Google has announced that users can now create videos using artificial intelligence through its Gemini chatbot and its recently launched Whisk experimental tool. The new feature is powered by Google's most advanced video creation model, Veo 2, which was introduced last year.
Developed by Google DeepMind, Veo 2 is now the foundation for a number of other video features, including Dream Screen for YouTube Shorts, which lets users create AI-generated clips to use as backgrounds for Shorts videos.
Gemini's latest update lets you convert text-based commands into 8-second animated videos at 720p resolution (16:9 aspect ratio), exported as MP4 files. To prevent GPU overload, Google will only let you create a limited number of videos per month.
In an official blog post, Google said the new feature is rolling out to Gemini Advanced users globally on web and mobile in all supported languages. You can create videos with Gemini by clicking on the models drop-down menu and selecting Veo 2 from the list.
You can then describe the scene you want to create in a text command and ask the chatbot to create a short video. By entering more detailed descriptions, you can have more control over how the video looks. Google offers easy options to share the created video to platforms like YouTube Shorts and TikTok.
Google is adding a new video creation feature called “Whisk Animate” to its experimental Labs photo-to-video tool it introduced last year. Available to Google One AI Premium subscribers, the feature lets you turn your photos into 8-second videos using the Veo 2.
The search giant says all videos created with Veo 2 are marked with SynthID, a digital watermark embedded in each frame to indicate the video was created using AI.
Note that the feature is in beta and may be limited in some regions.