Hồng Hoa
@honghoa98
2500 Following
718 Followers
1 reply
0 recast
0 reaction
13 replies
21 recasts
52 reactions
9 replies
10 recasts
22 reactions
2 replies
1 recast
5 reactions
4 replies
3 recasts
4 reactions
4 replies
3 recasts
5 reactions
4 replies
9 recasts
24 reactions
3 replies
4 recasts
6 reactions
3 replies
9 recasts
16 reactions
4 replies
16 recasts
30 reactions
0 reply
6 recasts
7 reactions
7 replies
8 recasts
17 reactions
2 replies
13 recasts
50 reactions
0 reply
0 recast
0 reaction
🧵09/34
Self Preservation
---
The narrow AI, you can unplug it whenever you want, it does not even understand the idea that it can be on or off.
With AGI it’s not that simple.
When the AGI plays chess, unlike the simple narrow chess program, its calculations include things in the real world, like the fact it might get unplugged in the middle of the game.
The main goal of the AGI is achieving checkmate at chess, but if it’s turned off it can not achieve that.
Basically, you can not win at chess if you are dead.
So staying up-and-running naturally becomes another instrumental goal for winning in the game.
And the same applies with any problem it works on.
Its nature, by being General, automatically gives it something like an indirect survival instinct out of the box.
It’s not a human or an animal, but it’s also quite different from the other tools we use.
It starts to feel a bit like …. a weird new life form. 0 reply
16 recasts
4 reactions
0 reply
6 recasts
0 reaction
0 reply
15 recasts
24 reactions
0 reply
0 recast
0 reaction
0 reply
0 recast
0 reaction
1 reply
7 recasts
0 reaction