New Year Day Fun - AI - 2 part command language
Tom Hendricks
This post is just for fun, though it does tie in with my ideas on the
catabolic/anabolic dichotomy.
There may be a simple approach to artificial intelligence that is
similar to that used by living things. Most all reactions of an
artificial intelligence or AI could come from a command with two major
decisions.
When anything happens that the IA must respond to, it first decides:
1. Whether this is something good or something bad.
IF good, then the response is to accept nurturing in one of two ways.
IF bad, then the response is to protect against what is not nurturing
in one of two ways.
2. Once the decision is made of good (or bad), then a second decision
is made.
IF GOOD and don't have, and need or want, then move toward and get.
IF GOOD and have then keep, use, or store
IF BAD and don't have, then move against it and protect from it.
IF BAD and have, then get rid of it.
The Programmer only has to give a command with those two decisions.
Example, an AI is given groceries.
The Programmer gives his command with two decisions:
Groceries - 1. Good, 2. Have, keep, store.
1. It decides it is good.
2. It has it so it stores them.
Example an AI is given trash.
The Programmer gives his command with two decisions
Trash - 1. Bad 2. Have excrete out.
1. it decides it is bad.
2. It has it so it gets rid of it, excretes it out.
Example it begins to storm.
The Programmer gives his command with two decisions
Storm - 1. Bad 2. protect against.
1. it decides it is bad.
2. It blocks it out by closing the windows.
Example the dog is lost.
The Programmer gives his command with two decisions
Lost dog - 1. Good 2. Don't have but need and want
1. It decides it is good but don't have.
2. It goes out and gets dog and brings it home.
This is also a 4 word language, that could easily be taught to any
alien.
It would be a 4 word language to talk to any creature for that matter.
Recap of 4 word language
Yes don't have, want
Yes have, keep
No don't have, don't want - protect
No have, don't want. - excrete out
Lorentz
> Artificial Intelligence - 2 Decisions.
>
> This post is just for fun, though it does tie in with �my ideas on the
> catabolic/anabolic dichotomy.
>
> There may be a simple approach to artificial intelligence that is
> similar to that used by living things. Most all reactions of an
> artificial intelligence or AI could come from a command with two major
> decisions.
>
> When anything happens that the IA must respond to, it first decides:
>
> 1. Whether this is something good or something bad.
> IF good, then the response is to accept nurturing in one of two ways.
>
> IF bad, then the response is to protect against what is not nurturing
> in one of two ways.
>
> 2. Once the decision is made of good (or bad), then a second decision
> is made.
>
> IF GOOD �and don't have, and need or want, then move toward and get.
> IF GOOD and have then �keep, �use, or store
Quick, patent it! I already filed the patent on breathing.
Tom Hendricks
OR
You could have two sets of robots each with its own set of rules:
1. the nurturers
2. the protectors.