Following CEO Neal Mohan's January announcement, YouTube has rolled out an AI avatar tool letting creators generate Shorts videos replicating their likeness and voice. The feature emphasizes user control, safety, and clear AI labeling on generated content.
YouTube creators can now create a digital twin via the platform's AI avatar tool in Shorts, without needing to appear on camera. Setup requires a live selfie and voice recording in the YouTube or YouTube Create app's AI Playground, followed by previewing a photorealistic avatar. Creators can then generate videos up to eight seconds long using text prompts or by remixing eligible Shorts with 'Reimagine' options.
Google advises account owners over 18 to use proper lighting, a quiet space, and center their face during setup. Only the creator controls their avatar, which can be deleted anytime (permanently removing data), with unused avatars auto-deleting after three years. Generated videos feature AI disclosures, SynthID and C2PA watermarks, and labels.
A Google spokesperson noted: 'Only the user themselves can create and control their avatar. No one else can use it.' The tool enables 'a digital version of yourself so you can generate videos that look and sound like you, safely and securely.' Creators can also limit video remixing. As Mohan stated earlier, 'AI will remain a tool for expression, not a replacement.' This builds on recent AI enhancements like video upscaling and editing tools.
