Instagram will begin limiting the reach of accounts that present AI-generated people without disclosing that the profile’s subject is synthetic. The platform is also renaming its existing “AI creator” label to “AI-generated profile,” language it says will make the warning easier to understand.
The policy targets a specific form of deception: an account that appears to belong to a real person when its central persona was generated or substantially created with AI. Instagram says creators who use the label will not be penalized simply because their featured person is artificial.
The change reflects growing tension around AI influencers and synthetic experts. Platforms can encourage creative tools while users increasingly demand to know when the person speaking to them does not exist.
The rule focuses on synthetic identities
Instagram says the label is not required for every use of AI. Editing a photograph, polishing a caption, creating graphics or making other creative changes does not automatically turn an account into an AI-generated profile under the announced policy.
Instead, the disclosure applies when the person featured by the account was generated or substantially produced with AI. Accounts that fail to label such a persona may have their reach reduced. Instagram has not detailed the size of that reduction, how violations will be detected or what appeal process will be available.
The company says users have complained about encountering profiles that looked human, only to discover later that the featured person was artificial. Renaming the label is meant to remove ambiguity about what is being disclosed.
That clarity matters because “AI creator” could describe a human creator who uses AI tools. “AI-generated profile” places the emphasis on the identity presented to viewers. The wording does not verify every post, but it gives users a more direct signal about the account’s central character.
AI influencers have crossed sensitive boundaries
Concern is not limited to fictional entertainment. Earlier this year, Wired investigated a network of apparently AI-generated male influencers promoting the gay dating app Goose on Instagram. More than two dozen accounts appeared to feature synthetic influencers, and some reportedly contacted potential users through direct messages encouraging them to register.
Health and wellness content presents another risk. The New York Times reported in July that it found hundreds of AI-generated doctors, healers and wellness personalities promoting supplements or making health claims on social media.
Those cases demonstrate why disclosure affects more than aesthetics. A synthetic persona can borrow the appearance of personal experience, expertise or intimacy. When the commercial or medical context is hidden as well, users may evaluate a message as if it came from a real individual with real credentials.
Instagram’s distribution penalty creates an incentive to disclose, but the outcome will depend on enforcement. The policy does not establish how accurately the platform can identify synthetic people or distinguish them from heavily edited photographs of real people.
Meta is narrowing several disputed features
The announcement follows criticism of an Instagram tool that let users create images using other people’s likenesses. Users objected that their public Instagram content could be used without an explicit opt-in, and Meta removed the feature.
Meta also reached an $18 billion settlement with US states last week over allegations about Facebook and Instagram’s effects on children and teenagers. Under that agreement, the company will introduce a default two-hour daily limit for teens across both services, a Night Mode block, muted notifications during school hours and other restrictions.
Those separate decisions share a pressure point: platform features that expand engagement or creation can also create harms users did not knowingly accept. The new profile label addresses one portion of that problem by making synthetic identity more visible.
Instagram is not banning AI personas. It is drawing a line between disclosed fiction and an account that gains attention by appearing human. Reduced reach gives the rule practical weight, although the company has not yet supplied enough detail to judge consistency or scale.
For users, the benefit will be measurable in whether fewer artificial personalities pass as real. For creators, the test will be whether a clearer label and enforcement process distinguish intentional synthetic characters from ordinary AI-assisted editing without punishing the latter.
