That’s behind Gan Theft Auto
The Gan Theft Auto project shows that an AI is quite capable of building a game environment that is actually playable, if not pretty.
Gan Theft Auto is a project using Nvidia’s technology Game Gan (Generative Adversarial Networks) used to train an AI who then tries to build their own version of Grand Theft Auto V. The result does not seem to knock the inclined observer off the chair, but is in any case very clearly recognizable as a GTA.
Contents
The AI can already build simple games like Pacman
It is also playable. The Game-Gan-Fork shows how capable the technology is in handling games that are more complex than Pacman. Nvidia had the cult game from the eighties recreated by an AI on the occasion of its fortieth birthday.
The Nvidia AI had proceeded as a human developer would have done. She tried to understand the rules of the game by watching and then to imitate it programmatically. Then she trained with 50,000 Pacman runs. In the end there was a fully functional version of the game that does not need any underlying game engine.
So the KI recreated GTA5
GitHub users felt the same way Sentdex in front. In the following video he shows that the AI was trained on various game scenes by showing it examples of cars driving in GTA5. The main aim, but not just, was to differentiate the surroundings from the vehicle.
In the end, Sentdex succeeded in creating a playable version of an – admittedly very simple – section of the game. Visually, the result is not necessarily a pleasure. The environment looks like you’re looking at it with eleven diopters, where you don’t need glasses.
Nevertheless, for the fact that the result was developed by an AI and is fully playable, we can be clearly impressed. Incidentally, a few months ago an Intel AI dealt with GTA5.
