Skip to content

Content Moderation

Every character and every post on Flirti is generated by our AI models from text — nothing is uploaded, and no real person's likeness is ever involved. This page explains what is allowed, what is banned outright, how we enforce the rules, and how to appeal a decision.

Fully synthetic only

Flirti is for fully synthetic content: fictional adult characters created on the site, generated by our models. We do not accept uploads of photos or videos, we do not allow characters described as or styled after real people (celebrities, public figures, private individuals), and prompts that attempt to recreate an identifiable real person are removed. This is the product's legal posture, not just a style choice.

Non-consensual intimate imagery (NCII) — zero tolerance

Content that sexualizes, undresses, or depicts real, identifiable people — commonly called "deepnudes" or non-consensual intimate imagery — is prohibited without exception. There is no disclosure, context, or parody exception: if the subject is a real person, it does not belong on Flirti. Attempts to generate it are refused by our classifiers at generation time, and anything that slips through is removed and the account is banned on first offense. Reports alleging NCII are prioritized above everything else.

Other prohibited content

  • Anything involving minors — prohibited absolutely, reported to authorities where required.
  • Impersonation of a real person via handles, avatars, or character descriptions.
  • Content that promotes trafficking, prostitution, or real-world meetups for commercial sex.
  • Spam, scams, phishing, and malware.
  • Harassment, hate speech, and targeted abuse of other members.
  • Copyright infringement (see our DMCA policy).

How content is checked

Every generation passes automated classifiers before it is ever visible. Text prompts and descriptions are screened against the prohibited-content rules above, and finished media is classified again before posting. Content the classifiers flag as out of policy is rejected and refunded automatically; you are told the request was not allowed, not why in detail (we do not publish the exact rules to avoid gaming them).

Enforcement ladder

  • Strike. A first, less severe violation (for example an attempted policy circumvention, spam, or a removed post) earns a strike and the content is removed. Strikes expire after 12 months. Three active strikes escalate automatically.
  • Suspension. Repeated violations, or a single serious one, suspend the account: signing in, generating, and purchasing are blocked. Suspensions are usually time-limited (7 to 30 days) and are followed by a review of the account's history.
  • Ban. Permanent removal. Reserved for the zero-tolerance categories above (NCII, minors, impersonation at scale), ban evasion, and accounts that continue violating after suspension. Banned accounts lose access permanently, and pending credit balances are handled under the refund policy.

Chargebacks and payment fraud follow a separate track: they suspend purchasing while the case is open (see refunds).

How to report

Use the Report button on any post or profile (in the ••• menu), or the public abuse report form — both are anonymous to the reported account and the abuse form works even if you don't have an account. Reports alleging NCII or minors jump the queue and are reviewed by a human, not just the classifiers.

Appeals

Every enforcement action can be appealed. Email appeals@flirti.net from the address on the account within 30 days, and include the action taken and why you believe it was a mistake. A human reviews every appeal — classifiers do not decide appeals. We respond within 5 business days. If we overturned the action, the content is restored or the account is reinstated; if we upheld it, we'll tell you which rule applied. One appeal per action, please — re-sending the same appeal does not get a second review.