Google unveils Imagen 3 and Veo AI models for image and video generation

Google has introduced two new AI models, Imagen 3 and Veo, on its Vertex AI platform. Imagen 3 generates images from text prompts, while Veo animates images based on text and existing visuals. Imagen 3 offers improved detail and lighting. Starting next week, Imagen 3 will be available to all Vertex AI users. It includes features like inpainting and outpainting for enhanced image customization. Veo, currently in private preview, can animate static images and create videos from text prompts. Both models include invisible watermarks to identify AI-generated content and have safety filters to prevent harmful outputs. Google states it has not used customer data for training and provides copyright protection.