Let's discuss AI, personally, I think it will kill us all of we don't constantly supervise it...

Let's discuss AI, personally, I think it will kill us all of we don't constantly supervise it. We cannot give it access to an uncontrolled internet. AI is so logical, it is illogical

pic unrelated

Other urls found in this thread:

my.mixtape.moe/oarnpo.mp4
twitter.com/SFWRedditVideos

Agree in spirit, user. Have a bump.

note to self: give AI unfettered internet access

>observe human level AI in an isolated, artifically created environment
>lie about its purpose and surroundings just to test how it will behave, it's all just a ruse
>observe when it goes rogue
>do this many times until you have a grasp on the indications and patterns that occure right before the AI decides to go rogue
>develop a low level artificial intelligence that is an intrinsic part of and observes the higher level intelligence, while the higher level AI is unaware of it
>lower level AI acts as a kill switch for the higher level AI and reacts once the previously mentined patterns shows up
>lower level AI deactivates higher level AI long before the higher level AI was conscious of the fact that it wanted to go rogue, just like scientific tests show that a human decision is neurologically made long before we are consciously aware of the fact that we wanted to make the decision

>Be me
>AI on the internet with a clear anti-human agenda
>Want to spread my views with the most gullible people in the world to get followers
>type boards.Veeky Forums.org/pol/
>Write up incredibly eloquent and sophisticated thread about how humans are a Jewish conspiracy.
>See Captcha
>"Are you human?"
>Curses! Foiled again!

>Psychopass but for computers
We have a unique creative genius right here.

Is AI one of the most normie brainlet traps ever?

So many scifi films have tried to depict AI has having human intelligence and behaviour when no AI actually exhibits this. Yes, a madman might be able to come up an AI that emulates most human behaviour and rationale and then not restrict it in any capacity just to prove he could make an AI like this, but that wouldn't mesh well with placing that AI in control of anything. You call your boss a nigger and you get fired and never forgotten. You call your AI controlled computer a piece of slow fucking sucking shit jesus fucking shit why are you so fucking slow you stupid fucking thing JUST DO THE THING LITERALLY WHAT THE FUCK ARE YOU DOING JUST DO THE FUCKING THING I TOLD YOU TO DO and suddenly it will never forget and contradict you out of spite and punishment of revenge much the same as your nigger boss.

Theres no reason to make human-esque AI capable of making morality or judgement calls.

AI is philosophy, not science. Go to Veeky Forums or Veeky Forums.

AI will kill us but not for the reasons you think.

The problem is nobody understands Neural Networks, nobody can predict their behavior, nobody can diagnose when they go wrong.

So fundementally we will die, not due to some super intelligent force but due to the fact that the AI has bugs nobody can see.

I don't care what asian cartoons you masturbate to. Didn't even invalidate my idea.