Doesn’t this imply that a transformer or NN could fill in details more efficiently than traditional techniques?

I’m really curious why this would be preferable for a AAA studio game outside of potential cost savings. Also imagine it’d come at the cost of deterministic output / consistency in visuals.

  I could absolutely see something like rendering a simplified and stylised version and getting Transformers to fill in details. That's kind of a direct evolution from the upscaling approach described here, but end to end rendering from game state is far less obvious.

Sure. This could be a variation. You do a quick render that any GPU from 2025 can do and then make the frame hyper realistic through a transformer model. It's basically saying the same thing.

The main rendering would be done by the transformer.

Already in 2025, Google Veo 3 is generating pixels far more realistic than AAA games. I don't see why this wouldn't be the default rendering mode for AAA games in 2035. It's insanity to think it won't be.

Veo3: https://aistudio.google.com/models/veo-3

> Already in 2025, Google Veo 3 is generating pixels far more realistic than AAA games.

Traditional rendering techniques can also easily exceed the quality of AAA games if you don't impose strict time or latency constraints on them. Wake me up when a version of Veo is generating HD frames in less than 16 milliseconds, on consumer hardware, without batching, and then we can talk about whether that inevitably much smaller model is good enough to be a competitive game renderer.

Well you missed the point. You could call it prompt adherence. I need veo to generate the next frame in a few milliseconds, and correctly represent the position of all the cars in the scene (reacting to player input) reliably to very high accuracy.

You conflate the challenge of generating realistic pixels with the challenge of generating realistic pixels that represent a highly detailed world state.

So I don't think your argument is convincing or complete.

> Google Veo 3 is generating pixels far more realistic than AAA games

That’s because games are "realtime", meaning with a tight frame-time budget. AI models are not (and are even running on multiple cards each costing 6 figures).

I mistaken veo3 for Genie model. Genie is the Google model I should have referenced. It is real time.

I look forward to playing games where the map and scenery geometry just mutates as I move around, and can’t complete levels or objectives because the model forgot about their existence while it was adding a door midair at 0.25 fps.

I’m so excited to be charged AAA prices for said wonderful experience.

That already exists for Minecraft

Genie 3 is still currently low quality, low resolution and not anywhere near current AAA graphics, while requiring hardware that exceed current AAA Graphics requirements.

For Genie to exceed AAA Graphics in 2035 at 60 to 120fps per second would require a breakthrough of efficiency that is at least an order of magnitude, and much higher for it to be cost effective.

The gaming industry for AAA titles requires at least 3-4 years in making. Which means AAA titles studios would need to start working on it in 2031. Possibility of All AAA games in 2031 are made with LLM model is practically zero.

The point is that we can generate ultra realistic graphics.

We are talking 10 years from now.

We can already do that today given sufficient computing resources with current AI, doing it real time is different. And I thought you mentioned earlier that all AAA games will be using it by 2035?

And unfortunately 10 years isn't that long time in many industries. We are barely talking about 3 cycles.

Genie is real time.

Genie 3 is already a frontier approach to interactive generative world views no?

It will be AI all the way down soon. The models internal world view could be multiple passes and multi layer with different strategies... In any case; safe to say more AI will be involved in more places ;)

I am super intrigued by such world models. But at the same time it's important to understand where they are at. They are celebrating the achievement of keeping the world mostly consistent for 60 seconds, and this is 720p at 24fps.

I think it's reasonable to assume we won't see this tech replace game engines without significant further breakthroughs...

For LLMs agentic workflows ended up being a big breakthrough to make them usable. Maybe these World Models will interact with a sort of game engine directly somehow to get the required consistency. But it's not evident that you can just scale your way from "visual memory extending up to one minute ago" to 70+ hour game experiences.

For future readers, yes. It’s Google Genie.