YouTube has launched a new “likeness detection” tool to combat the increasing threat of AI-generated deepfake videos, particularly affecting politicians, journalists, and public figures. This AI-driven technology scans uploaded videos for facial similarities, helping users identify potentially manipulated content that misrepresents them. Initially available to those in the YouTube Partner Program, the system is expanding access to more vulnerable groups amid rising concerns about misinformation and reputational harm associated with deepfakes during critical public events. Users must verify their identity to participate, and if a suspected deepfake is detected, they can review the video and request removal. YouTube emphasizes that submitted identity data will not be used to train AI models. As AI technology evolves, the company’s efforts to detect both facial and voice impersonations are crucial for preserving trust and authenticity in online content. This proactive approach highlights YouTube’s commitment to safeguarding its community from identity misuse and misinformation.
Source link
Share
Read more