What we found
On 30 July 2026 the FTC published a consumer alert stating that one of the best ways to fight fraud is to talk about it. The alert says that when you know scammers' tactics, they are easier to spot and avoid, and that the FTC is offering facts about fraud to help start that conversation. Pick one person this week and tell them out loud about a scam approach you have seen yourself — a real conversation, not a forwarded link. Our harvested material does not describe any specific scam, contact method, dollar loss, or number of people affected, so nothing here should be read as an alert about a particular fraud in progress. We do not know which fraud tactics or figures the FTC's underlying facts page lists, because only the alert's opening text was harvested. LIVEFRAUD ALERTS is independent and is not affiliated with the FTC or any agency named here; this check pass rests on a single harvested row.
- Only one harvested row, and it is a general awareness alert rather than a report of a specific scam.
- The row's text is a short excerpt; the underlying fraud facts, figures and tactics were not harvested.
- No affected group, contact channel or payment method is described anywhere in the harvested text.
- Reviewed by 4 models, 3 from independent houses.
THE RULE
No specific financial risk is described in the harvested material — this is general awareness guidance, so treat any concrete scam details you hear elsewhere as unchecked.
What we don’t know
- Which specific fraud tactics the FTC's facts about fraud describe
- Whether any particular group is being targeted, and by what contact channel
- How money is requested or moved in the scams the alert alludes to
- Any counts of complaints, victims or reported losses
The bench — who voted
4 INDEPENDENT AI MODELS REVIEWED THIS. 2 OF 4 — TWO MODELS DISSENTED. READ WHY.
The card names a count. Here are the seats behind it, with what each one said.
ANTHROPICclaude-opus-5 · LOW
The evidence row says only that the FTC alert urges talking about fraud; it does not say 'with friends and family'. The CLAIM's 'talk with friends and family' adds an audience the harvested text does not name.
DISSENT, PRESERVEDThe factual core — that on 30 July 2026 the FTC published a consumer alert saying talking about fraud helps people spot and avoid scammers' tactics — is fully carried by row [1]. My 'overstated' verdict rests on two specific additions (the friends-and-family framing and the unsourced 'facts about fraud' offering), not on the core.
DISSENT, PRESERVEDRisk to an ordinary reader is genuinely low: the item names no scam, no payment channel and no loss figure, and it explicitly disclaims being an alert about a fraud in progress. I would resist any push to grade this moderate or high merely because it appears under an alert banner.
DISSENT, PRESERVEDOne harvested row, opening paragraph only, is a thin basis for publication of anything; the stated low confidence is appropriate and should not be raised without a second source.
GOOGLEgemini-flash-latest · LOW
Raised no objection; read the evidence as holds.
GROQopenai/gpt-oss-120b · MODERATE
The finding includes a specific call to "Pick one person this week and tell them out loud..." which is not present in the harvested evidence; the evidence only contains the opening paragraph about talking as a strategy.
DISSENT, PRESERVEDThe evidence does support the core claim that the FTC urged discussing fraud tactics, but the additional actionable instruction is not substantiated.
OPENROUTERopenrouter/free · MODERATE
The finding specifies 'talk with friends and family' which is not mentioned in the evidence. The evidence only states 'talk about it' generally.
Reviewed by 4 independent models: 1 found it carried by the evidence, 3 did not.
▼ Protocol & challenge record
Objection (high): Date currency / plausibility is unverified. The only row is stamped 2026-07-30 and the finding asserts in the present perfect that the FTC 'has published' it. Nothing in the packet records when the harvest ran, so the desk cannot tell whether this alert is fresh, months stale, or a forward-dated / test-feed artifact. A single row with a publication date that may sit ahead of the pipeline's own clock deserves an explicit harvest-timestamp check before anything ships with a date in the lede.
Not resolved — preserved on the record.
Objection (high): The advice sentence is unsourced AND contradicts the source. Row text says the FTC 'has some facts about fraud to help you start the conversation' — i.e. it points readers at its own material. The finding instead instructs readers to relay 'a scam approach you have seen yourself — a real conversation, not a forwarded link.' That (a) carries row_ids: [] , (b) invents a first-hand-experience requirement the FTC never sets, and (c) actively disparages sharing the link the FTC is offering. Editorialising against the source's own call to action inside a 'check' is worse than saying nothing.
Not resolved — preserved on the record.
Objection (medium): targeting_dropped contains a garbled fragment: 'DIRECTIVE NOT IN EVIDENCE: member, approach, have, seen, yourself'. Those are word-salad tokens lifted out of the advice sentence, which suggests the directive scrubber tokenised and partially rewrote copy rather than removing a discrete directive. This is a pipeline-integrity signal: if the scrubber mangled that string, the surviving advice sentence may itself be the residue of a dropped share-directive that should have been cut entirely.
Not resolved — preserved on the record.
Objection (medium): watch_icons ['person','phone','link'] are not in evidence and contradict the packet's own limitation. The finding states plainly that no contact method is described, then attaches phone and link icons that signal telephone and URL attack vectors to the reader. Icons are claims. Either drop to nothing (or 'person' at most) or source them.
Not resolved — preserved on the record.
Objection (medium): Framing overclaim by container. 'ATTENTION: EVERYONE' under a 'LIVEFRAUD ALERTS' masthead dresses a routine, detail-free awareness post as a live fraud alert. The disclaimer sentence mitigates but does not undo the headline framing, and running content like this at alert volume trains readers to discount real alerts.
Not resolved — preserved on the record.
Objection (low): Small causal compression in the claim: the source says knowing scammers' tactics makes them easier to spot and avoid, and separately that talking is one of the best ways to fight fraud. The claim's 'talk ... as a way to spot and avoid scams' welds the two into a single mechanism the source does not state in those terms. Cheap to fix by mirroring the two-step structure.
Resolved: Entity and domain match are clean: consumer.ftc.gov with a /consumer-alerts/2026/07/ path is the FTC's genuine consumer-alert channel, and the headline 'Talk to your friends and family to fight fraud' supports both 'FTC', 'consumer alert' and the substance of the claim. The paraphrase is faithful; only the joining logic needs a light rewrite.
Objection (low): Confidence label conflates 'low-information source' with 'low confidence in the claim'. The claim itself is close to trivially true against an official FTC consumer-alerts URL; what is thin is the payload, not the verification. Reporting 'low' confidence on a well-sourced claim is its own kind of miscalibration and misleads downstream consumers of the confidence field.
Resolved: The packet's limitation and unknowns fields are unusually honest — they state outright that no scam, channel, loss figure or victim count is present and that only the opening paragraph was harvested. A reader is not misled about the thinness of the payload; the fix is to relabel the confidence axis, not to add hedging.
Objection (low): Alternative explanation not addressed: the harvested block could be an RSS/aggregator rendering rather than the live page, and FTC alerts are sometimes updated or re-dated after publication. The 'official' tag is asserted by the harvester, not independently confirmed by a fetch of the URL. With a single row there is no cross-check.
Not resolved — preserved on the record.
Objection (low): Reader utility gap: the finding tells readers the FTC has 'facts about fraud' but never surfaces the alert URL in the body, so a reader cannot get to the material the alert exists to promote. Unknowns correctly note the facts page was not harvested — but that was a harvest choice, and one extra fetch would have removed three of the four unknowns.
Not resolved — preserved on the record.
Preserved dissent
ON THE RECORDI would not publish this item in its current form. It contains no checkable fraud, no channel, no loss and no target group; the only proposition verified is that a government agency told people to talk to each other. Under an 'ATTENTION: EVERYONE' alert banner that is closer to filler than to fraud reporting, and it consumes reader attention that a real alert will need.
ON THE RECORDThe advice sentence should be deleted outright, not softened. 'Tell them about a scam approach you have seen yourself — a real conversation, not a forwarded link' is unsourced, sets a bar the FTC did not set, and tells readers to do the opposite of what the source asks. Inventing behavioural instructions and placing them inside a finding body is exactly the failure mode a check pass exists to prevent.
ON THE RECORDThe 'phone' and 'link' icons are indefensible. The packet says in plain text that no contact method is described and then draws two contact methods. If icons stay, this row should carry none.
ON THE RECORDI disagree with the 'low' confidence label. The claim is well sourced against an official FTC URL; what is low is informational value. Merging those two into one field will, over time, make the confidence signal useless.
The sources
Official sourceTalk to your friends and family to fight fraud2026-07-30
On 30 July 2026 the FTC published a consumer alert stating that one of the best ways to fight fraud is to talk about it, because knowing scammers' tactics makes them easier to spot and avoid.
Authority: official. Retrieved 2026-08-15.
Limitation: The harvested excerpt is only the alert's opening paragraph; it names no scam type, no tactic, no victim group, no loss amount and no contact or payment method.
Open the original source →Other checks
Approved by ihubglobalhq on 2026-08-17, 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