Google DeepMind’s Genie 3: What’s New in Real-Time 3D AI Creation
Humpy Adepu
Real-time 3D environments: Genie 3 is a world model that creates and renders interactive 3D worlds in real time from a single text prompt.
Higher fidelity and frame rate: The new model generates worlds at a much smoother 720p resolution and 24 frames per second, a significant leap from previous versions.
Improved visual memory: A key new feature is the AI's ability to remember and maintain the state of objects and details as users move around the environment.
Promptable world events: Users can instantly change the environment with simple text commands like "make it rain" or "add a character."
Longer interaction times: Genie 3 allows for several minutes of continuous interaction, whereas earlier models were limited to only a few seconds of engagement.
Versatility in creation: It can generate various environments, from realistic historical simulations to fantastical, imaginative worlds with dynamic details and characters.
Foundation for AGI: Google DeepMind positions Genie 3 as a crucial step towards Artificial General Intelligence (AGI), enabling AI agents to learn in unlimited simulated environments.
Empowering creators: Genie 3 could revolutionize game development, education, and entertainment by allowing developers to prototype and test new ideas rapidly.
Limited public access: The model is currently a research preview and not publicly available, with access limited to a small group of academics and creators for safety.