[Ivan Vodišek]

Hi all.

An idea came through my brain as I wanted to make plans for autonomous AI. The main problem was to recognize living beings so AI could help them in their striving. After a while I realized that I don't need to recognize if a unit is alive or not (which was a big problem for me). It is sufficient just to recognize that unit exists and to recognize its striving/purpose. A purpose or striving (it should be the same) is an attempt to shape up the Universe in a some way.

1. If a unit is alive, no problems, recognize striving and try to help.

2. if a unit is not alive the same case applies: recognize its striving and try to help because either:

    a) someone made the unit on purpose,

    b) the unit has purpose incidentally, or

    c) purpose is unknown / it has no purpose - in this case nothing would be done.

By this approach we reduced the problem to recognizing different units in the Universe and finding its striving to make an influence.

Conflicts should be managed in a classic way: in a case of a conflict neither side should be honored. Instead, both sides should find alternative solution that does not mess up with anyone else's striving. In my eyes, this conflict handling is perfectly safe for living beings because it does nothing that hurts anyone's opinion. Each striving is a result of it's parent striving, so alternative solution should be always able to be found.

Additionally, as the Universe has many, many units, some weighting function should be applied when choosing which unit to work out. This weighting function should embrace:

1. opportunity: possibility of finding solution fast enough

2. influenced unit count: more units -> the higher index

3. problem size: if many units spare more time on solving the same problem -> the higher index.

For the end I bring a part of autonomous AI algorithm (no worries, no copyright or restrictions):

​

Ok guys, this is all I've got on this subject for now and I wanted to share this with you.

Let me know how you like it :)

Ivan