Google has rolled out its latest video generation model, Veo 2, to users of Gemini Advanced, allowing the creation and sharing of AI-generated videos for the first time.
Starting today, the feature will be available globally, with support for multiple languages including English and Arabic. It enables users to produce short videos from text prompts through the Gemini platform, accessible via mobile or web.
Veo 2, which was first announced in 2024, has been designed to generate high-resolution video clips with cinematic detail. According to Google, the model exhibits improved understanding of real-world physics and human motion, resulting in more fluid movement, realistic scenes and enhanced visual accuracy.
Each generated video is eight seconds long, rendered in 720p resolution and delivered as an MP4 file in a 16:9 landscape format. The feature integrates directly within the Gemini interface and outputs videos embedded with a digital watermark, known as SynthID, to indicate that they have been created using artificial intelligence.
The development forms part of Google’s broader push to expand the creative applications of generative AI tools. While similar technologies have drawn scrutiny over their potential misuse, the inclusion of SynthID is aimed at enhancing transparency and responsible use.
Veo 2’s introduction to Gemini Advanced is expected to be the first in a series of updates bringing more AI-powered capabilities to Google’s suite of tools.
