AI NativeThe Brand Truth Studio is where you tell AI Native the truth about your brand, so it can tell when an AI answer gets you wrong. It has three parts on one screen: who you are, the facts that must be right, and the guardrails answers are checked against.
Brand identity
Identity is who you are in names and domains: your primary brand name, any aliases or sub-brands buyers might use, and the domains that belong to you. This matters because a scan has to judge whether an answer is really about your brand. If an alias is missing, a recommendation of you under a different name can be missed; if a wrong domain is listed, a competitor's mention can be counted as yours.
You confirm the identity here, and you can ask AI Native to suggest identity details and then accept or correct them. When you change the identity, a retag can re-evaluate existing answers against the corrected view, so the records line up with who you actually are.
Facts you verify and reject
Facts are the specific claims about your brand that an answer must get right: a settlement ratio, an entry age, a fee, a coverage detail. There are three ways to capture them:
- the AI fact interview, which asks brand-specific questions you answer in plain fields, saved as verified,
- bulk import, where you paste one fact per line as
Label: value, - and adding facts by hand.
Each fact carries a status. A fact you have confirmed is verified. AI Native can also gather candidate facts for you to check; those sit as unconfirmed until you act. You verify a candidate that is correct, which promotes it into the trusted set a scan checks answers against, or you reject one that is wrong or irrelevant so it never anchors a fact-check. Verifying and rejecting is how the trusted set stays clean rather than filling up with noise.
Once a fact is verified, branded scans check AI answers against it. When an answer states something that contradicts a verified fact, it is flagged as inaccurate, and the count of contradictions is one of the first things the command center shows you.
Guardrails
Guardrails are rules an answer is checked against beyond plain facts, the things that must or must not appear, often for compliance reasons in regulated industries. Each guardrail has a label, a kind, and a severity, so a hard requirement reads differently from a soft preference. You add, edit, and delete them here, and a separate compliance view shows how answers measure up against them.
Why this comes first
The Brand Truth Studio is the foundation the rest of the measurement rests on. Identity decides whether an answer counts as being about you. Facts decide whether an answer about you is accurate. Guardrails decide whether it is safe. Get these right early, and every scan that follows is judging answers against the truth rather than a guess.