YouTube has announced a new feature that gives creators more control over whether to allow third-party companies to use their videos for artificial intelligence model training. This move aims to balance the development of AI and the rights of creators, and bring new value to creators in the AI era. The feature is turned off by default, and creators can choose which companies are allowed to use their videos, or ban it entirely. The move responds to recent reports of large technology companies using YouTube videos to train AI models without authorization, and reflects YouTube's emphasis on creators' rights.
YouTube recently announced a new feature that will allow creators to choose whether to allow third-party companies to use their videos to train artificial intelligence (AI) models.
The default setting for this option is off, meaning creators don’t need to take any action if they don’t want third-party companies to scrape their videos for AI training. However, if a creator wishes to do so, YouTube will provide a clear option.
Rob, a member of the YouTube team, said in a support post: "We see this as an important first step in supporting creators and helping them realize new value for their YouTube content in the AI era." He also mentioned that YouTube will Continue to explore new ways to facilitate collaboration between creators and third-party companies, including options to grant access to content.
This setting will be gradually rolled out to YouTube Studio over the next few days, but unauthorized scraping is still prohibited. Creators can choose from a list of third-party companies to allow them to use their videos for training, or choose to allow all third-party companies to use their videos for training, according to a separate support page. According to TechCrunch, the companies initially listed include: AI21Labs, Adobe, Amazon, Anthropic, Apple, ByteDance, Cohere, IBM, Meta, Microsoft, Nvidia, OpenAI, Perplexity, Pika Labs, Runway, Stability AI, and xAI.
YouTube spokesman Jack Malon confirmed the accuracy of TechCrunch's list of companies to The Verge. He said the companies were chosen because they were building generative AI models and the potential partnerships with creators made sense.
The announcement follows reports that large companies, including OpenAI, Apple, and Anthropic, are using content and data sets scraped from YouTube to train AI models. Google itself already uses data from YouTube to help train its AI tools. "As we've done for years, we use content uploaded to YouTube to improve the product experience for creators and viewers on YouTube and Google, including through machine learning," the company said in September when it announced the feature was being developed. and AI applications to achieve this.” They emphasized that this approach is consistent with the terms agreed by the creator.
Highlights:
YouTube has launched a new feature that allows creators to choose to allow third-party companies to use their videos for AI training.
The default setting is off, and creators need to actively choose to allow use.
Allowed third-party companies include many well-known AI companies, such as OpenAI, Apple, and Microsoft.
All in all, YouTube’s move aims to better protect the rights of creators and promote cooperation between creators and AI companies to seek new business models and possibilities in the AI era. This move will continue to affect the development of the YouTube ecosystem and the AI industry.