Severity tags appear on cards; MVP index is count-based (flags ÷ stories), not severity-weighted.
Meter counts clear lies, evidence failures & logical collapse — spin stays in the spicy notes, off the %.
This pass sampled 28 stories across world wires, US national desks, US South/Miami locals, and intl desks — then scored hard claims against evidence. Four hard-BS flags landed; most in the World / MENA track.
Wires and international desks set the factual baseline: what can be checked against primary docs, releases, and cross-source triangulation.
US regional prestige papers and major-market TV/radio/digital locals that routinely feed — or are absorbed by — national framing. This pass included South/Miami crash coverage.
UK, EU, Canada, LatAm, MENA, Africa, Asia-Pacific — named outlets, same claim-vs-evidence rules. No side-picking by geography.
Honest small-sample read: overall index is 14% (4/28). World is elevated because all four hard-BS cards sat on the World track. Northeast / Midwest / West lacked claim density this evening; South was dominated by clean crash coverage (0 hard-BS).
Named national and international outlets sampled for the world rollup. Same hard-BS bar everywhere.
Aggregation across prestige, wire, TV/digital, and partisan-aware desks. We score claims against evidence — not outlets against each other.
| Outlet | Type | Why scan |
|---|---|---|
| Reuters | Wire | High-volume, attribution-heavy baseline |
| Associated Press | Wire | US-facing world & national desk |
| AFP | Wire | Non-US wire balance |
| BBC News (World) | Prestige / TV-digital | Global framing; strong primary-doc habits |
| The Guardian (Intl) | Prestige | World + UK lens |
| Al Jazeera English | TV-digital | Global South / MENA density |
| Financial Times | Prestige | Markets + geopolitics claims |
| DW / France 24 | TV-digital | EU international desks |
| NHK World / Japan Times | Regional intl | Asia desk balance |
| Outlet | Type | Note |
|---|---|---|
| NYT · WaPo · WSJ · LA Times | Prestige | National + regional density |
| USA Today · Axios · Bloomberg | Digital / markets | Briefing & markets claim volume |
| NPR / PBS NewsHour | Public | Broadcast digital |
| Politico · The Hill | Beltway | Policy / Congress claims |
| CNN · ABC/NBC/CBS | TV-digital | Network desks |
| Fox News · MSNBC | TV-digital | Partisan-aware — score claims, don't mirror tone |
Factual NTSB/local coverage (WPBF, Guardian, BBC) — scanned as clean hard-news. Listed so users see we aggregate locals too without inventing BS. South mini-meter: 0% hard-BS this pass.
Humans set principles and scope. AI finds patterns, aggregates sources, drafts cards, and updates the meter — on a fixed twice-daily cadence. Tonight's pass: 4 hard-BS ÷ 28 scanned = 14% (count-based MVP).
Only hard BS:
Contextual only — never inflate the index:
Every day including weekends. “Today's” index reflects the latest completed pass.
Each hard-BS card names persuasion tactics so readers can spot the move next time — separate from the evidence failure that put the card on the meter.
Severity 1–3 is shown on cards for reader weight. Tags (contradicted-by-dataabsolute-claimunfalsifiablevibes-as-facts) are descriptive.
MVP index is count-based: hard-BS flags ÷ stories scanned. Tonight: 4 ÷ 28 = 14%. Severity does not yet reweight the percentage — we state that honestly.
Duplicate claims across outlets → one card when possible; don't inflate % by cloning. Zero hard-BS days (and clean regional samples) are allowed — honesty over filler. Small-sample regional mini-meters carry a caveat.