AI and machine learning have been in development for quite some time, but MarI/O aims to bring them to video games. YouTube user "Seth Bling" first tried it on Super Mario World, and now, it's been reprogrammed from Super Mario Kart. Like the previous experiment, MarI/O wasn't taught anything beforehand, and only a few simple parameters were set. The AI uses a "fitness level" to determine what buttons to press and so forth, literally being incentivized to continue doing so. After many attempts, it finally managed to beat the computer, using several dirt paths to increase efficiency.

We have seen the future, and artificial intelligence takes over. MarI/O, created by gamer Seth Bling, is just the tip of the iceberg. This AI had no idea how to play the game at first, but after some simple parameters were set in the form of "fitness" levels, it was incentivized to continue trying new ideas. You read that right, each level used new ideas, and it remembers what worked, while discarding its mistakes. After 34 steps, using NeuroEvolution of Augmenting Topologies, MarI/O figured out that jumping was the key to complete the stage.