Instagram's AI label, explained: what it means, why you got it, and whether it hurts reach
In the first week of September 2026, two things happened at once: Instagram announced a new “AI-generated profile” label with reach penalties for accounts that ignore it, and photographers and brands started finding an “AI Content” badge on photos they never touched with generative AI. The result is a lot of people asking the same three questions: what does the label mean, how do I get rid of it, and will it hurt me?
This guide answers all three with the actual policy documents and dates, and then addresses the question a business owner really has: if I use AI to make my posts, what applies to me?
The short answer
Instagram's “AI info” label appears when Meta detects industry-standard AI markers in an image (the metadata that tools such as Photoshop, Firefly and most generators embed) or when you disclose AI use yourself. It was renamed from “Made with AI” on 1 July 2024, and labelled content is kept and shown normally rather than removed or demoted for carrying the label.
The rule that does affect reach is newer. Since 31 August 2026, a profile that features an AI-generated person must switch on the “AI-generated profile” label or face reduced reach and loss of recommendations. Instagram says that rule does not apply to people using AI to edit photos, polish captions or create graphics, which covers a business using Virallized AI to write and design its posts.
The two labels, and which one applies to you
AI info (content label). Sits on an individual post. Applied automatically when Meta detects AI markers in the file, or when you disclose it yourself. It is informational: Meta's stated approach is to keep the content up and add context rather than remove it. This is the one people see on photos and ask how to remove.
AI-generated profile (account label). Switched on by you in Edit profile. Required, since 31 August 2026, for any account that regularly features an AI-generated person rather than a real human. Miss it and Instagram says the profile may lose reach and eligibility for recommendations. Add it and Instagram says reach is unaffected.
The “AI Content” badge you did not ask for. The same content label, applied automatically. In September 2026 The Verge reported weeks of it landing on photos edited only with tools like background removers or basic photo apps, while some fully generated images passed unlabelled. If you have one on a real photo, this is why.
What triggers the AI info label
Meta's policy, published in April 2024 and updated since, names two triggers: industry-standard AI image indicators in the file (the C2PA and IPTC metadata that generators and editors such as Photoshop and Firefly write into an image), and self-disclosure when you upload. The label was called “Made with AI” from May 2024 and renamed to “AI info” on 1 July 2024 after complaints that it overstated how much of an image was generated. From September 2024, content that was only edited with an AI tool, rather than generated, carries the label less prominently, in the post's menu rather than under the username.
In practice that means a photo you retouched in a tool that writes AI metadata can be labelled even though you took it, and a fully generated image from a tool that writes no metadata can pass. The detection reads the file's markers, not the picture.
Why real photos are being labelled in September 2026
On 4 September 2026 The Verge reported that users had spent weeks watching Meta's automatic system attach an “AI Content” badge to pictures they neither generated nor edited with generative tools. The pattern matched 2024: a photographer who removed a speckle from a portrait got flagged; a brand that shot on an iPhone and touched the image up in a standard photo app got flagged; meanwhile a fresh account uploading fully generated pictures carrying machine-readable AI markers got nothing. Background removers, including Canva's, were the tool most often named by affected users.
That is the state of things at the time of writing. The label on a real photo is a detection error, not a judgement about you, and it does not change how the post is distributed.
Can you remove the AI label on Instagram?
There is no in-app control to remove a label Instagram applied automatically, and as of September 2026 no appeal button on the post. What people report working is re-exporting the original photo from a tool that does not write AI metadata (a plain export from the camera roll, for instance) and posting that file. What you should not do is strip the markers from an image that genuinely was AI-generated: the whole point of the disclosure system is that generated people and scenes are labelled, and undisclosed AI personas are now the thing Instagram penalises.
If you self-disclosed by mistake when uploading, the label you added yourself can be removed by editing the post. If it came from detection, it stays with that file.
Does the AI label hurt reach?
Not by itself. Meta's own policy statement is that it keeps labelled content on the platform in order to add informational labels and context, and only fact-checked false or altered content is shown lower in Feed. Instagram's September 2026 announcement repeated that accounts which proactively add the AI-generated profile label see no change to reach. The penalty is specifically for profiles that feature an AI-generated person and do not disclose it.
So: an AI info label on a post does not throttle the post. An undisclosed AI persona does throttle the account. Those are different things, and most of the anxiety online comes from conflating them.
The AI-generated profile rule (31 August 2026)
Instagram renamed its “AI creator” label to “AI-generated profile” and announced that profiles featuring an AI-generated person which do not carry the label will have their reach limited and may lose eligibility to appear in recommendations. The company's reasoning, in its own words: people do not like seeing a profile that seems human, only to find out later that the person featured is AI-generated. To add the label, edit your profile and turn on AI-generated profile; it then appears on the profile, next to it in search, and beside the account's content in Feed, Reels and Stories where the media does not already carry an AI label.
The rule is explicitly narrow. Instagram says it does not apply to creators using AI to edit photos, polish captions, create graphics or make other creative tweaks. It is about who the featured person is, not which tools made the post.
What this means for a business that uses AI to make posts
If you are a real business run by real people and you use AI to write captions, draft carousel slides, design graphics or plan a feed, no profile label is required and nothing in the 2026 rule applies to you. If a post contains a fully AI-generated image (a product scene, an illustration), it may carry the AI info label, which is informational and does not affect distribution. The line you should not cross is presenting an AI-generated person as a real staff member, model or customer without the profile label.
That is how Virallized AI is built to be used: it writes and designs branded carousels and posts from your topics, in your voice, with your logo and colours, as a tool in your hands rather than a persona in your place. Where it generates an image for a post, that image is yours to keep, replace with a photo, or label as you see fit, and the designed text-led carousels that most business accounts run day to day contain nothing the label system is looking for.
A separate July 2026 setting worth knowing about
Unrelated to labels but widely confused with them: in July 2026 Meta rolled out an image feature that can use public Instagram photos as references for AI-generated images, switched on by default. Several outlets, including Malwarebytes, reported the opt-out path at the time: in your profile menu, under Sharing and reuse, turn off the option that allows people to create with and reuse your content. If your account is public and you would rather not be a reference image, that is the setting to check.
Common questions
What does the AI label on Instagram mean?
The “AI info” label means Meta detected industry-standard AI markers in the file's metadata, or the uploader disclosed AI use. It is informational: the post stays up and is distributed normally. It was called “Made with AI” from May 2024 and renamed on 1 July 2024.
How do I remove the AI label from my Instagram post?
If you added it yourself at upload, edit the post and remove the disclosure. If Instagram applied it by detection, there is no in-app removal or appeal as of September 2026; people report success re-exporting the original photo from a tool that does not write AI metadata and posting that file. Do not strip markers from genuinely AI-generated images.
Does Instagram allow AI-generated content?
Yes. Meta's policy keeps AI-generated content on the platform with labels rather than removing it, unless it breaks other rules. The one requirement with a penalty attached is the AI-generated profile label for accounts that feature an AI-generated person, in force since 31 August 2026.
Does the AI label reduce reach on Instagram?
No. Meta says labelled content is kept and shown with context, and only fact-checked false or altered content is shown lower in Feed. The reach penalty announced in 2026 applies only to profiles featuring an AI-generated person that fail to switch on the AI-generated profile label; labelled profiles are unaffected.
Do I need the AI-generated profile label if I use AI to write and design my posts?
No. Instagram states the label does not apply to creators using AI to edit photos, polish captions, create graphics or make creative tweaks. It applies only to profiles that feature an AI-generated person instead of a real human.
Can Instagram detect AI-generated content?
It detects the AI markers written into a file's metadata by generators and editors, not the picture itself. That is why edited real photos sometimes get labelled while some fully generated images do not, as The Verge documented in September 2026.