One Disclosure, Five Destinations: A Cross-Platform Labelling Checklist
YouTube, TikTok, Instagram, Facebook, and Pinterest each label AI content differently. A working checklist for one asset published across all five.
"Did I disclose this is AI" isn't one checkbox when your content lands on five platforms — it's five different mechanisms, five different trigger conditions, and five implementations that don't talk to each other. Publish the same disclosed AI content across YouTube, TikTok, Instagram, Facebook, and Pinterest, and you'll interact with five distinct systems even though the underlying question — did a machine make or meaningfully alter this — never actually changes.
Five mechanisms, side by side
| Platform | Mechanism | What triggers it | Who applies it |
|---|---|---|---|
| YouTube | "Altered content" checkbox at upload; label appears in the expanded description | Realistic altered or synthetic content | Self-disclosed at upload |
| TikTok | Visible on-video AI label | Realistic AI-generated or significantly AI-edited people, audio, or scenes | Self-disclosed, or auto-applied when C2PA metadata is detected |
| New, opt-in profile-level "AI Creator" label: "This profile posts content that was generated or modified with AI" | Creator opts in at the account level | Self-disclosed only, voluntary | |
| "AI info" / "Made with AI" label at the post level | Photorealistic AI-generated or manipulated media | Metadata detection, classifiers, or self-disclosure | |
| "AI modified" indicator on closeup Pin view | Detected AI generation or modification | Metadata analysis and classifiers, with an appeals process |
The distinction that actually matters: self-declared versus auto-detected
The label-or-no-label question hides a more important split. TikTok's own labeling system applies automatically when it detects AI metadata, even if a creator never self-discloses — the label lands whether you check a box or not. Pinterest and Meta's "AI info" system work the same way, combining metadata analysis with content classifiers that can flag AI origin independent of creator input. YouTube's disclosure, by contrast, is self-declared at upload with no stated auto-detection layer, and Instagram's new profile-level label, announced in May 2026, is voluntary only.
That split has a real practical consequence. On auto-detecting platforms, skipping disclosure doesn't actually avoid a label — it just hands the decision to a classifier, which can get it wrong in either direction, and forfeits your chance to frame the disclosure in your own words. On self-declare-only platforms, skipping disclosure genuinely means no label appears, at least until enforcement catches up — a different, more exposed risk profile than a false sense of "I didn't check the box so nothing happened."
A second split worth knowing: profile-level versus content-level
Instagram's new test is unusual on this list specifically because it operates at the profile level rather than per post — a single, standing statement about an entire account rather than a toggle you set on each upload. Every other platform here labels content item by item. That's a meaningfully different disclosure UX, and it means an Instagram creator's choice is really "do I want this label visible everywhere I post," not "does this specific piece need it" — a broader commitment than the other four platforms ask for.
What the labels demonstrably do not do to reach
Here's the sharpest, most directly sourced point on this list: TikTok states plainly in its own newsroom announcement that "adding the label to content won't affect the distribution of a video." That's a first-party statement decoupling the disclosure mechanism from the recommendation algorithm, at least as officially framed.
The pattern holds, more cautiously, across the rest. None of the five platforms' own documentation frames its label as a reach or distribution penalty. Instagram's is explicitly opt-in and pitched as a trust-building option, not a compliance gate. Pinterest pairs its label with an appeals process specifically because the company acknowledges its classifiers aren't perfect — implying the goal is accurate labeling, not suppression. Facebook's "AI info" framing reads as descriptive rather than punitive.
Worth being precise about what that does and doesn't guarantee. None of this is a public commitment about how any platform's recommendation algorithm actually weighs labeled content today, since ranking mechanics are opaque and change constantly on every one of these platforms. The honest claim is narrower than "labels never hurt you": these platforms don't publicly document or frame their disclosure mechanisms as reach penalties, which is different from a guarantee that labeling carries zero downstream algorithmic effect. Disclose anyway — the downside of a false compliance shortcut is bigger than a risk none of these platforms admit to having built.
A practical five-platform checklist
Before the same piece of disclosed AI content goes out across all five:
- YouTube — check the "altered content" box at upload if the content is realistic and altered or synthetic; confirm the label shows in your expanded description before publishing.
- TikTok — self-apply the AI label rather than relying on auto-detection, since you control the framing that way; remember the label is permanent once applied and detection can catch unlabeled uploads regardless.
- Instagram — decide deliberately whether the new profile-level "AI Creator" label fits your account as a standing statement, separate from any per-post decision; it's not yet a universal requirement, but it's worth having a position on before it is.
- Facebook — expect an "AI info" label to appear on photorealistic AI content whether or not you self-disclose, since detection and classifiers run independent of your input; self-disclosing lets you control the framing rather than leaving it to a classifier.
- Pinterest — same logic: detection runs globally, and there's an appeals path if your content gets mislabeled, worth knowing about in advance rather than discovering it after a dispute.
Consult our AI content label and synthetic media disclosure glossary entries for the underlying definitions each platform's policy is built on before you finalize a cross-platform disclosure standard for your own content.
A Versely walkthrough: one caption, five native settings
Versely's post_to_social_media tool pushes one piece of content across multiple connected platforms in a single call — but the AI-disclosure toggle on each platform is a native, platform-side control, not something a scheduling tool can set on your behalf. Treat the two as separate layers rather than assuming one covers the other.
The workflow that actually respects both: ask the agent to check your connected accounts with get_social_accounts, then draft one caption that states the disclosure in plain language — something like "Made with AI" — directly in the text. That caption travels identically to every destination in a single post_to_social_media call, giving you one universal, human-readable disclosure that doesn't depend on any platform's native toggle existing or working correctly. Then handle each platform's own mechanism separately where it exists — YouTube's checkbox, TikTok's label, Instagram's opt-in — as a manual step per platform's own upload flow, since those settings live outside what a cross-platform posting call can reach. The caption is your universal fallback; the native toggles are the platform-specific layer on top of it.
FAQ
Do I need to disclose the same way on every platform?
No — each platform has its own mechanism, trigger condition, and detection method. YouTube and Instagram's new label are self-declared only; TikTok, Facebook, and Pinterest combine self-disclosure with automatic detection via metadata and classifiers. A single universal disclosure statement in your caption is a reasonable baseline, but each platform's native toggle still needs handling separately where one exists.
Does adding an AI disclosure label hurt my reach or distribution?
TikTok states directly that its label doesn't affect distribution. The other platforms on this list don't publicly frame their labels as reach penalties either, though none of them make an explicit public commitment about algorithmic treatment of labeled content. Treat disclosure as the safer choice regardless — the risk of skipping it outweighs an unconfirmed downside none of these platforms admit to.
What's different about Instagram's new AI Creator label compared to older AI labels?
It's profile-level rather than per-post — a standing statement on your account rather than a toggle you set on each upload. It's also entirely voluntary, distinct from Meta's older, narrower "AI info" labels that can be applied per post through detection.
If I don't self-disclose, will a platform label my content anyway?
On TikTok, Facebook, and Pinterest, yes — all three combine metadata detection and classifiers that can apply a label independent of your input. YouTube and Instagram's new label currently rely on self-disclosure, with no stated auto-detection layer for either.
Can Versely automatically set each platform's native AI disclosure toggle when I schedule a post?
No — post_to_social_media posts your content and caption across connected platforms in one call, but each platform's native disclosure toggle is a separate, platform-side setting outside what a cross-platform posting call can control. Put a plain-language disclosure in your caption as a universal baseline, and set each platform's own toggle manually where one exists.
Five mechanisms, one underlying question. Draft the disclosed caption once and schedule it across your connected accounts with Versely's posting and scheduling tools, then close the loop on each platform's native toggle before it goes live.