Google is testing new AI video creation features for Google Gemini, according to a new report. This means the Gemini app may soon be able to generate not only pictures but also realistic video clips based on simple text prompts.
As well-known tipster AssembleDebug discovered for Android Authority, the latest Google app version contains hidden mentions of new video generation features powered by “robin,” the company’s internal codename for Gemini-powered functions. The code also includes text snippets associated with the video generation process, including “Working ……” and “We will notify you when it’s ready.”
Google hasn’t yet made any AI video generation tools available to Gemini users. However, these new code fragments point to features Google is testing for possible inclusion in a future version of the app. We don’t yet know what these video generation tools might do, but there are a few possibilities based on Google’s current technologies.
Could Gemini Users Gain Access To Veo 2?
Perhaps the most exciting (or terrifying) possibility is that Google will give Gemini users access to Veo 2, Google DeepMind’s cutting-edge text-to-video and image-to-video AI model. Veo 2, currently available in early access form as VideoFX from Google Labs, can produce incredibly realistic video clips (like the example above) based on text descriptions or still images uploaded by the user.
Gemini can already generate high-quality still images from text prompts using Imagen 3, Google’s text-to-image model. However, extending Gemini’s capabilities to include Veo 2 would be a huge step forward. Unlike still images, high-quality AI videos currently require significant computing power to generate and are typically limited to paid-for services, although open-source alternatives are constantly improving.
Another possibility is that Google could deploy a significantly cut-down version of Veo 2 or perhaps an entirely different form of AI-powered video generation, such as the Gemini-powered video creation app Google Vids, currently available as part of Google’s business-focused Google Workspace subscriptions.
Google Vids helps users produce and collaborate on business-oriented videos, such as presentations and promotional content, using templates, stock media, AI-assisted script writing and voice synthesis. While some of these tools would be useful to general consumers, in its current form, Google Vids doesn’t seem like a good fit for the Gemini app.
How Much Will It Cost?
Google has not announced any AI video generation tools for the Gemini app. If they do appear, I’d expect to find Google’s most powerful AI models restricted to customers with paid subscriptions.
Follow @paul_monckton on Instagram.
https://www.forbes.com/sites/paulmonckton/2025/02/09/google-announces-much-needed-ai-protection-feature-for-google-photos/
https://www.forbes.com/sites/paulmonckton/2025/02/13/new-google-leak-reveals-immersive-google-photos-memories-upgrade/
