E
ChatBot cancel
smart_toy

Hello, I'm VAI. Are you new to Vease?

Example Image ChatBot cancel
Hello, I'm VAI. Are you new to Vease?

Google DeepMind’s Genie: Turning Images into Games

Genie AI by Google DeepMind transforming simple images into playable 2D platformer games, showcasing AI's potential in game development.

A Game-Changing AI

Google DeepMind is pushing boundaries again with Genie, an AI that’s like a magic wand for video games. Picture this: you take a photo or doodle something, and Genie transforms it into a game you can actually play. We’re talking about a single step from image to interactive fun, thanks to an 11 billion-parameter model that’s been fed over 200,000 hours of 2D platformer game videos. The model is currently running at 1 FPS, so right now it’s far away from real-time playable.

How Does It Work?

Imagine taking a snapshot or sketching a rough scene. Genie takes this input and, like magic, turns it into a playable 2D platformer game. Right now, the games are pretty basic, mainly because Genie’s been learning from low-res videos. But think about the possibilities as it starts understanding high-res images and gets more computing power to play with.

The Future of Interactive Entertainment

We’re looking at a horizon where AI doesn’t just create characters or landscapes but whole immersive, interactive experiences. Genie could be the first step toward AI-generated 3D worlds, characters that adapt and grow, and games that write themselves around your actions and words.

What’s Next for Genie?

The tech’s in its early days, with the generated games more novelty than next-gen for now. But the potential is huge. As Genie learns from more and higher quality data, and as DeepMind pours more resources into it, we’ll see games that are richer, more complex, and more engaging. Genie is opening the door to a future where anyone can create games and interactive experiences, no coding required. The question isn’t if this will change the game but how soon.

Image credit: MJ