Google is now making its its world generating model available to the general public under the name Project Genie.
According to 9to5Google, the model either uses a prebuilt world or lets you describe the environment you want to create, how you want to explore it (flying, driving, walking, etc.), and the point of view.
From there, you will specify your character (person, animal, project, etc.) Project Genie will then use the information provided to create a preview image (also called World Sketching) with Nano Banana Pro.
These AI-generated interactive worlds are available for 60 seconds and are set at 720p resolution at 20-24 frames per second.
From what I’ve read, it looks to be quite versatile. 9to5Google shared two videos that showed off what Project Genie could do: a goldfish swimming around in the ocean, and the POV of a surfboard riding a wave.
Going beyond realistic water simulations, The Verge’s Jay Peters recreated Nintendo titles like Super Mario 64, Metroid Prime 4: Beyond, and even The Legend of Zelda: Breath of the Wild, which, to me, is impressive.
As expected, Project Genie is currently limited to U.S. users with Google’s AI Ultra plan (priced at $340 in Canada). I wouldn’t get too upset, though, as this isn’t the first time Google has limited AI features to the U.S. before they debut north of the border.
Source: Engadget, 9to5Google
Image Credit: Google DeepMind
