Brand Hallucination
A brand hallucination is a confident but false statement an AI model makes about a company or product — invented pricing, features that don't exist, wrong integrations, or fabricated policies. Unlike a bad review, it carries the authority of a neutral assistant and repeats to every buyer who asks a similar question.
Hallucinations about brands have specific, recurring shapes: prices interpolated from competitors or old pages, features assumed from category norms ('all CRMs have lead scoring, so X probably does'), discontinued plans still quoted, mergers and shutdowns invented from ambiguous entity names. The model is not lying; it is filling gaps with plausible pattern-completion.
The business damage is asymmetric. A hallucinated missing feature quietly removes you from shortlists; a hallucinated low price creates sales calls that start with disappointment; a hallucinated compliance claim can create real risk if buyers rely on it. None of these show up in analytics — the buyer simply acts on the wrong answer.
Mitigation has two halves. Reduce the gap-filling: publish the facts extractably, keep them current, make the entity unambiguous. And detect what slips through: monitor answers with claim extraction so false statements are caught and traced to their source instead of being discovered by a prospect. Perciva classifies extracted claims as accurate, outdated, false, or missing for exactly this reason.
Related terms
See what AI currently says about your brand — free snapshot.