Using AI on Instagram is easy.
Keeping track of when you are supposed to disclose it is becoming considerably less so.
A new Hootsuite guide lays out Instagram’s increasingly detailed AI disclosure system, which now includes separate labels for individual pieces of AI content and entire accounts built around synthetic people. For brands and agencies, that means AI disclosure is starting to look less like a publishing toggle and more like another step in the approval process.
Not Every Use Of AI Needs A Label
Instagram currently requires users to disclose realistic-looking video or realistic-sounding audio that has been created or materially altered using AI.
That includes things like cloned voices, realistic AI voiceovers, generated people appearing in video, or real footage where people or environments have been convincingly replaced.
AI-written captions and scripts do not require disclosure. Neither do AI-generated or AI-edited still images under Instagram’s current self-disclosure rules, although Meta may automatically add an “AI info” label when its systems detect AI-related metadata or other signals.
Instagram also now has a second layer: the “AI-generated profile” label for accounts whose featured person is synthetic.
And that one comes with a distribution consequence.
If Instagram identifies an AI-generated profile that has not properly disclosed itself, the account can stop being recommended to new users until the issue is corrected or successfully appealed.
For Brands, The Problem Moves Upstream
The interesting part here is not another label appearing in the feed.
It is what happens before the post goes live.
A brand might approve a creator video without realizing the voiceover was generated. An agency might use AI to alter a background. A virtual influencer might appear in branded content. The person publishing the post may not even be the person who created the asset.
That makes disclosure a workflow question.
Hootsuite recommends brands explicitly address AI in creator briefs and add an AI-disclosure check to approval processes before content is published. Meta says it may apply penalties when users fail to disclose qualifying realistic AI video or audio, although it does not publicly spell out every potential penalty.
This comes while Meta is simultaneously pushing generative AI deeper into Instagram creation and advertising.
So the platform is effectively encouraging more AI use while building a growing system of signals to explain when that AI use becomes visible enough to require context.
For social teams, that creates a fairly simple new question to add before approving anything:
What did AI actually do here?
Because increasingly, “we used AI” is not enough information.