No Widgets found in the Sidebar

Google DeepMind’s Generative AI team has introduced “Genie,” an AI system that can generate playable video game worlds from images and text prompts.

How Genie Works

Genie is trained on over 200,000 hours of publicly available 2D platformers. This training allows Genie to understand the physics and mechanics of video games, such as player control, actions, and movement.

To create a game world, users simply provide Genie with a prompt, sketch, or image. Genie then uses this information to generate a virtual world that users can play around in. The AI can create assets from scratch and generate pixels based on player actions.

Genie’s Potential Applications

Genie has the potential to revolutionize the way video games are made. It could allow developers to create new games more quickly and easily, and it could also open up the possibility of creating games that are more personalized and interactive.

Genie could also be used in the field of robotics to help train robots on how to navigate environments.

Concerns about Genie

While Genie is an impressive generative AI, it also raises concerns about the future job security of video game developers and the potential for spam games and asset flips on marketplaces like Steam.

Overall, Genie is a promising new AI technology with the potential to change the way video games are made and played.