YouTube introduces automatic AI labels below the player and in Shorts
YouTube moved AI-use labels to more visible spots and says it will automatically label significant undisclosed AI use in photorealistic or meaningfully altered videos. Creators can appeal mistaken labels in Studio, so channels should review uploads for false positives and spam workarounds.
TL;DR
- YouTube moved its AI disclosure labels into the main viewing surface, with long-form labels now sitting below the player and Shorts labels appearing as an overlay, according to YouTube's policy summary and the Official blog post.
- Starting in May, YouTube says it will automatically label videos when its systems detect significant photorealistic AI use that a creator did not disclose, per YouTube's policy summary.
- Creators can change an incorrect disclosure in Studio, but YouTube says labels stay permanent for videos made with its own tools, including Veo and Dream Screen, as YouTube's policy summary notes and the YouTube Help explainer expands.
- The label by itself does not change recommendation or monetization eligibility, according to YouTube's policy summary and the official announcement.
- In the discussion roundup, the practical worries were familiar: false positives, weak appeals, and how easily spam channels might route around disclosure while still flooding feeds.
You can read the official announcement, check YouTube's "How this content was made" help page, and skim the HN thread where the conversation jumped straight to false positives and AI-slop evasion.
Label placement
Improving AI labels for viewers and creators
YouTube is updating its generative AI disclosure framework to improve transparency for viewers and simplify the process for creators. The platform is moving labels for photorealistic and meaningfully AI-altered content to more prominent positions: directly below the player for long-form videos and as an overlay on Shorts. Additionally, YouTube is introducing internal systems to automatically detect and label significant photorealistic AI use if a creator fails to disclose it manually. While creators remain in control and can update disclosure statuses in YouTube Studio if they believe a label was applied incorrectly, labels are permanent for content created with YouTube's own AI tools, such as Veo or Dream Screen. These labels do not impact video recommendations or monetization eligibility.
The visible change is simple. For photorealistic and meaningfully altered videos, the AI label moves below the player on standard uploads and onto the video itself for Shorts, per the Official blog post.
YouTube kept the lighter-touch case separate. Unrealistic, animated, or only slightly altered content can still leave the disclosure in the expanded description, according to the announcement and YouTube's older disclosure policy page.
Auto-labeling
YouTube to automatically label AI-generated videos
For creators, this is a YouTube policy/workflow update: AI disclosures become more visible, and YouTube may auto-label photorealistic AI-heavy uploads even when creators don’t self-disclose. The practical questions raised in the thread are about false positives, appealability, and how easy it will be for AI-slop channels to evade detection while still flooding recommendations.
The bigger policy shift is that YouTube is no longer relying only on self-reporting. The company says it will automatically apply a disclosure when it detects significant photorealistic AI use that was not manually declared, as described in the official post.
Creators still get an escape hatch. If a label was applied incorrectly, they can update the disclosure status in Studio, while the label alone does not affect recommendations or monetization, according to the official post.
A second path comes from provenance metadata. YouTube's help explainer says the platform can also surface disclosures from Content Credentials, with an "Info from" signer attached when secure metadata is present.
Permanent labels
Improving AI labels for viewers and creators
YouTube is updating its generative AI disclosure framework to improve transparency for viewers and simplify the process for creators. The platform is moving labels for photorealistic and meaningfully AI-altered content to more prominent positions: directly below the player for long-form videos and as an overlay on Shorts. Additionally, YouTube is introducing internal systems to automatically detect and label significant photorealistic AI use if a creator fails to disclose it manually. While creators remain in control and can update disclosure statuses in YouTube Studio if they believe a label was applied incorrectly, labels are permanent for content created with YouTube's own AI tools, such as Veo or Dream Screen. These labels do not impact video recommendations or monetization eligibility.
Some disclosures are not meant to be reversible. YouTube says labels remain permanent for videos created with its own AI tools, including Veo and Dream Screen, and for content carrying C2PA metadata that marks the whole video as fully generative AI, per the official post and the help explainer.
That creates three distinct routes into the label system:
- creator disclosure in Studio
- YouTube's own automatic detection
- attached Content Credentials metadata
False positives and slop
Discussion around YouTube to automatically label AI-generated videos
Thread discussion highlights: - zahlman on automatic detection and false positives: I can't wait for their detection to repeatedly get this completely wrong (as it does for many other things) and for innocent content creators to complain on social media about how their appeals get automatically dismissed by AI-powered bots. - CM30 on creator appeals and correction: The fact this status can be removed by the uploader certainly helps fix this issue, but then it feels like something any good conman will be able to work their way around really easily... But something like this is needed. - Willish42 on AI slop and creator transparency: I've been thinking for some time that it wouldn't be too hard to create a third-party browser extension to crowdsource detection of channels that use primarily AI-generated content... and hide them from suggestions or home feeds.
In the discussion roundup, the strongest reaction was not about transparency as a principle. It was about whether YouTube can label accurately enough to avoid hitting legitimate creators while still catching the channels people already describe as AI slop.
One commenter in zahlman's HN reply predicted misfires and botched appeals, while CM30's reply argued that creator-side correction helps but also makes evasion easier. A third comment, Willish42's suggestion, proposed a crowd-sourced browser extension to hide AI-heavy channels from recommendations altogether.
That last bit is the most revealing new workflow signal here. YouTube says the label itself will not affect distribution, but the thread immediately turned the label into a filtering primitive that viewers or third-party tools could use on their own.