Reinforcement learning not only works with humans, but apparently with artificial intelligence as well to get the results you're looking for.

OpenAI, founded by (among others) SpaceX and Tesla founder and CEO, Elon Musk; created a program that simulated 481 million games of hide-and-seek.  The AI agents consisted of "seekers" and "hiders."  Both of which developed many strategies, including breaking the game's physics, to win the game and earn the points that both are working to claim hence, reinforcement learning.

https://www.vox.com/future-perfect/2019/9/20/20872672/ai-learn-play-hide-and-seek

What are your thoughts on machine learning and the inevitable future of intelligent machines becoming apart of even the most basic aspects of your daily life? 

I've seen Battlestar Galactica, so I'm totally against the creation of the Cylons...

I've seen Battlestar Galactica, so I'm totally against the creation of the Cylons...

LOL Yeah I guess that's fair.  There will have to be extremely heavy regulations on AI to prevent them for executing commands they were not meant to do.  As this simulation showed, many were in place but the AI still found a weakness.  I suppose we could make another AI to find these and fix them.  And yet another AI to fix that one