Google has added a new feature to its Veo 3 video generation tool that lets users create videos using just an image and a text prompt. This update is available through the Gemini website and is currently being rolled out in select countries. However, it is not yet available in India.
With this new feature, users can upload a reference image and write a description of what they want to see in the video. The Veo 3 model, powered by artificial intelligence (AI), will then turn that image into a video based on the prompt.
To try it out, users need to visit the Gemini web platform, click on the “Video” button below the text box, and then upload a photo. After adding a written prompt—like what actions or scenes they want to create—Veo 3 will generate a short video.
Google shared a demo showing some fun and creative results. For example, it used a simple image of a cardboard box and turned it into different video scenes. These included a hamster cooking a meal inside the box, a man jumping into the box, and even an elevator coming out of it. This shows how creative and flexible the new feature can be.
However, the image-to-video feature is only available to those with a Google AI Pro or Ultra subscription. Also, this feature is not yet supported on Gemini mobile apps—you must use the desktop website.
To ensure transparency and safety, every video made using Veo 3 will include two types of watermarks:
- A visible watermark to clearly show that the video was created using AI.
- An invisible watermark using Google’s SynthID technology, which cannot be removed or edited. This invisible mark can only be seen with special software and helps identify AI-generated content.
In summary, Veo 3’s latest update makes video creation more interactive and fun, offering powerful tools to bring images to life through AI. Once fully rolled out, it could open exciting creative possibilities for users worldwide.
Also Read: Elon Musk launches Grok 4 AI chatbot after controversy
