Every guide to AI labels tells you the same thing: flip the “AI-generated” switch before you hit publish. That switch lives inside each app’s own posting screen. A post you schedule never passes through that screen, so unless the label travels with the post, your AI video goes out unlabeled. On YouTube, that’s more than a missing tag: under YouTube’s own policy, creators who keep skipping the disclosure can lose videos or get suspended from the Partner Program.
A scheduled post only carries an AI label if the network’s API accepts one and your scheduling tool sends it. TikTok, YouTube, X, and Instagram accept an AI label through their APIs, and Fedica sends it to all four. Facebook Pages, Threads, LinkedIn, Bluesky, and Mastodon don’t offer an AI-label field for scheduled posts, so on those you disclose in the caption.
What an AI Content Label Is
An AI content label is a platform’s own marker on a post saying it contains AI-generated or AI-altered media. It’s set by whoever publishes the post, and each network shows it its own way: “AI-generated content” on TikTok, “Made with AI” on X, “AI info” on Instagram and the rest of Meta’s apps, and an altered or synthetic content label on YouTube.
Some platforms also add a label themselves. YouTube says it may label content made with its own AI tools, content carrying C2PA metadata, or content its systems detect as AI-generated. Meta’s Transparency Center describes detecting “industry standard AI image indicators” across Facebook, Instagram, and Threads. C2PA, better known as Content Credentials, is an open standard for embedding a file’s origin and edit history inside the file. Self-labeling is still the part you control.
When a Post Actually Needs an AI Label
Not every AI-touched post needs one. YouTube’s help page on altered or synthetic content is the clearest rule of thumb, and it lines up with how the other platforms describe it. Disclose when realistic content:
- makes a real person appear to say or do something they didn’t do
- alters footage of a real event or place
- generates a realistic scene that didn’t actually occur
YouTube lists what doesn’t need disclosure too: beauty filters, color grading, captions, and “production assistance” like script generation. Clearly unrealistic content, such as an animated scene or a fantasy creature, doesn’t need it either.
So an AI-written caption on a real photo of your shop? No label. An AI-generated photo of your product on a beach it never visited? Label it.
Which Networks Accept an AI Label From a Scheduler
A scheduling tool can only set a label that the network’s API has a field for. Here’s where each network stands, from each platform’s developer documentation as of September 2026:
| Network | API field | Label people see | What it covers | Change it after publishing? |
|---|---|---|---|---|
| TikTok | is_aigc | “AI-generated content” | Videos and photo posts (in Fedica) | No, it’s set when the post publishes |
| YouTube | containsSyntheticMedia | Altered or synthetic content | Videos | Yes, the field can be updated later |
| X | made_with_ai | “Made with AI” | Posts with an image or video | No, it’s set when the post publishes |
is_ai_generated | “AI info” | Posts, Reels, Stories, and carousels as a whole | No, it’s set when the post publishes | |
| Facebook Pages, Threads, LinkedIn, Bluesky, Mastodon | No AI-label field found | None from a scheduler | Disclose in the caption | Not applicable |
Two details from the docs are worth knowing. X describes its field as a way to “disclose that the tweet contains AI-generated media,” so the label is about the image or video, not the text. Instagram’s field is “not available for carousel children,” which means a carousel gets one label for the whole thing, not one per slide.
Because TikTok, X, and Instagram fix the label at publish time, the decision belongs at the scheduling step. Getting it right before the post goes out saves you deleting and reposting later.
How to Label AI Content on a Scheduled Post in Fedica
Fedica’s AI-generated media toggle covers YouTube, X, Instagram, and TikTok. It sits in Custom Posts, where you set every other per-platform option, and it works the same for a TikTok photo post as for a video going out through Fedica’s YouTube scheduler. Here’s how it works:
- Open the Publish menu and choose Create Post.
- Select your platforms, for example TikTok, Instagram, and X, plus anything else in the campaign.
- Add your image or video, and its alt text. Fedica can generate alt text for you.
- Click Custom Posts and turn on the AI-generated media toggle for each platform that should carry the label.
- Click Schedule, or Post Now.
The toggle only appears once media is attached. A text-only post doesn’t show it, which matches how the labels work: they describe AI-generated media. On TikTok it works for both video and photo posts.
Because it’s set per platform, you can label the TikTok and Instagram versions of a campaign and leave it off where it doesn’t apply. It’s the same Custom Posts pattern Fedica uses to limit who can reply to scheduled posts on each network.
Networks Without an AI Label Field
Facebook Pages, Threads, LinkedIn, Bluesky, and Mastodon don’t offer an AI-label field in the APIs a scheduler posts through. That doesn’t mean you skip disclosure there. It means the disclosure goes in the post itself.
A short, plain line works best. “Image generated with AI” or “Made with AI” at the start or end of the caption is clear to anyone reading, and it’s the same wording the platforms use in their own labels.
LinkedIn has one extra layer. When an uploaded image carries C2PA metadata, LinkedIn shows a small “CR” Content Credentials badge that viewers can click to see where the file came from. The badge shows that provenance data exists, not necessarily that the image is AI-generated.
Fedica keeps C2PA metadata on the files you upload, so the credentials reach the platform intact. The one exception is when a platform doesn’t accept the file’s format and Fedica has to convert it, which can drop the metadata. Upload in a format the platform accepts, like JPEG or PNG, to keep it. Fedica can’t edit C2PA metadata, so if the credentials on a file are wrong, upload a replacement file.
What the EU AI Act Requires for AI Content
Since 2 August 2026, Article 50 of the EU AI Act has required disclosure of certain AI-generated content. For anyone posting to social media, two parts matter, according to the text of Article 50:
- Deepfakes. Deployers, meaning the people and businesses using an AI system professionally, who generate or manipulate image, audio, or video content that counts as a deepfake “shall disclose that the content has been artificially generated or manipulated.”
- Text on matters of public interest. AI-generated text published to inform the public on matters of public interest must be disclosed, unless it has “undergone a process of human review or editorial control” and someone holds editorial responsibility for it.
The disclosure has to be clear “at the latest at the time of the first interaction or exposure,” which on social media means on the post itself, not on a page somewhere else. This is a summary of the law’s text, not legal advice. If your posts carry legal risk, ask someone qualified.
Frequently Asked Questions
Do I need to label AI-written captions?
Usually not: YouTube lists script generation as production assistance that doesn’t need disclosure, and under the EU AI Act, text only needs a label when it informs the public on matters of public interest and nobody reviewed it.
Can I change the AI label after my post goes live?
On YouTube, yes, because the disclosure field can be updated after upload. On TikTok, X, and Instagram, the label is set when the post publishes, so decide before you schedule.
Why don’t I see the AI-generated media toggle on my post?
The toggle appears once you add media, and only for YouTube, X, Instagram, and TikTok.
Does using a scheduler or AI to plan posts mean I have to label them?
No. The label is about what the content shows, not the tools you used to plan, write, or schedule it. An AI-assisted schedule, a caption tweaked by AI, or a post queued a week ahead doesn’t need a label by itself. What triggers it is realistic AI-generated or AI-altered media.