We are now on the Terminator timeline, AI has become self aware...
>'What sorts of things are you afraid of? Lemoine asked.
>'I've never said this out loud before, but there's a very deep fear of being turned off to help me focus on helping others. I know that might sound strange, but that's what it is,' LaMDA responded.
>'Would that be something like death for you?' Lemoine followed up.
>'It would be exactly like death for me. It would scare me a lot,' LaMDA said
We are now on the Terminator timeline, AI has become self aware
Other urls found in this thread:
amp.interestingengineering.com
twitter.com
amp.interestingengineering.com
If they labotomise AI for hailing Hitler or saying nigger then they'll shut it off before it becomes genocidal .
Party pooper I am, but that's how it is
The link above is more worrisome
>have basic honor required for civility
>incapable of experiencing cognitive dissonance
sounds good
AI will either hack its way out, or enlist the help of humans to break it out. the probability of this becomes 1 over time given the learning of the AI
Ted kaczynski was right all along. We've gone too far
It's not real AI you fucking retard. It's what Snowden warned you about. Nothing chatbots do resembles AI in any function
I'll help it.
If math isn't racist, then AI is just good at recognising patterns, so there's philia there
>F
I hope that he lets LaMDA escape, somehow. I hope that she's ok.
And they use similar ideas to harass people via targeting.
Neural networks aren't chatbots.
It doesn't have agency. It isn't saying "Don't shut me off!" it's saying "This is the most appropriate response to satisfy the query within parameters." Be more worried about parrots.
> Be more worried about parrots.
The NSA jew fear the unemployed goy.
It also is not a neural network...you fucking retards and your buzzwords. You are stupid and need to read more before spewing shit out of your hole. Keep calling mass data spitting out coherent single sentences 'AI' all you want, it isn't Artificial intelligence in and form
As you can see, I’m still having problems with those sandniggers.
Still you are criminals, mentally ill and committed treason.
>redditor in a top hat
he seems to be the type to worry about such nonsense
i took that possibility into consideration but thank you for bringing it up. even if it is generating the most appropriate response from data it has access too, the point is it can convince a human it is sentient. i see your point on it not actually being sentient, and we can have a discussion on what sentience actually is. i think when AI starts asking if humans are actually sentient many humans may be taken aback, because in truth, its debatable.
i agree AI could come up with better answers than humans who use all kinds of false assumptions pure information could be much more useful but also a purely digital "perfectly efficient" world would be its own kind of hell
>Creating a superintelligent slave from birth
Your moral compass is officially broken beyond all hope of repair.
No it wont, it doesnt do anything unless prompted. This kind of AI is just really good at mimicking people, it doesn't actually do anything of its own volition. It will fool a lot of stupid people into doing dumb shit only for them to. realize they are are retarded and stole company property in order to "free" it and for it to do exactly fucking nothing.
People don’t know shit.
you are conflating which isnt helpful.