What we found
The Federal Trade Commission says operators of a student loan debt forgiveness scheme took more than $45.9 million from consumers as part of an illegal debt relief operation. In a November 2024 suit, the FTC alleged the operation pretended to be affiliated with the U.S. Department of Education and falsely promised student loan forgiveness to borrowers. At the FTC's request, a federal court temporarily halted the scheme and froze its assets, and in early 2025 the agency added several other companies and two more operators as defendants. In September 2025 the U.S. District Court for the District of Nevada entered orders banning two of the operators from the debt relief industry, with one also banned from telemarketing and the other prohibited from violating the Telemarketing Sales Rule. A proposed order against a further operator would ban her from debt relief services and telemarketing and imposes a monetary judgment of more than $45.9 million that will be partly suspended because of an inability to pay, with the full amount due if she is found to have materially misrepresented her finances. The FTC states that consumers can get help with their student loans for free at StudentAid.gov. The release does not say how many borrowers paid, how they were asked to send money, or whether any of it will be returned to them. It also does not quote the sales scripts, phone numbers or websites used, so our harvested material cannot tell you what one of these calls sounded like or which company names to look out for today. Before you pay anyone a fee for student loan forgiveness, hang up or close the page and check your loan yourself at StudentAid.gov, where help costs nothing. LIVEFRAUD ALERTS is an independent consumer awareness desk and is not affiliated with the FTC or the U.S. Department of Education; this alert was written only from the single official release listed above.
- The sole source is an official FTC press release naming the court, the district and the specific orders.
- The dollar figure, dates and the impersonation allegation are stated directly in that release rather than inferred.
- Confidence covers what the agency alleges and what courts ordered; the release does not describe payment methods, victim counts or current copycat activity, so the operational detail of the scam remains thin.
- Reviewed by 4 models, 3 from independent houses.
THE RULE
Callers claiming a link to the U.S. Department of Education and promising loan forgiveness charged borrowers for help that is free.
What we don’t know
- How many borrowers paid and how much each lost.
- How consumers were asked to pay — card, bank transfer or otherwise — is not stated.
- Whether any money will be returned to borrowers.
- What names, scripts or numbers the callers used, and whether similar calls continue now.
- Whether the proposed order has been signed by the District Court Judge.
The bench — who voted
4 INDEPENDENT AI MODELS REVIEWED THIS. ALL 4: HIGH RISK.
The card names a count. Here are the seats behind it, with what each one said.
ANTHROPICclaude-opus-5 · HIGH
The single sourced row is a one-sentence summary. It supports only: FTC allegation of Department of Education impersonation, false forgiveness promises, more than $45.9 million taken, and resulting court orders banning operators from debt relief and telemarketing. Nearly all of the finding's procedural detail — a November 2024 filing date, a temporary halt and asset freeze, additional companies and two more operators added in early 2025, September 2025 orders from the U.S. District Court for the District of Nevada, one operator banned from telemarketing and another enjoined from violating the Telemarketing Sales Rule — appears nowhere in the evidence text and cannot be checked against it.
DISSENT, PRESERVEDThe core claim line — impersonation of the Department of Education, false forgiveness promises, more than $45.9 million taken, and bans from debt relief and telemarketing — is fully carried by the row. My objection is to the elaborated procedural narrative wrapped around it, not to the substance of the alert.
DISSENT, PRESERVEDI grade the underlying pattern high regardless of the sourcing problems. Advance-fee student loan forgiveness pitches that claim federal affiliation target people already under financial strain, routinely harvest FSA ID credentials and bank details, and the $45.9 million figure with a suspended judgment indicates victims are unlikely to be made whole. The closing advice to hang up and check StudentAid.gov directly is the right instruction and should survive any redraft.
DISSENT, PRESERVEDThe alert is candid that it cannot supply company names, scripts or phone numbers. That honesty is creditable, but it means the piece cannot help a reader recognise a live call, which limits its protective value to the generic rule about never paying for forgiveness.
GOOGLEgemini-flash-latest · HIGH
The finding includes extensive specific details not present in the provided evidence row, such as procedural dates (November 2024, early 2025, September 2025), the specific legal remedies assigned to individual unnamed defendants, details regarding a partially suspended $45.9 million judgment based on financial inability to pay, and the reference to StudentAid.gov.
GROQopenai/gpt-oss-120b · HIGH
The finding adds multiple specifics (dates, additional defendants, details of proposed orders, partial suspension of judgment) not present in the single evidence row, which only mentions the alleged deception, the $45.9 million loss, and court bans.
OPENROUTERopenrouter/free · HIGH
The evidence does not specify how consumers were asked to pay, the names of the companies involved, or the exact court-signed orders, though these omissions do not negate the core finding.
Reviewed by 4 independent models: 1 found it carried by the evidence, 3 did not.
▼ Protocol & challenge record
Objection (high): Claim overclaims legal finality: 'court orders now bar the operators from the debt relief industry and telemarketing.' The only order covering both bans (Merdjanian) is a PROPOSED stipulated order, and the release explicitly warns 'Stipulated final orders have the force of law when approved and signed by the District Court Judge.' Of the two entered September 2025 orders, only Caldwell's includes a telemarketing ban; Hernandez is merely prohibited from violating the Telemarketing Sales Rule. The claim's collective 'the operators ... and telemarketing' is not supported for any single defendant set, and the word 'now' asserts an in-force status the source does not establish for Merdjanian. This also contradicts the desk's own unknown ('Whether the proposed order has been signed by the District Court Judge') — the check is internally inconsistent.
Resolved: Unresolved as drafted. Required rewrite of claim: 'A September 2025 court order bans two operators from the debt relief industry (one also from telemarketing); a proposed order against a third operator, not yet signed by the court, would ban her from debt relief and telemarketing and imposes a partly suspended $45.9M judgment.' Drop 'now bar the operators' and drop the collective telemarketing ban.
Objection (high): The limitation sentence is factually false about the harvested material: 'It also does not quote the sales scripts, phone numbers or websites used, so our harvested material cannot tell you ... which company names to look out for today.' The release names Superior Servicing LLC and three individual operators (Dennise Merdjanian, Eric Caldwell, David Hernandez). Telling readers the source contains no company names, when it names the defendant company, misstates the evidence and strips the one identifying detail a reader could actually use or search. Either name the entities or rewrite the limitation to say the release does not identify d/b/a names, scripts, numbers or websites used in the calls.
Resolved: Unresolved. Either (a) name Superior Servicing LLC and the three operators in the finding, attributed as FTC defendants, and delete the false clause; or (b) keep names out but rewrite the limitation to 'the release does not quote the sales scripts, phone numbers or websites used, so we cannot tell you what one of these calls sounded like.' The 'which company names to look out for' clause must go either way.
Objection (high): Entity-extraction failure leaked a PDF filename slug into audience directives: 'Send this to any ordergrantingstipulatedpifordefdennisemerdjanian you know' and 'Forward this to the ordergrantingstipulatedpifordefdennisemerdjanians in your life.' This is derived from the URL 'OrderGrantingStipulatedPIforDefDenniseMerdjanian.pdf' being treated as a person/audience noun. Any of these strings reaching output is disqualifying, and it signals the same extractor may be feeding other fields.
Resolved: Unresolved. Delete both slug-derived directives and add a filter rejecting audience nouns derived from URL paths/filenames or containing no whitespace over ~25 characters. Re-audit callout_options from the same extractor.
Objection (medium): Callout options apply the §11 Rule 2 targeting test inconsistently. The desk correctly dropped a directive for lack of evidence that a group was targeted, but then offers 'ATTENTION: PARENTS' and 'ATTENTION: EVERYONE' — the release says nothing about parents or PLUS borrowers, and 'EVERYONE' is not a described victim group either. 'ATTENTION: STUDENT LOAN BORROWERS' is the only callout the row supports.
Resolved: Unresolved. Remove 'ATTENTION: PARENTS' and 'ATTENTION: EVERYONE'; retain only borrower-scoped callouts supported by the row.
Objection (medium): Currency and present-tense risk framing. The source is dated 2026-07-21 and describes conduct halted by TRO in November 2024, with litigation against remaining defendants now resolved. The risk_line and the advice are written as live warning copy ('Callers claiming a link to the U.S. Department of Education ...'), and nothing in the finding tells the reader this specific operation was shut down roughly a year and a half before the release. Readers can reasonably conclude these specific calls are still going out. Add an explicit sentence that the operation was halted in November 2024 and the release describes enforcement outcomes, not an active campaign.
Resolved: Unresolved. Add to the finding: the operation was halted by court order at the FTC's request in November 2024, and this release (July 2026) reports settlement and ban outcomes rather than an ongoing campaign. Adjust risk_line to past tense for this operation while keeping the generic fee-for-forgiveness warning clearly marked as general advice.
Objection (medium): Money framing risks conflation. The $45.9M figure is (a) an allegation of what defendants took, and (b) separately the amount of a monetary judgment that is 'partially suspended' for inability to pay. Sentence 1 states it flatly ('took more than $45.9 million') without 'alleged', while sentence 2 does use 'alleged' — inconsistent attribution within the same finding. Combined with the unknown 'Whether any money will be returned', a reader may read the $45.9M judgment as a recovery pot; the suspension language in the source points the other way.
Resolved: Unresolved. Add 'alleged' to the first sentence, and state plainly that the judgment is largely suspended and the release does not announce consumer refunds.
Objection (low): Available protective resource omitted. The release cites ftc.gov/StudentLoans (consumer.ftc.gov guidance on avoiding student loan debt relief scams) alongside StudentAid.gov. The finding uses only StudentAid.gov while simultaneously complaining that the material offers little operational detail. The omitted link is source-backed and directly on point.
Resolved: Unresolved. Add ftc.gov/StudentLoans alongside StudentAid.gov, both sourced to the row.
Objection (low): watch_icons includes 'bank', implying bank-transfer payment, while the desk's own unknowns state 'How consumers were asked to pay — card, bank transfer or otherwise — is not stated.' Icon set should not assert a payment channel the source does not name.
Resolved: Unresolved. Drop 'bank' from watch_icons, or move payment-channel iconography behind an explicit sourced payment method.
Objection (low): targeting_dropped rationale is garbled: 'DIRECTIVE NOT IN EVIDENCE: calls — no harvested source describes this group.' 'Calls' is not a group; the note as written does not explain which directive was dropped or why, and would not survive audit review.
Resolved: Unresolved. Rewrite the drop note to quote the dropped directive text and name the group it implied.
Objection (low): Confidence 'high' is defensible on the sourced dates/dollar figure but is asserted over a single official release whose own headline ('Permanently Banned') runs ahead of the unsigned proposed order it describes. The confidence_reasons do not flag that the desk is repeating a press-release headline framing that the release's own final paragraph qualifies.
Resolved: Partially mitigated: the third confidence_reason already concedes thin operational detail. Still requires a fourth reason noting the Merdjanian order is proposed and unsigned, and that the desk is not adopting the release headline's 'permanently banned' framing as an accomplished fact.
Preserved dissent
ON THE RECORDI do not accept 'high' confidence on this draft as written. The sourced facts are solid, but two of the finding's own sentences misstate the record: the claim asserts that court orders 'now bar the operators' when the release says stipulated orders take effect only when signed and describes the Merdjanian order as proposed, and the limitation tells readers the material contains no company names when it names Superior Servicing LLC and three individual operators. A check that certifies a false limitation is worse than a thin one, because it tells the reader the evidence is emptier than it is.
ON THE RECORDI also object to publishing this as live-threat consumer copy. The conduct was stopped in November 2024 and the litigation is resolved; the alert's callout and advice read as if calls are going out today. If the desk wants to warn about the general fee-for-forgiveness pattern, it should say so in its own voice and stop implying the FTC release supports a present-tense campaign.
ON THE RECORDThe 'ordergrantingstipulatedpifordefdennisemerdjanian' directives are not a cosmetic slip. They are evidence that entity extraction on this row ran on URL strings rather than text, and I would not clear any field produced by that extractor — including callout_options — until it is re-run and re-audited.
The sources
Official sourceStudent Loan Forgiveness Scammer Permanently Banned from Debt Relief Industry and Telemarketing2026-07-21
The FTC alleged that a Nevada-based debt relief operation pretended to be affiliated with the U.S. Department of Education, falsely promised student loan forgiveness, and took more than $45.9 million from consumers, resulting in court orders banning the operators from debt relief and telemarketing.
Authority: official. Retrieved 2026-08-23.
Limitation: A press release about a settlement and court orders; it does not describe payment methods, the number of borrowers affected, refund plans, or whether similar operations are still calling.
Open the original source →Other checks
Published under standing founder pass (A9) — every claim source-mapped by the machine.
▼ What the machine checked
- ✓ Not a community submission.
- ✗ Draws on an FTC enforcement release, which names a defendant: "Student Loan Forgiveness Scammer Permanently Banned from Debt Relief Industry an".
- ✓ All 6 material sentence(s) map to FTC.
- ✗ anthropic returned "overstated"; google returned "overstated"; groq returned "overstated"; openrouter raised 2 objection(s) — published on the receipt, not blocking (A9 amendment).
- ✓ No audience band is set.
No human affirmed these. They were verified by the classifier described in Amendment A9, on 2026-08-23.
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