The AI field is evolving over time, as companies develop such software to enhance customer experience. Recently, Google’s AI research organization, DeepMind, created a technology to generate AI video games.
DeepMind latest AI model Genie 2 is capable of generating an “endless variety of action-controllable, playable 3D environments for training and evaluating embodied agents.” It can create videos based on simple prompt text. Humans and AI agents itself can play games using keyboard keys.
According to a Google DeepMind post, “Genie 2 is a world model, meaning it can simulate virtual worlds, including the consequences of taking any action (e.g. jump, swim, etc.). It was trained on a large-scale video dataset and, like other generative models, demonstrates various emergent capabilities at scale, such as object interactions, complex character animation, physics, and the ability to model and thus predict the behavior of other agents.”
However, the company claimed that Genie 2 can generate consistent worlds for up to one minute, while the video examples only last for 10–20 seconds. It will provide different perspectives like first-person, isometric views, or third-person driving videos.
With cutting-edge features, Genie 2 can remember parts of simulated scenes or pictures that aren’t in view. Then, generate the objects accurately when they appear in the video game. Even though, it wouldn’t give fun to players. Genie 2 could erase your progress every minute or second, which may frustrate users. As a result, DeepMind evaluated its AI model for unlimited workflows and “prototyping interactive experiences.”
Along with Genie 2, World Labs’ computer scientist Fei-Fei Li introduced the AI image generator that transforms classical paintings and still images into 3D images.
Thanks for choosing to leave a comment. Please keep in mind that all comments are moderated according to our comment Policy.
