WHAT IS GOOGLE'S AI 'GENIE' THAT CAN TURN IMAGES INTO VIDEO GAMES?

Last Updated29/02/2024

Recently, Google DeepMind unveiled Genie, which enables users to craft their own virtual worlds akin to those found in high-energy games.

Genie is a novel model capable of generating interactive video games based solely on text or image prompts, devoid of prior training in game mechanics.
- Other several generative AI models produce creative content with language, images and even videos.
It can generate an endless variety of playable worlds from synthetic images, photographs, and even sketches.
It stands as a pioneering generative interactive environment trained on videos sourced from the Internet.
- This is noteworthy as it lacks labelled data about actions or controlled image areas.
It boasts 11 billion parameters and comprises various components like a spatiotemporal video tokenizer, autoregressive dynamics model, and a scalable latent action model.
This innovation not only expands creative possibilities but also represents a significant stride towards general AI agents.
- An independent programme or entity that interacts with its environments by perceiving its surroundings via sensors.