How to disclose AI content on Instagram: turn on the 'made with AI' label on posts, Reels, and Stories, avoid the reach penalty, and appeal a wrong flag.
Last verified · 2026-09-05 · by Moe Ameen
Instagram won't reliably label your AI content for you — its detector reads file metadata rather than pixels, so it tags real photos that only touched a background-remover while genuinely synthetic images slip through once that metadata is stripped. That makes self-disclosure the part you actually control, and Instagram gives you a switch to set it: an in-app toggle on feed posts, Reels, and Stories that marks content as made with AI.
This walkthrough shows exactly where that toggle lives, which content actually needs it (the realism test Meta and every other platform converge on), how the per-post "AI info" label differs from the account-level "AI-generated profile" label that now carries a reach penalty for undisclosed synthetic personas, what to do when a real photo gets a wrong flag, and how to keep the disclosure consistent when the same clip also ships to TikTok and YouTube. For the policy background, the [Instagram AI-content detection and labeling guide](/guides/instagram-ai-content-detection-and-labeling) is the deep dive; this is the do-it version.
This is platform-policy guidance, not legal advice. Beyond Instagram's terms, the EU AI Act's Article 50 imposes a legal disclosure obligation for synthetic content that reaches EU users, and AI content used in advertising or endorsements can fall under advertising and consumer-protection rules (in the US, the FTC's deception and endorsement standards). Realistic AI that could mislead an audience about a person, product, or event carries the highest disclosure risk regardless of which platform it runs on.
Setting the AI toggle once is easy; keeping it right across a week of Instagram posts, Reels, and Stories that also fan out to other platforms is where disclosure quietly breaks. That daily operation is where Kompozy fits. Kompozy is a full AI content generation and multi-platform publishing engine, and its edge here is simple: because you generate the asset inside the engine, you already know whether it is realistic synthetic media — you never have to defer that call to Instagram's unreliable metadata reader. A [Persona Short](/glossary/persona-shorts) or Persona HeyGen avatar clip is photorealistic AI and gets the label; a clip cut from your own real footage, a quote graphic, or a design carousel does not. That determination travels with the asset instead of being re-guessed at upload.
Where that becomes an operational win is the review gate. From one [Persona Brief](/glossary/persona-brief), Kompozy produces a batch of on-brand content, and [Autopilot](/glossary/autopilot) schedules it across eight social platforms plus blog and email — but every post passes through a per-post review step before it ships. That step is the single checkpoint where you confirm the honest disclosure once and know it carries: you never mark an item as AI on the Instagram upload and forget it on the TikTok and YouTube cross-posts, and you never let separate detectors reach different verdicts on the same clip. The honest limit: Kompozy does not flip Instagram's Advanced-settings toggle for you or file your Account Status appeal — those actions live in your Instagram account and stay yours. What it removes is the reason disclosure drifts at scale, by keeping every output tied to a known format and a single review pass. Creator ($49/mo for 2,500 credits) fits a solo creator running one disclosed Instagram presence; Pro ($299/mo for 18,000 credits) suits a brand publishing AI-assisted content across every platform on a schedule; Enterprise is custom for agencies keeping disclosure consistent across many client accounts.
On the final screen before you publish a feed post or Reel, scroll to "Advanced settings" near the bottom and turn on "Label this content as made with AI." For a Story, tap the three-dot menu after choosing your content and toggle the AI label on. Instagram may also apply the label automatically if it detects AI signals in your file, but self-disclosing is more reliable than depending on that detection.
Yes. Your disclosure duty is separate from Instagram's detection. Meta requires labeling realistic AI-generated or substantially altered media and warns of penalties for not labeling content that needs it, and the EU AI Act adds a legal obligation for content reaching EU users. The detector failing to catch something does not remove your responsibility to disclose realistic AI you actually made.
Because Meta detects AI by reading file metadata, not the image itself. Editing tools like Photoshop's generative fill or Canva's background remover write C2PA Content Credentials or IPTC "AI-generated" fields into the exported file, and Instagram reads those and applies the "AI info" label. It can't tell a dust-spot removal from a full generation, so lightly edited real photos routinely get flagged.
Use the Account Status dashboard to review and contest the label rather than stripping your file's metadata. Scrubbing C2PA and IPTC data can clear the flag but destroys the provenance that proves your content is genuine and puts you in the same bucket as deliberate fakes. Keep originals and unfiltered footage as evidence when you appeal.
The "AI info" label is per-post — it marks an individual image or video as made or edited with AI. The "AI-generated profile" label is account-level — it declares that the person a whole profile is built around is synthetic, and since August 31, 2026 accounts that hide it may see reduced reach. A real creator's single AI-edited post can get an "AI info" tag without ever needing the profile label.