Your Digital Twin Just Minted Its First NFT: YouTube's AI Avatar Feature Is Live

Google is dropping a fresh AI feature for YouTube Shorts that lets creators spawn short videos using a digital avatar of themselves. Dubbed "Make a video with my avatar," the tool is currently deploying through the YouTube app and YouTube Create. Think of it as a non-fungible version of yourself—minus the gas fees.

The feature started rolling out on Wednesday and should hit most users within the coming days. It's powered by Google's Veo 3.1 video model and is currently only available in Shorts and the YouTube Create app. Because apparently, even AI needs a home venue.

"The avatar feature gives users an easier way to include themselves safely and securely in videos, building on our existing creative tools," a YouTube spokesperson told Decrypt. Safety and security are the buzzwords here, which in tech-speak usually means "we've thought about liability for at least fifteen minutes."

Each prompt-based clip can run up to about eight seconds, though users can combine clips to produce longer videos. Eight seconds to condense your entire personality into a looping avatar—this is basically speedrunning self-expression. The feature is available to users who own a YouTube channel and are at least 18 years old. Sorry, toddlers—you'll have to wait for Web3 childhood.

At launch, it's rolling out to mobile users globally outside of Europe, with broader availability expected in the coming days. Only the account holder can use their avatar to generate videos, which can be deleted at any time, though previously created videos remain online unless manually removed. So your digital twin can be deleted, but its content lives on like a blockchain transaction you regret.

The clips will also include AI disclosures and digital watermarks to indicate that the content was generated using artificial intelligence. In the age of "was this real or AI," at least they're leaving a receipt.

The news comes as generative video tools are spreading across the tech industry, with companies including Synthesia, ElevenLabs, and HeyGen offering platforms that create videos using AI presenters. The rise of realistic AI videos has also raised concerns about deepfakes, prompting companies to add labels and watermarks to indicate when content is AI-generated. It's the wild west out there, but at least everyone's wearing digital name tags.

However, bringing realistic video AI generation to the masses comes with a hefty cost. In March, OpenAI shut down its Sora video app after only six months. While the company said it was refining its focus on developing AGI, Sora reportedly cost the company $15 million per day to run. That's roughly $450 million per month to let people generate slightly weird videos of cats juggling. At least the compute bills make sense now.

The rollout fits into YouTube's broader push to expand AI tools for creators. In a January letter outlining the platform's priorities for 2026, CEO Neal Mohan said the company plans to expand AI-powered