What’s worse than rogue artificial intelligence? How about a rogue artificial intelligence that could win Ninja Warrior? You know this and I know this, but perhaps someone needs to tell the folks over at Google’s Deepmind project.
Our benevolent overlords just released a video of their DeepMind artificial intelligence system learning how to navigate a virtual parkour course through a process of reinforcement learning.
At its most basic level, the system was as follows: the faster the AI moved across the terrain, the greater the rewards. Additional incentives and penalties were added for more complex programs.
The AI used a trial and error system to figure out how to move forward as fast as possible without “terminating.”