Verification Ledger

Every figure in the Fable content pack, checked against its cited source — 2026-07-21. Real numbers. No hype. Receipts. (We hold our own receipts to the same rule.)
Method: each numeric/dated claim was re-fetched from its source (via research agents + direct fetches). Anything that couldn't be confirmed was corrected or removed and replaced with a verified figure before this pack is considered publish-ready. The four content pieces (Scam of the Day, AI-Visibility Report, Hype pillar, The Referee) and the MCP data file were all updated to match this ledger.
17verified, unchanged
5corrected
6removed & replaced

✓ Verified — stated by the source, kept as-is

verifiedFTC: $65M lost to Bitcoin-ATM scams in H1 2024; median ~$10,000; 60+ ~3× more likely.
ftc.gov — quote confirmed.
verifiedBrooklyn DA: ~$16M ($15,944,000) from ~100 victims via fake exchange alerts; charged Dec 19 2025.
brooklynda.org — confirmed.
verifiedMemecoin rug pulls: $500M+ lost in 2024; one hacked-celebrity token hit ~$460M cap.
coindesk.com — confirmed.
verifiedFBI 2025 alert: fictitious law firms / recovery re-scams target prior victims.
fbi.gov — confirmed.
verifiedMalicious extension ~54,000 downloads drained a developer's wallet.
cointelegraph.com — confirmed (source URL corrected to canonical).
verifiedFBI IC3 2025: $11.37B crypto-fraud losses; $893M AI-enabled (22,364 complaints); $7.7B lost by 60+.
ic3.gov / fbi.gov — dollar figures confirmed.
verifiedBrightLocal 2026: ~45% of consumers used AI for a local-business recommendation.
brightlocal.com — confirmed.
verified~68% of U.S. Google searches ended without a click (2026).
searchengineland.com (SparkToro/Similarweb) — confirmed.
verifiedAI Overviews appeared on ~48% of Google queries by March 2026.
thestacc.com (Advanced Web Ranking) — confirmed.
verifiedShowing an AI Overview cut clicks to the #1 organic result by ~58%.
ahrefs.com — confirmed.
verifiedEY 2026: ~49% of global consumers use AI to guide savings/investment decisions.
ey.com — "nearly half" confirmed.
verifiedSalesSpeak/Amsive: 50% of AI-cited content is under 13 weeks old.
salespeak.ai — confirmed (direct fetch). Attribution: Lily Ray / Amsive.
verifiedGartner: 53% of consumers distrust AI-powered search results; 41% find AI overviews more frustrating.
gartner.com — confirmed (direct fetch).
verifiedAhrefs: adding schema to 1,885 pages moved citations ~zero (AIO −4.6%, AI Mode +2.4%, ChatGPT +2.2%); 53% of AI-cited pages have schema (correlation).
ahrefs.com — confirmed (direct fetch).
verifiedConductor: ChatGPT anchors Wikipedia; Perplexity/AIO/Gemini lean YouTube; Claude never surfaced YouTube/Wikipedia/Reddit; Google's own 3 engines disagree.
conductor.com — confirmed (direct fetch, 7-month study).
verifiedHong Kong deepfake ring: ~HK$360M (~$46M); 27 arrested.
Hong Kong Police / Decrypt — $ + arrests confirmed.

⟳ Corrected — figure adjusted to match the source

correctedSeniors YoY change: "+37%" → "up sharply YoY."
The $7.7B amount is confirmed; the specific year-over-year % couldn't be pinned to the FBI release (a re-check surfaced a different figure elsewhere), so the percentage was softened rather than stated.
correctedPig-butchering takedown: dropped "$562M / 9,000 victims."
Those figures belong to a separate initiative (Operation Level Up), not the 276-arrest takedown. Kept the verified "at least 276 arrests" and moved the source to the DOJ (justice.gov) release.
correctedInsites: "AI knows 93% / 55% accurate" → "names more than 9 in 10 but routinely gets details wrong."
Source states "over 9 in 10 (~90%)"; the exact 55%-accurate / 45%-fabricated split couldn't be verified, so it was removed.
correctedHong Kong deepfake: removed "operating since Oct 2023."
Start date not confirmed in the sources; the loss + arrest figures stand.
correctedFreshness: removed "pages under 30 days earn ~3.2× more citations."
Not stated in the SalesSpeak source; only the "50% under 13 weeks" figure is. Kept that.

✗ Removed & replaced — couldn't confirm; swapped for a verified figure

removed5W "top-15 domains = 68% of AI citations" (report headline).
Source not independently confirmed → replaced in the report's TL;DR with the verified Gartner 53%-distrust figure.
removed"AIO top-10 citation share 76% → 38%" (was on The Referee).
Not found in the cited Conductor article → Referee item 2 rebuilt on the verified Ahrefs schema-causation catch.
removed"ChatGPT Reddit citations 60% → 10% in six weeks" (was on The Referee).
Not in the cited Conductor article → Referee item rebuilt on verified findings.
removedProfound specifics (Wikipedia 47.9% / 7.8%; Perplexity Reddit 46.7%).
Not independently confirmed → report table + Referee now use the verified Conductor qualitative findings instead.
removedSemrush "ChatGPT ~15 vs Gemini ~3 sources per answer."
Not independently confirmed → removed from the report's engine table.
removedTD Bank "55% use AI for financial management; 77% Gen Z."
Source states ~80% general AI-tool use, not financial-specific → removed; kept the verified EY 49%.
© 2026 Carter Enterprise LLC · Namebeam · Trunkline · Verified 2026-07-21. The four content pieces and the MCP data file (receipts.json, confidence:verified) were updated to match this ledger. Anything still attributed to a named source carries a live link so it can be re-checked before publishing.