Frame from Playing a Neural Network's version of GTA V: GAN Theft Auto Plays with YouTube's privacy-enhanced embed

Playing a Neural Network's version of GTA V: GAN Theft Auto

Open on YouTube First HN thread

3 HN users recommended this demo in 4 different months between 2021 and 2025.

2007–2026
4 recommendations 8 citations HackerTube score 4

Cited as an early, striking demo of neural networks acting as game engines and world models.

Sentdex demonstrates GAN Theft Auto, a GameGAN-based neural network that recreates GTA V so the network itself is the playable environment. He shows the results and releases the code.

Pull quote: “This YouTube video, to me, shows the promise of things to come.” (HN, 2022)

Also tagged: neural network game simulation, gan, gamegan, gta v, world models, generative ai

What Hacker News said

12 labeled comments from the HN archive; up to 20 are kept per video, favoring different commenters. Each one links back to the original thread.

Recommends ALittleLight August 2021 under Where are the robotic bricklayers?
Shows promise of automated game generation.
This video really amazed me in terms of automated game development. My fantasies are something like this and GPT-3 plus a couple decades of progress. I think we'll get to genuinely open ended sand box games. [this video]
Recommends ALittleLight April 2022 under DALL-E, the Metaverse, and Zero Marginal Content
Glimpse of AI-generated game worlds.
This YouTube video, to me, shows the promise of things to come. AI generated game worlds. Language models to generate plots and dialog, transformers and GANs to create illustrations. Imagine a game, a truly open world sand box, Grand Theft Auto meets AI Dungeon - every NPC is a "real" person with unlimited dialog options, the buildings you drive by you could easily walk in and investigate, unlimited play space - you could type in more general instructions and ideas to the plot generator ("add in a vampire romance and murder mystery angle") on the fly. [this video]
Recommends eru December 2022 under Netflix: 30% bitrate saving using Film Grain Synthesis
Neural net encodes game world rather than bitmaps.
> You could directly encode the elements as understanded by the process used to generate them, rather than attempting to reconstitute things from a stream of bitmaps. You might like Gan Theft Auto: [this video]

Citations, criticism and jokes

Cites it magic_quotes June 2021 under GAN Theft Auto [video]
Neural network generates a playable GTA simulation.
Have you seen the video? The author even goes as far as suggesting the technique might useful for (generating?) entire operating systems at [this video] . That's just wild.
Cites it bentcorner July 2021 under The art of asking nicely
Neural-net generated GTA has dream-like quality.
GAN Theft Auto also had a similar dream-like quality, especially with anything involving collisions (cars and how "big" the road was). [this video]
Neural net simulating GTA from controller input.
Read the full comment
> I'm not going to be so stupid as to say 8 cores should be enough for anyone (while attached to a machine with 128) but you have to wonder if the stable diffusion style apps running on your desktop are going to be mainstream, or isolated to the few who choose to _need_ them as a hobby or a smaller part of the public that uses them for commercial success. AKA, I can utilize just about every core i'm given with parallel compiles, or rendering a 4K video, but I'm pretty sure i'm the only one in my immediate family that needs that. My wife in the past might have done some simulation work, but these days the heaviest thing she runs on her PC is office products. See GAN Theft Auto at [this video] Someone trained a neural network to convert controller input into video output that simulates the game Grand Theft Auto. If technology keeps improving, I expect many future games will be such 'dreams' of neural networks. You are right that running Word or Excel won't really benefit from more cores.
Cites it SXX November 2023 under Stable Video Diffusion
Earlier GAN-based generated game video.
Actually it was already done by sentdex with GAN Theft Auto: [this video] To an extent... PS: Video is 2 years old, but still really impressive.
Cites it KhoomeiK August 2024 under Diffusion models are real-time game engines
GAN-based GTA demo preceding diffusion game engines.
NVIDIA did something similar with GANs in 2020 [1], except users could actually play those games (unlike in this diffusion work which just plays back simulated video). Sentdex later adapted this to play GTA with a really cool demo [2]. [1] [link] [2] [this video]
Cites it spyder September 2024 under Leak claims RTX 5090 has 600W TGP, RTX 5080 hits 400W
Example of running a whole game inside a neural network.
True, but also AI could be the reason why games would need more powerful GPUs in the future (but not yet). For example LLMs for NPC dialogs, or the wilder things like running the whole game inside a neural network, or running an AI style filter over the output of the regular game engines. [link] [this video] [link]
Cites it imtringued November 2025 under Spatial intelligence is AI’s next frontier
Learned world model controlled by a human.
>There's some progress in video generation which takes a short clip and extrapolates what happens next. That's a promising line of development. The key to "common sense" is being able to predict what happens next well enough to avoid big mistakes in the short term, a few seconds. How's that coming along? And what's the internal world model, assuming we even know? [this video] This has been a thing for a while. It's actually a funny way to demonstrate model based control by replacing the controller with a human.

Keyboard shortcuts

/
Search
j k
Next / previous video in a list
Enter
Open the focused video
l
Like (on a video page)
w
Watch later (on a video page)
?
This help
Esc
Close