Not a rank tracker with an AI badge.
Fama asks the AI engines the questions your buyers actually ask, scores the answers with a rules pass you can read, and builds the content that closes the gaps. This page is the mechanism, step by step, with the honest edges left in. Everything below grounds in the same verified brand record as every Jinn product: how the record works.
One rule governs this page: every mechanism described here is verified against the code that runs it. Where a capability is still in build, the copy says so.
How prominently six AI engines name your brand when asked real buyer questions, blended into one 0 to 100 Brand IQ.
A deterministic rules pass reads where your name lands in each answer and scores it 0 to 5. No AI grades another AI, so the same answers always produce the same number.
A page-level Playbook of rewrite blocks grounded in your verified Brand DNA. Every block passes a hard compliance gate and a voice check before you see it.
Shortcuts. Confirmed Corrections never move the score, accusations require a quality source, and no accuracy number publishes until the method clears its own floor.
Measure, fix, measure again.
Four stages, and every one of them is inspectable. The loop only works because the measurement half is deterministic: when the number moves, the answers changed, not the ruler.
- AskYour real buyer questions go to six AI engines: ChatGPT, Perplexity, Claude, Gemini, Grok, and Google AI Overviews. The scored question set is identical on every plan tier.
- ScoreA rules-based reader scores how prominently each answer names you, 0 to 5: from not mentioned at all up to named first and called a top pick. Per-engine scores blend into one 0 to 100 Brand IQ, and a run with a missing engine flags itself degraded instead of quietly pretending full coverage.
- FixFama selects the pages behind your gaps and drafts rewrite blocks grounded in your verified Brand DNA and your confirmed Corrections: FAQ blocks, About prose, structured data. Nothing reaches you until it clears the gates below.
- ProveRe-run the audit and the same questions go back to the same engines, scored by the same rules. You publish better content and the number answers honestly, either direction.
A separate fact-audit lane grades what the engines say about you for accuracy. Its machinery is engineering-verified and described below, and it is in build: it publishes nothing to customers yet.
Two engine rosters, on purpose.
Visibility and accuracy are different questions, so Fama runs them on different rosters and never bends one roster to match the other.
- The visibility laneThis lane answers how prominently the engines mention you, and rolls up to Brand IQ. Google AI Overviews counts here. Engine weights renormalize over the engines that actually answered, and the result carries a degraded flag whenever coverage was short.
- The fact-audit laneThis lane answers whether the engines get your facts right, on the engines proven to elicit factual claims. Google AI Overviews is excluded from this lane by design, not by omission, and the exclusion is written into the roster itself.
- A floor, not a shrugA fact-audit run may publish or rank only when at least five of the six engines answered every question. A brand below that floor is re-queued for a fresh run, never scored on thin coverage.
- Census, not vibesEvery engine in the lane answers every question, twice per run. The grid is a census of the roster, never a convenience sample of whichever engines were easy that day.
The fact-audit lane is in build and publishes nothing yet. These rules are how it is engineered, not a live customer surface.
Adjudication, not one model’s opinion.
When the fact-audit lane grades a claim an engine makes about you, the grade happens in two stages. Stage one grades each claim against DNA-grounded truth where your record covers it, or against cross-engine consensus where it does not. Anything suspicious gets flagged.
Stage two re-verifies every flagged claim with a live web search. A source only counts if the search tool itself surfaced it: a URL the model typed from memory never satisfies sourcing, so a hallucinated citation cannot launder a verdict.
The verdicts have a floor. Calling a claim wrong or fabricated requires an allow-listed authority-class source: your own domain, government or education domains, or named wire and encyclopedic sources. Anything less and the verdict is forced down to an unverified disagreement. Fama never accuses without a quality source, and that is enforced in the verifier, not promised in an editorial policy.
Claims that cannot be sourced either way are excluded from the accuracy denominator entirely. You are never scored down for a claim nobody could verify. And the lane does not pretend to a census it does not run: every suspicious claim is individually verified, plus a statistically sampled check on the rest, sized to catch the failure consensus cannot see, which is every engine repeating the same wrong fact.
When a score does publish, it will carry a 95 percent confidence interval built from that sample, and with too few confident observations the code claims no estimate at all rather than a shaky one.
No accuracy percentage appears on this page, on purpose. The method must clear its own agreement floor against a human-labeled golden set before any number publishes. Until it does, you get the discipline, not the number.
The gates are code, not another AI.
Every Playbook block passes checks that are plain string and regex math. They run at zero token cost, they cannot be sweet-talked, and the same text always gets the same verdict.
Built so you cannot cheat it. Neither can we.
Fama sits on the record.
The brand record, the Brand IQ math, and the study methods are shared substrate, documented once and linked here rather than re-explained. The terms this page leans on are defined in the dictionary.
The questions people ask.
- Does an AI grade my Brand IQ?
- No. The score is a deterministic rules pass over where your name lands in each engine’s answer. No LLM judge sits in the scoring path, so the same answers always produce the same number.
- Can I pay, or optimize, to move the score?
- You can publish better content and re-run the audit. You cannot lobby the number: the scored question set is identical free and paid, and confirmed Corrections are structurally unable to change the score.
- Does Fama watch my site around the clock?
- Monitoring is off by default on a fresh account; you switch on scheduled re-audits when you want the trend line. Crawl checks are different: they run inside engagements, on demand, with no standing cron. Fama gives you reproducible measurement, not always-on theater.
- Where does the fix content come from?
- From your verified Brand DNA plus the Corrections you have confirmed, drafted into page-level rewrite blocks and gated before you see them. Nothing is scraped from competitors, and nothing ships unchecked.
- Why is there no accuracy percentage on this page?
- Because the fact-audit method has not yet cleared its own publish floor against a human-labeled golden set. The floor is code, the discipline is real, and the number stays unpublished until the method earns it.
Run the loop once. Keep the number.
The full audit is free: your Brand IQ across all six engines and the specific gaps behind it. No card.