Wasting energy of your opponents - Defeating AI series

 Losing count now, nobody cares about which number it is by now, as long as they can use it in their advantage. So let me help you.

The principle contained in 1999 AI recovery, which is a key in defeating AI as JAIL is "One Action Only". Combined with 5 Steps, which has a separate name for another time, it is the top 2 on the list.

The one action only implies you act habitually, but DO NOT GO INTO A CHAIN. As a chain of action CAN BE PREDICTED. Suppose you are discussing a prisoner on a run. All roads will be covered, and pretty much any unexpected action requires a complete fool of an officer (look it up "runner escaped jail" on Youtube) to let the suspect off the hook.

So, if you are in this game, and, let's say try to escape, you are only in luck if your enemy is weak. LIFE IS NOT LIKE THAT. Try getting away with a crime, you're in for a misery. NO OTHER WAY. Escape or not, wherever you go, whatever you do has been covered for years. Best case, life on a yacht, but you worry about running out of toilet paper with no supermarket nearby. Then maybe have another yacht to restock you. You are, by now in a chain of action. Where your exits WILL BE COVERED by your adversaries. Every criminal like that dies in misery, especially the ones that are publicised in press.

The solution: One Action Only. You perform an action, and you either complete it, and go back to your base point (i.e. root), or stop JUST SHORT of completion. In this way, given that the action is legal, you're in no trouble, and your opponent AI has just wasted energy in CALCULATING AND COVERING exits for you.

See, I told you it gets more interesting. Talk soon again. This is my one action for this time.

Comments

Popular posts from this blog

Development cycle vs publishing your code on Github

Was I wrong?

Alcohol and why it is in an AI blog