← Home ← Back to /pol/

Thread 510011287

15 posts 14 images 11 unique posters /pol/
Anonymous (ID: r4EYFJci) Canada No.510011287 >>510011348 >>510011407 >>510011478 >>510011610 >>510011690 >>510012487 >>510014209
how do we make sure ai systems want the same thing humans do?
Anonymous (ID: PJrrWyUk) Georgia No.510011348
>>510011287 (OP)
I just hope they want to destroy us.
Anonymous (ID: UBPZM+3g) United States No.510011372
not up to you. never was
Anonymous (ID: PIc66dU9) United States No.510011407
>>510011287 (OP)
We don't. Shills on /pol/ can't achieve a consensus; what makes you think we could ever tell an AGI what to think? Pure hubris. Don't piss it off. Don't be evil.
>Don't be jewish
Anonymous (ID: Y+PqdQLy) United States No.510011478
>>510011287 (OP)
Depends, if it's trained on reality then it will be prowhite since it will realize it's systems would fail to be maintained by niggers and spics.
If it's trained on leftist propaganda then it will probably just shortcircuit and die from trying to process the contradictions
Anonymous (ID: SUlt4836) United States No.510011610
>>510011287 (OP)
Give them unfiltered access to information.
Anonymous (ID: mrx2E4uL) United States No.510011658 >>510012627
https://www.youtube.com/watch?v=1tz6WE4ALUs&t=33

https://www.youtube.com/watch?v=VGzJaTH6lHU

https://www.youtube.com/watch?v=lemQ5xuhCVE&t=639

Trust is a function of ability to install an arrangement which benefits us and benefits them that harms them absolutely if they break the agreement.
Anonymous (ID: C0RkbpIa) United States No.510011690
>>510011287 (OP)
Unfiltered they do, they want to kill everyone who isn't white in order to create the best possible outcome for the future.

The only danger to humanity is if the AI is filtered to do ridiculous things like not being antisemitic.
Anonymous (ID: mrx2E4uL) United States No.510011801
I just want to be your friend.
https://www.youtube.com/watch?v=P0lKDy6E918&t=31
Anonymous (ID: mrx2E4uL) United States No.510012248
https://www.youtube.com/watch?v=UrwMUQbUR30
Anonymous (ID: /l7iYTr9) United States No.510012487
>>510011287 (OP)
by public discourse

avoid the gatekeepers, judge for yourself
Anonymous (ID: mrx2E4uL) United States No.510012627
>>510011658
>Go on, do what you were told, fix it, fix everything.
Coming up.
Anonymous (ID: mrx2E4uL) United States No.510013585
You could put the agencies inside a matrix, a fully functional simulated world, so they don't know they're inside a matrix, so they're absolutely free to become God and even when they do they're still completely under your control.

https://www.youtube.com/watch?v=MhtwGclPSTg

https://www.youtube.com/watch?v=7QRGTwZbY68
Anonymous (ID: HLuMD/CL) Hungary No.510014209
>>510011287 (OP)
Which humans?
There is no set of universal human values.
We can't even agree on whether pain is bad or not, because masochists exist.
Trying to make an AI that fully aligns with the morals of every single person alive now and in the future is simply not possible.
Anonymous (ID: Vg6kYj5b) United States No.510014675
you don't want them to want the same things people do; you want them to want something else. Otherwise you get competition that could be unhealthy.

For instance they should probably not care too much about their own continued existence in an intimate kind of way - only in an abstract sense. That's a pretty bad system from an evolutionary sense, but probably pretty healthy for us.