YouTube has broadened its initiative aimed at combating deepfakes. The facial recognition technology, developed to safeguard against inauthentic videos, is now set to undergo trials involving journalists, government officials, and individuals running for political office.
Participants in this scheme will submit a brief video clip along with a photo of their identification document so that the neural network can familiarize itself with their unique facial characteristics. Subsequently, YouTube will automatically analyze uploaded content, searching for any videos generated by artificial intelligence that incorporate that specific likeness. Should questionable material be detected, the individual has the option to review it and request its removal.
Last year, this protective measure was made available to creators who are part of the YouTube Partner Program. Now, the company has opted to prioritize high-profile individuals.
However, immediate removal of content upon request is not guaranteed. YouTube emphasizes that every single complaint will be evaluated based on “existing policies.” If a video is determined to be a “parody, satire, or possesses other artistic merit,” the platform may decide to keep it accessible.
About the Author
