Google Adds Video Generation to Gemini

Google has introduced new video generation features using the “Veo 2” model in the Gemini app and the experimental Whisk platform. Starting April 15, Gemini Advanced subscribers can create 8-second video clips at 720p resolution from simple text prompts. These videos can be easily shared on social media directly from mobile devices.

Veo 2 produces realistic scenes with smooth character motion and a cinematic visual style. Users describe the scene in natural language, and the system generates a short film based on the given details, genre, and storyline.

Google has also integrated Veo 2 into the Whisk platform, which previously allowed image generation from text or reference pictures. Now, with the new Whisk Animate feature, Google One AI Premium subscribers can bring these images to life as short, animated videos.

All generated videos are automatically marked with a SynthID watermark embedded in every frame — clearly indicating they were AI-generated. According to Google, the model has undergone safety testing to prevent the generation of harmful content.

Examples include scenes like a candy figurine in an icy cave, a mouse with glasses under a glowing mushroom, or a pixelated ice cream melting against a blue sky. Early feedback has been positive, with users praising the visual quality and ease of use.