Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Sergarr]

Well you could say the same about humans, really. The difference here is that AI's behavior would be much easier to test, and that successful behaviors would be much easier to propagate to other AI, thus making AIs theoretically much better at the whole "not making obvious errors" thing.

[Graknorke]

Not really. Humans already have all of these biases and preconceptions that we don't even really think about normally. With an AI you have to consider those and create an objective measure of them from the ground up.

[wierd]

You dont need to model the humans, if you directly interface with the humans, and gather live input.  An AI can gather and process data faster than human government officials can.

By directly, I DO NOT mean with brain implants mind.

I mean, more or less, consider the already extant mass surveillance conducted by global intelligence agencies. Now, imagine the AI reading everyone's candid thoughts, including this post I am making right now, and performing weighted analyses on that data, in real time.

[Bohandas]

AI implies the ability to learn; that seems a bit overengineered for a lot of decisions, which could probably be made by plugging the right statistics into a fixed equation

EDIT:
Some of he less important decisions could even be given to a "Magic 8-Ball"

[Spehss _]

And that's a standard sci-fi trope to justify the robot revolution against humans. "Humans are flawed, beep boop, better to remove humans than serve flawed meatbags, beep boop. Kill kill kill beeeeep."

That... isn't what I meant. It's not so much that they'd want to get rid of humans as they wouldn't consider them the same way other humans do.
See: Stamp collecting AI thought experiment.

I was responding to MonkeyHead's comment on AI identifying humans as the weak link and stuff, not what you said or meant. :v

Good video. From the example in the video, I'd think the reason the stamp AI does that is because it was made with no limiters. Its goal was to get as many stamps as possible as fast as possible in whatever way it could think of. No limits on how to get stamps, or if it could make stamps, or if it could make stamps out of people. That's pretty much why MonkeyHead's AI would determine to wipe out humans. It was given no limiters and determines that the best way to reach its goal is to wipe out humans.

You could get the same kind of thinking from a human, theoretically. Make them free to do whatever, with the power to do anything they can think of, and remove all their ethics and morals and any mental hindrances, and give them a single broad goal for them to fulfill singlemindedly.

[Sergarr]

Not really. Humans already have all of these biases and preconceptions that we don't even really think about normally. With an AI you have to consider those and create an objective measure of them from the ground up.

>implying AIs won't have biases and preconceptions

[wierd]

Human satisfaction is not an easily quantized function.

It is both the goal, and the limiter.

(It would be the bane of capitalism though. Capitalism sequesters and concentrates resources and capital into a small group of hands-- The AI would assume control over ALL capital and ALL resources, to assure maximum human satisfaction can be attained. That means wealthy business people would be unable to exist; The AI would not permit them to, as their existence implies a reduction of human satisfaction elsewhere that does not balance out; Millions suffering in 3rd world conditions, so that a few thousand can live it up-- The AI would decide against this.)

[Bohandas]

And that's a standard sci-fi trope to justify the robot revolution against humans. "Humans are flawed, beep boop, better to remove humans than serve flawed meatbags, beep boop. Kill kill kill beeeeep."

That... isn't what I meant. It's not so much that they'd want to get rid of humans as they wouldn't consider them the same way other humans do.
See: Stamp collecting AI thought experiment.

I was responding to MonkeyHead's comment on AI identifying humans as the weak link and stuff, not what you said or meant. :v

Good video. From the example in the video, I'd think the reason the stamp AI does that is because it was made with no limiters. Its goal was to get as many stamps as possible as fast as possible in whatever way it could think of. No limits on how to get stamps, or if it could make stamps, or if it could make stamps out of people. That's pretty much why MonkeyHead's AI would determine to wipe out humans. It was given no limiters and determines that the best way to reach its goal is to wipe out humans.

You could get the same kind of thinking from a human, theoretically. Make them free to do whatever, with the power to do anything they can think of, and remove all their ethics and morals and any mental hindrances, and give them a single broad goal for them to fulfill singlemindedly.

Indeed. A lot of the people in power are already sociopaths anyway.

[Loud Whispers]

Science has gone too far
Fuck you Zhakarov, load the planet busters, space war now

Nyet, nothing can stahp the progress!  Nothing- SWEET UNMERCIFUL SCIENCE, THE MINDWORMS

I TRIED TO WARN YOU ABOUT THE DRUMBEAT OF TECHNOLOGICAL PROCESS BRO
IT KEEPS HAPPENING

[Spehss _]

Human satisfaction is not an easily quantized function.

It is both the goal, and the limiter.

I feel like giving an AI a goal of "make humans happy" or the like would result in the ai wiping out humans so no one can ever be not happy. Or make people constantly happy and remove any ability for humans to feel anything less than the ai's definition of "happy".

Indeed. A lot of the people in power are already sociopaths anyway.

I was thinking of adding the whole "sociopath in power" to the end of my post. Except even existing sociopaths have ideas of social norms and values, if they were raised among other humans. The norms and values would effect the way the sociopath would think or respond to outside data they receive as inputs. The hypothetical stamp collecting ai doesn't have that, it goes purely by the data it receives and reacts to it based on the goal of getting stamps.

[Sergarr]

Human satisfaction is not an easily quantized function.

It is both the goal, and the limiter.

(It would be the bane of capitalism though. Capitalism sequesters and concentrates resources and capital into a small group of hands-- The AI would assume control over ALL capital and ALL resources, to assure maximum human satisfaction can be attained. That means wealthy business people would be unable to exist; The AI would not permit them to, as their existence implies a reduction of human satisfaction elsewhere that does not balance out; Millions suffering in 3rd world conditions, so that a few thousand can live it up-- The AI would decide against this.)

That implies that the AI would want to maximize human satisfaction. Why would it want to do that? Why would humans want to maximize human satisfaction?

[Tawa]

I feel like giving an AI a goal of "make humans happy" or the like would result in the ai wiping out humans so no one can ever be not happy.

That would be a roundabout and illogical way to achieve the goal. It would be like building the greatest house in the world by knocking down all the other ones and building a shack--it ultimately requires much more effort than building a better house.

[wierd]

Because without that goal, the AI will be very deleterious to humans.

Note SATISFACTION, NOT HAPPINESS.  Maximizing human happiness is easy-- strap them in a chair and pump them full of euphorics. (Or give them all brain implants and stimulate away. Even cheaper.)

Satisfaction requires that humans be able to respond to questions, and be active in their daily lives. A dead human is neither satisfied, nor dissatisfied-- it is no longer a human, it is a corpse.

Telling the AI it must respect human rights (including right to life) means the AI cannot cheat and murder all other humans except for a sociopath, which it then keeps alive forever in a pampered palace planet.

This forces the AI to make decisions that result in optimized local maxima against all of its directives.

[Sergarr]

Because without that goal, the AI will be very deleterious to humans.

Why? Why can't AIs be simply like normal humans, with their own little wants, wishes and dreams, like all of us?

[wierd]

Because that is not how an artificial mind works.