YouTube To Automatically Detect, Label AI-Generated Videos (variety.com) 36
YouTube will begin automatically labeling videos when its systems detect "significant" photorealistic AI use, while also making AI-content disclosures more visible below long-form videos and directly on Shorts. "We've heard consistently from our community that they value transparency when it comes to generative AI content," YouTube said in a blog post. "These changes are designed to balance transparency with creator control." Variety reports: Under YouTube's guidelines, creators will still be required to manually disclose when they use realistic AI. But starting this week, it also will roll out a new internal system to help identify AI-generated content. "If a creator doesn't specify whether or not they used AI, but our systems detect significant photorealistic AI use, we will now automatically apply a label," YouTube said.
YouTube creators who believe their content was incorrectly flagged as AI-generated can modify the disclosure status using the YouTube Studio tool. However, according to YouTube, the AI labels will "remain permanent" in some cases, including for content created using YouTube's own AI tools (such as Veo or Dream Screen) and for content that contains C2PA metadata (based on standards from the Coalition for Content Provenance and Authenticity) that indicates it was fully AI-generated.
In addition, YouTube is moving the disclosure label for photorealistic and meaningfully AI-altered or AI-generated content to a more prominent position. Until now, YouTube labeled AI content in a video's expanded description. Going forward, for long-form videos, the AI label will now appear directly below the video player and above the description. For YouTube Shorts, the label will appear as an overlay on the video itself. "The goal here is context at a glance. If it looks real but was made with AI, viewers will know immediately," said Rene Ritchie, YouTube head of editorial and creator liaison. He added that the AI labels alone "do not affect how our videos are recommended or whether they can earn money. This is purely about giving viewers the right information at the right time."
YouTube creators who believe their content was incorrectly flagged as AI-generated can modify the disclosure status using the YouTube Studio tool. However, according to YouTube, the AI labels will "remain permanent" in some cases, including for content created using YouTube's own AI tools (such as Veo or Dream Screen) and for content that contains C2PA metadata (based on standards from the Coalition for Content Provenance and Authenticity) that indicates it was fully AI-generated.
In addition, YouTube is moving the disclosure label for photorealistic and meaningfully AI-altered or AI-generated content to a more prominent position. Until now, YouTube labeled AI content in a video's expanded description. Going forward, for long-form videos, the AI label will now appear directly below the video player and above the description. For YouTube Shorts, the label will appear as an overlay on the video itself. "The goal here is context at a glance. If it looks real but was made with AI, viewers will know immediately," said Rene Ritchie, YouTube head of editorial and creator liaison. He added that the AI labels alone "do not affect how our videos are recommended or whether they can earn money. This is purely about giving viewers the right information at the right time."
Alternate headline (Score:5, Insightful)
Youtube creates vast adversarial network to make fake video undectable
Re: (Score:2)
A contest between AI video creators and people using AI to detect AI was inevitable.
Re: Alternate headline (Score:2)
We all win when we step away from the YouTubes because of the world wide slop fest. Making money shaving away hours of our life will become less viable the less we engage with it.
Re: (Score:2)
And lawsuits over false positives in 5 4 3 2 ...
Also youtube needs a second URL (Score:2)
www-no-ai-videos . youtube . com
And also require the "try this one trick" advertisements to clearly state what the 1 trick is within the first 10 seconds of the video.
Not enough, by far. (Score:5, Insightful)
They need to block the AI slop entirely and ban the frauds posting it.
Re: (Score:2)
including for content created using YouTube's own AI tools (such as Veo or Dream Screen)
Good luck wishing for that, when it's already making them money.
Re: (Score:2)
Re: (Score:2)
As much as I thumbs-up and thumbs-down content, it still has opinions on what it thinks I want to see that are entirely wrong. The feedback loop is short to get new content into the feed. Getting the slop out takes a lot longer.
Re: (Score:1)
They need to block the AI slop entirely and ban the frauds posting it.
They're not even going to be able to detect it reliably, nor avoid false positives. It's also not all fraud. Also as sibling points out, ALL content is profitable for them, AI or not.
Prompt Injection (Score:2)
So they just have to incorporate the phrase "not produced with AI" somewhere in the video and it'll be undetectable.
Self-serving process (Score:5, Insightful)
Re: (Score:2)
Re: (Score:2)
Re: (Score:3)
Probably already is. I saw a story several months ago that 90% of the new music submissions to Spotify is AI slop.
There's new "bands" that, within a month of "forming" have 80+ songs submitted under their names. It's easy to flood the zone when each new song or album is just a few words and a few minutes of machine time. Much harder for us lazy human musicians to keep up in comparison.
Re: (Score:2)
Remember that until recently most people were watching that dreg.
Re: (Score:2)
Just as they flag up copyright material, they'll flag up humans talking to each other as "AI". There'll be no way to talk to a human about the mistaken flag, there'll be no human verification of your appeal and your account will be closed without human oversight.
Honestly, self-serving or none, it's going to get harder to actually put any sort of content onto YT. On aggregate, that's probably a good thing because there's a lot of dross, but it's going to piss off the "influencers" no end.
Re: (Score:1)
Movies are not real ...
Who cares if a non real story is story tale-ed by an AI created movie or a movie that costed a fortune to make and the people involved got paid bread crumbs?
Thinking about Avatar, Lord of the Rings of Game of Thrones.
Books, media, etc (Score:2)
No Adverts
No tracking
Physical media can not later be denied access to it
So much less "noise', BS, Slop, Distraction, etc etc etc...
And MUCH MUCH healthier for the brain.
The price for free (Score:2)
Great! What About A Filter? (Score:4, Interesting)
Can I also have a filter to exclude all those AI tagged videos?
Re: Great! What About A Filter? (Score:2)
Not trying very hard (Score:4, Insightful)
Why aren't they putting this information alongside the thumbnail so we can totally skip AI content if we want to. Only finding out once you've clicked on the video and the player has loaded is stupid -- being both a waste of the viewer's time and bandwidth.
Re: Not trying very hard (Score:3)
Full Auto (Score:3)
The best thing about automated moderation is that it's never wrong and is quick and easy to appeal.
This is necessary (Score:4, Insightful)
I sometimes watch AI generated videos and find them amusing. I don't hate AI videos, I just want honesty.
Someone who honestly uses a tool should be honest about it.
Someone who hides their tool use is probably up to no good.
Re: (Score:2)
I agree that just labeling the AI-generated stuff *should* be enough, but I wonder if it really is enough. Yesterday someone sent me a video of Brian Cox describing some concept. Right in the text of the original post it said, "This video features an AI-generated voice for storytelling and educational purposes. It is not the real Brian Cox." So, fully disclosed, but it didn't stop people from forwarding it.
When I complained, the person who sent it to me said that the idea's interesting regardless of where
Start by labelling every video with sparkly 'snow' (Score:2)
Give us a filter or a toggle (Score:2)