apply yourself through the VA directly at no cost. Our harvested material does not say how these approaches reach people, what payment methods are requested, how much is asked for, or how many people have paid. The Department of Veterans Affairs is named here only as the body these impersonators claim to represent, and LIVEFRAUD ALERTS is not affiliated with it or with any agency named in this alert. This alert was written only from the harvested row listed below and contains nothing beyond it.
- The single supporting row is an official consumer advisory describing the tactic directly.
- Only one row was harvested, with no second source and no independent reporting.
- The row gives no figures on losses, victim counts, contact methods or payment channels.
- Reviewed by 3 models from independent houses.
THE RULE
Money handed over as a fee for something the VA lets you apply for free, with no benefit application necessarily filed in return.
What we don’t know
- How the impersonators make first contact with people leaving the military
- What payment methods or amounts are requested
- Whether any application is filed after payment
- How many people have been approached or have paid
- Whether particular locations or units are being targeted
The bench — who voted
3 INDEPENDENT AI MODELS REVIEWED THIS. ALL 3: HIGH RISK.
The card names a count. Here are the seats behind it, with what each one said.
GOOGLEgemini-flash-latest · HIGH
The finding asserts that the advisory states it is free to apply with the VA, and that scammers target individuals preparing to leave the military because applying for benefits is top of mind. Neither detail appears in Evidence Row 1.
GROQopenai/gpt-oss-120b · HIGH
The finding adds that the advisory states it is free to apply with the VA and that scammers target people because benefits are at the top of the to‑do list, but the sole evidence row only mentions scammers pretending to be VA and asking for payment; it does not contain those additional statements.
OPENROUTERopenrouter/free · HIGH
The evidence row does not state that the advisory says it is free to apply with the VA; that claim is not present in the source.
Reviewed by 3 independent models; all judged the finding to go beyond the evidence.
▼ Protocol & challenge record
Objection (high): Internal contradiction on targeting. targeting_dropped asserts "no harvested source describes this group" for service members leaving the military — but the harvested row says verbatim that applying for veterans benefits "is at the top of the to-do list when you're getting ready to leave the military," and the finding itself repeats that as a sourced claim sentence, and the share_directive tells readers to forward it to "anyone you know who is getting ready to leave the military." Either the group is sourced (in which case the drop rationale is factually false and should be removed) or it is not (in which case sentence 3 and the share directive are unsourced targeting). As drafted the packet does both at once.
Resolved: Unresolved by me — Desk must pick one. My recommendation: delete the targeting_dropped entry, since the row does describe people getting ready to leave the military; keep sentence 3 and the share directive as sourced.
Objection (high): risk_line contains a fact not in evidence. "with no benefit application necessarily filed in return" is pure inference — the row says nothing about whether an application is filed, and the packet's own unknowns list flags exactly this ("Whether any application is filed after payment"). Asserting the harm mechanism in the risk line while listing it as unknown is an overclaim by the alert's own standard.
Resolved: Proposed rewrite of risk_line: "Money paid as a fee to someone claiming to be from the VA, for an application the VA accepts free of charge." Drop the clause about no application being filed.
Objection (high): The advice sentence is over-broad and could cause real harm. "If anyone asks you for money to file or speed up a VA benefits application, stop" sweeps in VA-accredited attorneys and claims agents, who may lawfully charge fees in some circumstances (e.g., appeals), and it contradicts the plain fact that fee-charging claim consultants exist and are not necessarily VA impersonators. The row's warning is specifically about people *pretending to be with the VA*. The advice must be narrowed to that, or at minimum flagged that the harvested row does not address accredited representatives.
Resolved: Proposed rewrite of the advice sentence: "If someone claiming to be from the VA asks you for money to file your benefits application, stop — the VA does not charge to apply, and you can apply directly at no cost." Add limitation noting the harvested row does not address fees charged by VA-accredited attorneys or agents.
Objection (medium): Truncated source, and the truncation is at the point where the qualification would appear. The harvested text ends "get the help you need while avoiding scams. Here's how." — the FTC's actual operative guidance (almost certainly pointing to free VSOs and accredited representatives) was not harvested. Writing an imperative advice line from the half of the advisory that precedes "Here's how" risks stating advice the source itself qualifies immediately after the harvest boundary.
Not resolved — preserved on the record.
Objection (medium): watch_icons "bank" and "card" assert payment channels the packet explicitly says it does not know ("our harvested material does not say ... what payment methods are requested"). Icons are claims to a reader. Drop them or keep only a generic money/person icon.
Resolved: Recommend reducing watch_icons to ["person"] (impersonation) unless a payment channel is harvested.
Objection (medium): Claim-line stretch and present-tense framing. "People are asked to pay a fee" implies documented, ongoing solicitations. The row is a prospective consumer warning describing a tactic; it reports no incidents, dates, counts or losses. The claim also drops the defining element — impersonation of the VA — which is what makes the conduct a scam rather than a paid-consultant transaction.
Resolved: Proposed claim rewrite: "Scammers posing as the VA offer to help you apply for veterans benefits for a fee, though applying with the VA is free." Keeps the impersonation element and matches the source's prospective framing.
Objection (medium): Date currency needs an explicit check. The single source is stamped 2026-07-27 and the URL path is /2026/07/. If that date is in the future relative to publication of this alert, the row is not verifiable and the whole packet fails; if it is recent past, say so. A one-source packet resting entirely on a date that looks forward-dated cannot be waved through without confirming the harvest date against today.
Not resolved — preserved on the record.
Objection (low): Attribution is vague and the disclaimer misfires. The finding says "An official consumer advisory" three times but never names the FTC, so a reader cannot judge the issuer or find it. Meanwhile the disclaimer disclaims affiliation with "any agency named in this alert" — the only agency named is the VA, the impersonated body, not the issuer. Name the FTC as the source.
Resolved: Name the FTC as the issuing body in the first finding sentence and extend the non-affiliation disclaimer to cover the FTC as well as the VA.
Objection (low): Scope unstated. An FTC alert and VA benefits are US-specific; nothing in the finding limits the geography, so non-US readers may take the "free to apply" and "apply through the VA directly" instructions as applying to them.
Not resolved — preserved on the record.
Preserved dissent
ON THE RECORDThe targeting_dropped line is simply wrong on the facts of the harvest. The row states, in its first clause, that scammers know applying for veterans benefits is at the top of the to-do list "when you're getting ready to leave the military." That is a sourced description of who is being targeted. Recording a rule-based drop for a group the source names, while simultaneously naming that group twice elsewhere in the same packet, is not caution — it is incoherence, and it undermines the credibility of the drop rule everywhere else it is applied.
ON THE RECORDI do not accept the risk_line as written. "With no benefit application necessarily filed in return" is the packet inventing the loss mechanism, and the packet's own unknowns list concedes it does not know this. If a line appears in unknowns it cannot appear as an assertion in the risk line.
ON THE RECORDThe advice sentence as drafted is the most likely thing in this alert to do harm. Told flatly that anyone asking for money to file or speed up a VA claim should be stopped, a reader could walk away from a lawfully accredited attorney or agent. The source warns about impersonators of the VA; the alert should say that and only that. I would hold the alert until this sentence is narrowed.
ON THE RECORDOn sufficiency generally: one official FTC advisory is adequate to support the narrow proposition that this tactic is being warned about, and I would not block on single-sourcing. What I would block on is the packet claiming more than the tactic — ongoing solicitations, unfiled applications, payment channels implied by the bank and card icons. Strip those and the remainder stands.
The sources
Official sourceVeterans: Don’t pay to apply for VA benefits2026-07-27
An official consumer advisory describes scammers pretending to be with the Department of Veterans Affairs and offering to help you apply for benefits if you pay them.
Authority: official. Retrieved 2026-08-15.
Limitation: The row is a short advisory opening; it does not describe contact channels, payment methods, amounts, dates of incidents or numbers of people affected.
Open the original source →Other checks
Approved by ihubglobalhq on 2026-08-16, after the six-point evidence checklist.
Something wrong here? Tell us and we'll correct it — corrections are published, not quietly edited.
Phishy? Send it → sharelivefraud.com/squire-it
Not affiliated with any government agency, credit bureau, bank, platform, or law-enforcement agency. Informational only — not legal or financial advice.
Naming a source is not an endorsement, and being named here is not an accusation against any company.
Powered by SquireIt™
Verify this receipt at squireit.com