The Twins Made This Hyper-Realistic Deepfake Video of Me in Just a Few Minutes.
During its major presentation at I/O 2026, Google announced Gemini Omni , an AI model capable of creating videos from virtually anything. In the weeks since its announcement, paid subscribers have been able to use a combination of text, images, documents, and video clips to create new AI-powered videos. Now, Google is launching a new feature, Omni , that lets you create hyper-realistic avatars of yourself. While it’s currently limited to videos (AI-powered profile photos aren’t available yet), it does a surprisingly good job of creating talking-head videos using just a couple of selfies as a reference—outperforming the short-lived app Sora . With Omni, Google has released a tool that lets you create deepfake videos of yourself (and only yourself) in just a few minutes. Are we ready?
With Gemini Omni, you can create your own deepfakes today.
Luckily, you can’t turn a selfie into a deepfake video. You need to go through Google’s verification process, which involves scanning your face. Once you’ve accessed this Omni feature, go to the Gemini sidebar and click “Video.” You’ll see a pop-up window to create your own avatar. If you don’t see it, click the Plus button in the Gemini prompt, then click “Avatar.” Gemini will redirect you to Google’s website to scan your face. Here, you take a few selfies and move your face from side to side. Then, say a few numbers out loud, and you’re done. You don’t even need to say any words or sentences. Once the process is complete, your avatar is ready, and you can return to Gemini.
Now, return to the Videos section and enter “@[your name]” to use your avatar in your video. The video will take a few minutes to create, and you’ll receive a notification when it’s ready. You can play the video in the app, save it to your gallery, or share it via a link or along with the video file itself. There are some additional restrictions: the new Gemini avatar feature is only available to AI Pro or AI Ultra subscribers using personal accounts. Furthermore, this feature is only available to users over 18 and is not available in the European Economic Area, Switzerland, or the UK. It can currently only generate audio in English. Each video created includes a prominent Gemini watermark and is also encoded with SynthID, Google’s new industry standard for seamlessly watermarking AI-generated videos and images using metadata that can be tracked even if the video is cropped.
Would you use deepfake to create an image of yourself?
After a full day of experimenting with this feature and creating a series of videos, it became clear that this tool is astonishingly good. We are not ready for an era where deepfake technology is so widespread and easily accessible. You can see this in the example below: I asked Gemini to create a video of me reviewing the iPhone 17 Pro and asked it to include a specific phrase. And it did.
Of course, there are limitations, and I can find plenty of flaws too. I asked for an iPhone 17 Pro review, but the video used an iPhone 16 Pro. At one point, something spontaneously appears over one of the many headphones in the background. And while the sound resembles my real voice, there’s no rhythm. No personality, just a monotonous delivery. And this extends to the videos themselves. It’s me, a human, speaking the words, but the videos feel rather lifeless. They’re too clean, too sharp, and my hair doesn’t always look good . If you know what to look for, it’s not hard to tell that this is ultimately an AI-generated deepfake (if the Gemini watermark and the “AI” label on YouTube didn’t give it away). But the question is, how long will this last? And who will check it thoroughly enough?
Currently, videos are limited to 10 seconds in length and can’t be edited or modified in any way, so the potential for abuse is still quite small. However, Google says they are currently collecting feedback. The company states, “Regarding video editing to change audio and speech, we’re still testing this feature and better understanding how we can responsibly deliver this capability to users,” so an interactive editing interface is possible in the future. Soon, users will be able to insert themselves into videos, which they can customize to their preferences before sharing them online. We might not be ready for that yet.