First, look at the evidence
Open the flag. It names the type (for example Hate Speech or Undisclosed Sponsorship), the post, the excerpt or frame it came from, and when it was published. If the evidence is not there, the flag should not be either — that is our rule, and a flag without evidence is itself a bug worth reporting.
Brands
Use Report as false positive on the flag. You will see an acknowledgement on the card; the flag moves to the reported list, stops counting in alerts and digests, and feeds the detector's accuracy review.
Creators
Email [email protected] with the link to the flag and, in a sentence, why it is wrong (a quoted lyric, a joke read literally, a translation error, a video frame misread). A person reviews it. A flag confirmed as a false positive is retired everywhere it appears and your score is recomputed at the next refresh.
What we will not do
We do not remove accurate findings because they are unflattering, and we do not lower a score on request. We do fix mistakes, quickly, and we say so in the report.
Still need help? Message us and a person will answer.