Google upgrades Veo with more realistic AI-generated video capabilities

Google has released updates to its AI video generation model, Veo 3.1, adding native vertical video output and expanding its integration with YouTube.

The update introduces 9:16 vertical video generation, allowing users to create full-screen portrait videos without cropping landscape footage. The feature is aimed at short-form and mobile viewing formats, including YouTube Shorts, where vertical video is standard.

The new version also improves visual consistency across generated clips. Characters can retain the same appearance across scenes, even when settings change. Objects, textures and environments can be reused, supporting longer and more continuous narratives. Motion and transitions have also been refined, including in videos generated from brief prompts.

Veo 3.1 expands output quality options, allowing videos to be generated in 1080p or 4K resolution. The 1080p is intended for social media and post-production use, while 4K is designed for large-screen and professional applications. These features are available through Flow, the Gemini API and Vertex AI.

The updated Ingredients to Video feature is now available to consumers through the Gemini app. YouTube Shorts and the YouTube Create app are gaining access for the first time. Professional users can use the same tools across Google’s enterprise and creative platforms.

All videos generated with its tools will include an embedded SynthID watermark. Users can also upload videos to the Gemini app to check whether content was created using Google’s AI, extending its existing verification tools from images to video.