Google is increasingly looking into the world of AI image and video generation. The company announced new versions of Veo and Imagen, its AI-powered image and video generators.
Veo 2 is a new video generator that Google says “better understands real-world physics and the nuances of human movement and expression” The AI video generator can create videos (like similar tools) from a prompt. Interestingly, Veo 2 can also receive prompts for specific goals and can simulate their use.
Videos can be created at resolutions up to 4K and durations “up to a few minutes,” according to Google. Google also claims that Veo 2 inserts quirks (like extra fingers or weird objects) less frequently than other models. Veo 2 will also be rolled out to YouTube Shorts sometime in 2025.
Along with Veo 2, Google DeepMind’s AI imaging model has been updated. Imagen 3 is said to “render more varied art styles with greater accuracy” The generator can receive prompts that include art styles, like anime or photorealism. Google says images match the prompt better and are more detailed. Like Veo 2, Imagen achieved “state-of-the-art results” when compared to human-created images, according to Google.
Finally, a new experience from Google Labs allows users to edit images, both input and output, using AI. The tool allows users to mix and match topics, scenes, and styles in a drag-and-drop interface to trick the AI into creating something new. Users can also give text instructions to refine the result.
These templates and tools are available today, but you may need to join a waiting list before you can use them.
Buy a Lenovo IdeaPad 3i Chromebook on Amazon.
