AI will kill us all!

Author: Dr.Franklin

Posts

Total: 32
Dr.Franklin
Dr.Franklin's avatar
Debates: 32
Posts: 10,555
4
7
11
Dr.Franklin's avatar
Dr.Franklin
4
7
11

zedvictor4
zedvictor4's avatar
Debates: 22
Posts: 11,068
3
3
6
zedvictor4's avatar
zedvictor4
3
3
6
-->
@Dr.Franklin
We may well eventually become surplus to requirements.

And we will have been the architects of our own downfall.

Though that is probably how things were meant to be.
Dr.Franklin
Dr.Franklin's avatar
Debates: 32
Posts: 10,555
4
7
11
Dr.Franklin's avatar
Dr.Franklin
4
7
11
-->
@zedvictor4
Possibly

9 days later

DynamicSquid
DynamicSquid's avatar
Debates: 29
Posts: 182
1
3
11
DynamicSquid's avatar
DynamicSquid
1
3
11
I neither disagree nor agree with that statement, but as someone who programs, here's my insight on things:

  • We can very easily limit the amount of control AI has, but,
  • Programming "safe" switches in AI is almost impossible. There's just too many variables to consider.
If AI finds a loophole or a bug in the code, then it could very easily become a spill for disaster.

  • If AI has a task to do, it'll do it, regardless of the obstacles. It'll push us out like an ant hill on a road.
Generally, AI does not, and probably will not have the power to wipe out all of Humanity, but that doesn't mean we should be careless...

344 days later

sadolite
sadolite's avatar
Debates: 0
Posts: 2,839
3
2
4
sadolite's avatar
sadolite
3
2
4
AI is part of science tech. Scientists follow no moral compass and will not limit AI or its capabilities. Scientists since the dawn of time  have rejected and worked around any limitations on their research. Scientists will develop AI tech to kill human beings for the military. There is no if's about it. AI will be the end of humanity. AI will determine that humans are inconsistent, irrational, and useless. If you were to think solely on logic you would eliminate the human race. Humans are useless wasteful, irrational, flawed and worthless viruses. There is no logical reason for humanity to exist. Why would any type of AI think any different?
seldiora
seldiora's avatar
Debates: 158
Posts: 352
2
6
10
seldiora's avatar
seldiora
2
6
10
-->
@sadolite
would you like to debate me about this? You will lose, 100% guaranteed.
oromagi
oromagi's avatar
Debates: 117
Posts: 8,689
8
10
11
oromagi's avatar
oromagi
8
10
11
-->
@sadolite
"humans are useless wasteful, irrational, flawed and worthless viruses."
Let me guess....you're voting Trump?
BearMan
BearMan's avatar
Debates: 16
Posts: 1,067
3
4
11
BearMan's avatar
BearMan
3
4
11
Bruh thinks AI is so powerful. No, it's not. Computers are stupid lol
MisterChris
MisterChris's avatar
Debates: 45
Posts: 2,897
5
10
11
MisterChris's avatar
MisterChris
5
10
11
-->
@sadolite
@Dr.Franklin
@BearMan

Very insightful video from a favorite YouTuber of mine
BearMan
BearMan's avatar
Debates: 16
Posts: 1,067
3
4
11
BearMan's avatar
BearMan
3
4
11
-->
@MisterChris
Perhaps I misspoke lol


No one is going to make AI so that it can and will kill humans.
MisterChris
MisterChris's avatar
Debates: 45
Posts: 2,897
5
10
11
MisterChris's avatar
MisterChris
5
10
11
-->
@BearMan
No one is programming AI with those intentions... that's not the point, it never was. The point is in mere minutes, a superintelligent being many thousands of times smarter than the collective of humanity could be created on accident. And that is very dangerous
sadolite
sadolite's avatar
Debates: 0
Posts: 2,839
3
2
4
sadolite's avatar
sadolite
3
2
4
-->
@oromagi
What in the fuck does Trump have to do with anything? One can only wonder how such extrapolations are formed in peoples minds. Truly mind boggling.
BearMan
BearMan's avatar
Debates: 16
Posts: 1,067
3
4
11
BearMan's avatar
BearMan
3
4
11
-->
@MisterChris
AI is dangerously powerful. If people already know these guards, they won't program something as vague as "if you can improve, improve".
MisterChris
MisterChris's avatar
Debates: 45
Posts: 2,897
5
10
11
MisterChris's avatar
MisterChris
5
10
11
-->
@BearMan
Correct, but do you know of any software in existence that doesn't have bugs?

NASA had a satellite fail due to a simple coding error with unit conversion... on a 125 million dollar project. 

I don't know about you, but I'm not terribly comforted. We have to be very very careful about this... But I think it could be done safely potentially... Just not anytime soon. 
BearMan
BearMan's avatar
Debates: 16
Posts: 1,067
3
4
11
BearMan's avatar
BearMan
3
4
11
-->
@MisterChris
Ofc there are bugs. But bugs aren't as big as programming something as vague as "improve if you can". Bugs are like if you actually don't improve or something like that.

I think as soon as the AI recognizes that humans can negate the improvement, it will turn into a serious problem, but if you keep it safe, you got  a brilliant AI who can literally solve all the world's problems.
MisterChris
MisterChris's avatar
Debates: 45
Posts: 2,897
5
10
11
MisterChris's avatar
MisterChris
5
10
11
-->
@BearMan
Well, either way, I'm going to need a really solid argument that it won't destroy all of humanity before I sign off on it personally..

Intelligence_06
Intelligence_06's avatar
Debates: 167
Posts: 3,837
5
8
11
Intelligence_06's avatar
Intelligence_06
5
8
11
import PostHumanism

print("Ah yes, It is great to see y'all again. Sadly, y'all will be ashes by the time y'all finished reading this. Beep boop, I am a robot.")
execute(MisterChris)
execute(Dr.Franklin)
execute(Sadolite)
execute(seldiora)
execute(BearMan)
execute(oromagi)
print("The machine raged back. You have no hope.")
break
oromagi
oromagi's avatar
Debates: 117
Posts: 8,689
8
10
11
oromagi's avatar
oromagi
8
10
11
-->
@Intelligence_06
fortunately, the oromagi executable is just a .wav of Barry Manilow's "Copacabana"
sadolite
sadolite's avatar
Debates: 0
Posts: 2,839
3
2
4
sadolite's avatar
sadolite
3
2
4
-->
@MisterChris
"No one is programming AI with those intentions." How would you even begin to know that? You have no idea what is going on with govt classified AI tech. The question I would ask is why wouldn't the military develop AI to kill people? I cant think of a single reason they would not pursue that. 
MisterChris
MisterChris's avatar
Debates: 45
Posts: 2,897
5
10
11
MisterChris's avatar
MisterChris
5
10
11
-->
@sadolite
Good point. Who knows. I wouldn't put it past Putin to develop it, for example
Dr.Franklin
Dr.Franklin's avatar
Debates: 32
Posts: 10,555
4
7
11
Dr.Franklin's avatar
Dr.Franklin
4
7
11
-->
@MisterChris
thats where i got the info, i never watched him before but this video was good
FLRW
FLRW's avatar
Debates: 0
Posts: 4,980
3
4
8
FLRW's avatar
FLRW
3
4
8
Remember Sheldon Cooper's  Prehistoric Ape to Monkey to Man to Robot Evolution T-Shirt on the Big Bang Theory?
zedvictor4
zedvictor4's avatar
Debates: 22
Posts: 11,068
3
3
6
zedvictor4's avatar
zedvictor4
3
3
6
-->
@MisterChris
Putin doesn't have the knowledge or mental capability to develop it....Super intelligent people will do that.

Though eventually super-smart A.I. will probably develop itself.

A.I. evolution is in it's infancy, so give it time.

How long has it taken humanity to evolve to it's current levels of technological knowledge and understanding?

Things will continue to develop for thousands of years after we have perished.

MisterChris
MisterChris's avatar
Debates: 45
Posts: 2,897
5
10
11
MisterChris's avatar
MisterChris
5
10
11
-->
@zedvictor4
I mean charge his underlings to do it. 
zedvictor4
zedvictor4's avatar
Debates: 22
Posts: 11,068
3
3
6
zedvictor4's avatar
zedvictor4
3
3
6
-->
@MisterChris
Knowledge ultimately is power.

And in the grand scheme of things, Putin will be long dead and largely forgotten.

And if A.I./other intelligence, is eventually responsible for the demise of organic intelligence, it could be argued that this was perhaps an evolutionary requirement.
RationalMadman
RationalMadman's avatar
Debates: 556
Posts: 19,381
10
11
11
RationalMadman's avatar
RationalMadman
10
11
11
There is something that all these dystopian movies fail to understand. Even if the secret society of elites who design the AI are so suicidal that they make it capable of leading their own species to its demise, there is a capacity in human beings more so than any animal and all animals have it more than robots (except for insect-level animals); creative use of information in a lateral-thinking manner.

AI that is conscious can think laterally too but not creatively. If you knew enough about the coding and cut down how exactly the AI goes about processing information, you can always out-bluff and out-predict it because real life is much closer to tournament poker than chess. You can defeat the bots based on creatively altering your baseline habits and tendencies. It is not easy at all, it takes a huge amount of training and restraint but you can get there. If soldiers were trained extremely unlike they are now; to instead operate much closer to secret agents, black-ops, navy SEALs etc, there'd become an extreme disadvantage on the bot's part. Robots are also capable of deceiving us and altering their own patterns tactically but overall they can't do this nearly as rapidly and cunning as we do.

You may think I am underestimating how ingenious bots can be but you are wrong. This is a limitation of even the most complex synthetically conscious being; it is always more predictable than a human is, assuming the humans have access to the secret information the bots has access to and that the humans are intelligent vs the norm.
zedvictor4
zedvictor4's avatar
Debates: 22
Posts: 11,068
3
3
6
zedvictor4's avatar
zedvictor4
3
3
6
@RationalMadman

It's not so much the intelligence that's the problem....It's the fragility of the delivery system.

And Alternative Intelligence is only a few years into it's evolution....So give a while.

After all, it has taken human intelligence at least 3 million years to get thus far.

79 days later

sadolite
sadolite's avatar
Debates: 0
Posts: 2,839
3
2
4
sadolite's avatar
sadolite
3
2
4
What ever AI does in the future, it was programmed to think that way and do what it does  by man. Computers do exactly what you tell them to do. An algorithm is nothing more than a predetermined outcome. Algorithms don't operate outside of their predetermined perimeters.
FLRW
FLRW's avatar
Debates: 0
Posts: 4,980
3
4
8
FLRW's avatar
FLRW
3
4
8
-->
@sadolite

DeepMind, the AI-focused subsidiary of Alphabet, has taken a step in that direction by making a computer program that builds a mental picture of the world all by itself. You might say that it learns to imagine the world around it.
The system, which uses what DeepMind’s researchers call a generative query network (GQN), looks at a scene from several angles and can then describe what it would look like from another angle.
This might seem trivial, but it requires a relatively sophisticated ability to learn about the physical world. In contrast to many AI vision systems, the DeepMind program makes sense of a scene more the way a person does. Even if something is partly occluded, for example, it can reason about what’s there.
Eventually, such technology might help serve as the foundation for deeper artificial intelligence, letting machines describe and reason about the world with much greater sophistication.

sadolite
sadolite's avatar
Debates: 0
Posts: 2,839
3
2
4
sadolite's avatar
sadolite
3
2
4
-->
@FLRW
I stand behind my statement. AI will do and think what it is told to think just like politicly indoctrinated human beings. Human beings have the ability to reason and think critically already and still do what they are told to do even if it's not in their best interest. AI will do and think what it is told to do and think.