By using this site, you agree to our Privacy Policy and our Terms of Use. Close

Forums - General - Genie 3 and the impact of AI on video games

A lot of you have probably already seen this but just in case here's a thread on it. Genie 3 is a new very impressive looking AI model from Google DeepMind that can create interactive 3D worlds. It's not available to the public yet but the applications for this sort of tech are huge. It's not ready for VR one reason being it's only 24fps but when it is ready you'll be able to walk around countless generated environments and use speech to prompt it to make basically whatever you want happen on the fly which sounds so cool that I need it to happen within a few years. Thankfully with the rate of improvement going on that happening that soon wouldn't surprise me at all. Just compare the video below to Genie 2 in this tweet from only eight months ago and it seems clear that things are gonna get truly crazy by the 2030's.

Last edited by Norion - on 07 August 2025

Around the Network

So yeah, these world models are the next leap. Not just in games, but in general (lossy, but still semi-accurate) representations of the world which can be used with reinforcement learning or genetic algorithms to train (or "grow" rather) embodied intelligence and to essentially eliminate spatial reasoning hallucinations for LLMs and other general purposed neuro-symbolic models.  

I'm starting to think the technological singularity started in 2008 with GPGPU and big-data collection via Web 2.0. Everything else has just been accelerating from then. 

As for gaming, it will be interesting to see how this will be configured and combined with traditional game engines and how controllable constraints can be set. 



I need to see more before I'm sold on its capabilities (ie how interactive, consistent, and accurate it is) but if genuine then it's definitely one of the most impressive examples of where AI is headed. Like HoloDust said when I posted that vid in the PC thread, imagine prompting an AI to scan your favorite book/movie/show and then generate a world out of it that you can explore and interact with. 



Yeah, this is looking crazy cool. I can't wait to see how it will impact gaming.

The next decade could be very interesting.



TallSilhouette said:

I need to see more before I'm sold on its capabilities (ie how interactive, consistent, and accurate it is) but if genuine then it's definitely one of the most impressive examples of where AI is headed. Like HoloDust said when I posted that vid in the PC thread, imagine prompting an AI to scan your favorite book/movie/show and then generate a world out of it that you can explore and interact with. 

Describe it, and it shall be made. Kinda like Stanislaw Lem envisioned back in 60s in The Cyberiad.