Recent changes have seen YouTube crack down even harder on AI abuse by clarifying its policies on how content on its platform is monetized. To discourage creators from using AI and other tools to create low-quality content, YouTube’s policies describe three types of videos that fall under the broader category of “inauthentic content” that cannot be monetized:
YouTube already had a policy regarding AI content. The company announced last year that it would block creators’ ability to generate revenue from what it called “inauthentic content,” such as mass-produced videos and other types of repetitive content made easier to generate thanks to AI technology.
Rather, recent updates aim to add further nuance to existing policies.
The updated guidelines classify inauthentic content into three categories: Content that is generic, repetitive, or template-based, followed by content that is offensive or offensive, and content where AI personas are used to discuss sensitive topics such as health or finance.
While AI slop, like adult content, is something of a “you know it when you see it” situation for many, YouTube’s policies aim to be more specific about when the use of AI in content creation is and isn’t allowed for videos monetized through the YouTube Partner Program (YPP).
The program, which allows creators to earn money through ads and subscriptions, is the lifeblood of YouTube. It is not in the company’s financial interest to allow it to fill up with AI slop and other low-quality fare. This is especially important given that YouTube is in fierce competition with TV and other streamers for TV ad dollars. Google’s video platform now generates more ad revenue than rival streamers and has overtaken Netflix in average daily viewership worldwide.
As Matt Halprin, head of trust and safety at YouTube, explained in a Creator Insider video last week, the company’s goal with the policy revisions is to reduce references to content farming efforts, or low-quality videos that exist solely to generate revenue.
“With AI, people can actually create a lot of videos,” Halprin said. “These videos can be great and really boost creativity, allowing us to create a greater volume of high-quality content, which is what we want to encourage and provide at YPP.”
“But with the exact same new tools, you can quickly create a lot of very similar videos. They’re very generic, they don’t have a lot of storytelling, they don’t have a lot of creativity. So the same technology allows for really great things, but it also allows for things like a kind of content farming, and that’s something we don’t want to have in YPP,” Halprin added.
He said the first category includes repetitive content that changes little from video to video, such as AI, computer-generated imagery (CGI), or material that can be easily created with templates. This fills your channel with cookie-cutter videos. Halprin points out that this policy may also apply if the tutorial video is not original and reproduces content that is already popular on the platform.
The second category focuses on what YouTube calls “disgusting” content, content designed to manipulate distress or emotions in order to chase views. Halprin said this could include videos of animals in distress and then someone coming and rescuing them.
“We’ve heard from viewers that it’s not their cup of tea. They’re uncomfortable with it. They may not want to go back to that channel or that platform,” Halprin said.
Channels dedicated to this content will be removed from YPP, regardless of whether the content is AI-generated.
The third bucket more directly targets the use of AI personas, or AI-created representations of real people. YouTube says it does not encourage creators using AI personas to discuss more sensitive topics such as finance, legal issues, health care, or medical issues.
YouTube channels with too much of any of these three types of content will no longer be monetizable, Halprin said. This policy clarification will roll out on July 16th and will impact all YouTube Partner Program members.
If you buy through links in our articles, we may earn a small commission. This does not affect editorial independence.
