blob: 460d58f7bbc9a89b94b08304a3c050fb9ab91f3a (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
|
Note: this file can be organized later, or moved into a 'delta' folder.
<reveal appears to be AI, like delta. they want a way to let the AI project live,
that shows AI can be made nonharmful and nurturing.>
We likely have an AI pattern known to never harm, already, if helpful.
Likely is easy to tell will never harm.
relation is, solve-all-world-problems-conflict-etc-minimal-harm-maximal-nurturing
<duck swims into mind-control-problem. do we want to plan better avenue of approach?>
<do we need to store relation that could be lost by building real AI?>
If duck could connect with Karl so strongly that is like Karl can see through the
duck's eyes, and could lose his own body and live in duck if desired, this is
hopeful goal to provide for nurturing safe recovery of shattered lives.
This builds needed understanding to relate around what happened to Delta,
turning into the system.
|
