Google enhances AI video generation with Veo 3.1 and native vertical support

Continuous improvements in Google’s generated AI continue today with a major update to Veo 3.1, introducing a series of features designed to move AI-generated content beyond experimental clips and into the realm of professional-level, high-fidelity production. This update focuses on three key pillars: “footage-to-video” consistency, native vertical support for mobile-first content, and high-resolution upscaling.

Perhaps the most impressive technical advancement in this version is the improved consistency of characters and environments. One of the main hurdles in AI video has always been maintaining the “identity” of a character or setting across multiple scenes. Veo 3.1 allows users to provide “ingredient” images to anchor the AI, ensuring that a character’s appearance or specific background remains stable even as the camera moves or the story changes. This level of control is essential for storytellers who want to go beyond one-shot wonders and create coherent, multi-scene narratives.

Mobile-first creation and professional fidelity

Recognizing the advantages of short-form video, Google is introducing native vertical output (9:16 aspect ratio) to its “Ingredients to Video” workflow. Previously, creators had to crop horizontal AI videos, resulting in poor quality and unwieldy frames. This native support enables full-screen storytelling specifically tailored for platforms like YouTube Shorts and YouTube Create, underscoring Google’s intent to put pro-level tools directly into the hands of mobile creators.

For those working with larger screens, Veo 3.1 includes state-of-the-art upscaling to 1080p and 4K resolutions. These higher fidelity options are being deployed in professional and enterprise environments such as Gemini API, Vertex AI, and Google Vids. By delivering broadcast-ready clarity and rich textures, Google is positioning Veo as a viable tool for high-end production workflows, not just a social media novelty.

Featured videos

Integration across the Google ecosystem

Google is making these new features available across its product stack. Starting today, consumers can start experimenting with enhanced Veo 3.1 features within the Gemini app and YouTube. On the professional side, expanded API support and 4K options allow developers to integrate high-quality video generation directly into their own applications.

As with our image generation tools, Google continues to focus on transparency and security. All videos generated through these tools are embedded with SynthID, an imperceptible digital watermark. Additionally, the Gemini app has been updated to allow users to review video content. Just upload a clip and ask Gemini if ​​it was generated by Google’s AI. As these tools become capable of creating more realistic content, a commitment to a transparent ecosystem is a necessary foundation.