The platform Youtube unveiled new transparency and accountability measures for generative AI on Tuesday. Viewers, creators and artists will be able to request the removal of videos generated by the technology.
YouTube announced on Tuesday, November 14, major changes in its approach to generative artificial intelligence, highlighting its potential to revolutionize creativity on the video service, but also its many dangers, both for creators, viewers and artists. The platform seems determined to strike a balance between innovation and protection of the YouTube community, an issue where it knows it is particularly expected.
Creators must indicate that their content was created synthetically
If generative AI promises to redefine the user experience on YouTube, the impact of the platform now requires regulating this technology to avoid abuse. The service plans to introduce information requirements for creators who use generative AI. For example, they will have to indicate whether their content has been modified or created synthetically, particularly for videos touching on sensitive subjects such as elections, ongoing conflicts, public health crises, or public figures.
In the coming months, users will be able to report manipulated, artificially created, or AI-generated content that violates privacy. YouTube will take various factors into account when processing these reports, including whether they are parodies, satire, or whether the person concerned can be directly identified.
Music industry partners will also have the ability to request the removal of AI-generated audio content imitating an artist’s voice. To process requests, YouTube promises to base itself on various criteria. The platform will ask itself, for example, whether it is a news report, an analysis or a criticism linked to the voices created by artificial intelligence.
A very valuable AI in the moderation of YouTube videos
YouTube will also intensify the use of AI to facilitate content moderation, a task that has already largely fallen to it for several years, to support the 20,000 human moderators working for the Mountain View firm around the world. This technology, integrated into existing moderation systems, accelerates the identification and removal of harmful content. Generative AI helps to rapidly expand the database on which AI classifiers are trained.
Google’s video service takes a meticulous approach to developing AI tools for creators. It takes care to integrate protection measures to avoid the generation of shocking or illegal content. Dedicated teams actively test these tools against adversarial attacks and potential threats, improving protections month after month.
In any case, YouTube recognizes that it is at the start of a major transformation thanks to generative AI. The platform wants to find a balance between the opportunities offered by this technology and its responsibility towards the community. Close collaboration with videographers, artists, and other players in the creative industry is now envisaged in the long term to shape a future beneficial to all.
2