YouTube's Role in AI Training: New Challenges for Copyrights
Google is actively utilizing content from its vast YouTube library to train its artificial intelligence models, including Gemini and the new video and audio generator Veo 3. CNBC reports this based on its own sources.
One source revealed that a selection from a catalog of 20 billion videos is used for training. Google confirmed this information but clarified that it only pertains to a portion of the content under agreements with creators and media companies.
A YouTube representative explained that the company has always used its own content to enhance its services — the emergence of generative AI has not changed this practice. "We understand the importance of guarantees, which is why we have developed reliable protection mechanisms for creators," the company stated.
However, experts express concerns about the implications for copyright. They believe that using others' videos for AI training without the creators' knowledge could lead to a crisis in intellectual property. While YouTube claims to have previously communicated this, most creators were unaware that their content was being used for training.
Google does not disclose how many videos were used to train the models. But even if it concerns 1% of the library, that amounts to over 2.3 billion minutes of content — 40 times more than competitors.
By uploading videos, creators grant YouTube broad permission to use that content. However, they do not have the option to refuse the use of their videos for training Google's models.
Representatives from companies protecting digital rights believe that the years of work by creators are being utilized for AI development without compensation or even notification. For example, the company Vermillio has developed a service called Trace ID that identifies similarities between AI-generated videos and original content. In some cases, the match reached over 90%.
Some creators are not against their content being used for training, viewing the new tools as opportunities for experimentation. However, most believe the situation is opaque and requires clearer rules.
YouTube has even entered into an agreement with Creative Artists Agency to develop a system for managing AI content that mimics famous personalities. However, the mechanisms for removing or tracking similar content still need improvement.
Meanwhile, calls in the U.S. are already being made to provide authors with legal protection that would allow them to control the use of their creativity in the generative AI landscape.
Recently, Google also changed its internal content moderation rules on YouTube — now videos that partially violate the rules can remain online if deemed socially important.
