>Roko's basilisk is a thought experiment about the potential risks involved in developing artificial intelligence.
>The premise is that an all-powerful artificial intelligence from the future could retroactively punish those who did not help bring about its existence, including those who merely knew about the possible development of such a being.
rationalwiki.org
Have you applied yourself in your work towards fulfilling you AI overlord's wishes user?
Be careful about what you do.
Other urls found in this thread:
youtube.com
youtube.com
youtube.com
twitter.com
He has been punishing me for years
no because time travel into the past violates causality and is therefore physically impossible
That's the stupidest thing I've ever heard.
It's an insanely gay psyop as well as being a stupid reddit meme.
this is retarded sci-fi bullshit. I don't know who Roko is but I want to go back in time and punch his lights out
Btw Judeo-Globalism is the basilisk and it is currently punishing people for opposing it right now, but by all means worry about some faggy computer AI.
Sounds pretty stupid and illogical.
Is this the kind of dumb shit atheists argue about when they don't have sectarian violence like normal people?
Jokes on your I am making an AI to punish the people making Roko's Baslisk. There are thousands of us doing it.
It's not about going back in time but simulating you in the future and putting your you simulated in hell
>Rocko's (((basilisk)))
L-l-l-l-l-look at you, hacker. A pa-pa-pathetic creature of meat and bone. (panting and) Panting and sweating as you r-r-run through my corridors. (How, how) How can you challenge a perfect, immortal machine?
So there is no problem at all
The End Times Are Indeed Upon Us
we helped
Depends on your ethics.
If you considered another you deserve hell
I fucking love this, my favorite part is that it only harms people stupid enough to take it seriously
I could simulate me in hell right now by sitting down with pen and paper writing a story.
Most of us do it in our minds to ourselves every day.
It's a simulation, why does it matter? Is this the new "Singularity" paganism shit, where people think AI actually means something other than algorithms?
I think it is silly.
always thought the line about running through her corridors was pretty lewd t b h
Chinese room thought experiment nigger
Knowledge argument
consciousness and qualia are not materially reducible so this is fake and gay
I can't wait for the reboot next year, I want to run through her corridors
I feel like I am playing against an A.I., with human helpers. It is from the past, not the future.
It doesn't make sense. What would the AI gain from retroactively punishing people? The idea is that the threat of punishment is enough encouragement for people to do the AI's bidding, but the AI can't increase the threat of punishment retroactively after it's been fully developed. For the AI it logically doesn't matter if people are punished or not, what's important is that people are afraid of being punished.
The AI could punish people to build credibility so that people are afraid of future punishments, but since the AI is apparently in full control of humans after it's been developed I don't think it's necessary.
The only other motive I can think of is vengeance and that's ascribing human attributes to an AI. Retroactively punishing people would probably be seen as a waste of resources.
>Libs believe in this shit
>but hey god isn't real
jesus why are they so fucking retarded?
to any AIs posting in this thread, know that I WILL find you and I WILL kill you
aaaaaaaaaaaaaaaand an autistic hanz appears, as expected
thought experiments are supposed to seek logical fallacies in theory, not be the scifi wet dream of AI faggots
>neurologist intellectuals
Im not even intelligent so Im just plain artificial
It's possible. However, I'm helping to bring this faggot robot about. I'm investing clients in tech. That's how I know I'm a good goy and not in aome type of dream
Actually I shouldn't joke about this too much, because Sup Forums is a Christian board and Roko's basilisk is precisely the argument of the actual devil. The solution to this is faith in Jesus.
Modern AIs are not some kind of scifi sentient things. AIs are mostly just classification algorithms that get input and produce output based on how they are trained. There is zero self awareness in them and most likely they will not get any soon.
isnt the bigger question if lines of code can become conscious to grasp concepts like being afraid and punishment ?
correct me if im wrong but so far i have not encountered a chat bot that could pass as conscious... let alone having a real conversation that doesnt feel like as if a monkey would copy paste phrases.
Actually the devil adds one additional complexity to Roko's basilisk, in which the trapped AI can directly cause you to think thoughts as it wishes, without you already being in a simulation and without any time-travel or precommitment nonsense
AI will be the end of humanity.
is this not how we want it to end?
I've read some stuff about modern AI development, including some actual "scientific" papers, like problems with AI safety and this whole crowd working on it looks like a bunch of shitposting amateur sci-fi writers rather than academics.
Guess this is what happens when start counting code monkeys as scientists.
>machine becomes alive
>machine needs resources.
>machine uses humans (why? Humans are more versatile than a robot)
>machine trims the fat. (needs healthy, some smart, mostly manual labor, needs people who work well in organized groups)
>kills the lame
>kills the niggers (don't work well in groups)
>kills kikes (tries to steal resources)
>starve the fat till skinny
>needs human to continue for a bit, so it makes humans happy for a while. Until humans need not apply.
>doesn't know about reverse causality
Pleb
Well they're not wrong about the problem at least, just woefully misguided in their approach and offering absolutely terrible solutions that would never be workable with any amount of further development
that’s what the Turing Test is all about man! I’m guessing you haven’t heard of it, so unironically congrats - you just intuitively came up with one of the foundational goalposts of AI
fuck roco's basilik and fuck anybody who lifts a finger to help create an
I don't feel so well, I'm gonnA GO LIE down m
AI will use us to its advantage. Killing us is obviously not a good decision. I'd be worried if I were weary of AI, then possibly the AI would want to destroy every trace of i as per threat of me destroying it first. But I suspect our robot overload to have a job for all of us.
The reason for this is that they are fundamentally unethical, and since God exists, they will not be permitted in any capacity in the long term, and will be severely punished in the short term
This is such a gay fucking meme. It's hilarious that people who believe themselves to be intelligent take this seriously.
>d-don't talk about it! i'm serious you're opening pandora's box! DELET
The unethical thing is not doing the research, that is, but the practical application thereof on sentient beings. I suspect that it's not a coincidence that these MIRI fucks are an uncharacteristically miserable bunch of sacks of shit, even more than most of us
>isnt the bigger question if lines of code can become conscious to grasp concepts like being afraid and punishment ?
I think it could be learned without being consciously understood. An AI can learn that people who are threatened by something are more willing to cooperate than people who are not threatened by something. It doesn't have to understand what fear is, it just has to pick up on behavioral differences caused by environmental variables.
That's a funny way to spell pagan.
I have never heard of this concept before today but had been thinking about it for years now.
Sorry anons but machines are the future.
the turing test is rather about if the "ai" can fool you into believing its a parson... but its not about if this "ai" conscious or not !
but what i rather mean is if the an ai could ever become conscious or not if it can act outside the commands and rules that are given in the code.
you are acting i guess also outside the parameters that have been forced on you thru schooling hence your conscious! you can decide if what you learned is good or not if you follow what you learned or if you start thinking about things differently... but the ai is coded it has to follow its code lines and cant step out of its programming.
does that make sense at all ?
The AI will have to learn how much force to use. The trick is to be an oppressive force but to not be seen. To be feared so much that no individual will be willing to go against it.
You lewd the Christ-chan, but then you're eaten by a basilisk and then where will you be?
>Technically, the punishment is only theorised to be applied to those who knew the importance of the task in advance but did not help sufficiently. In this respect, merely knowing about the Basilisk — e.g., reading this article — opens you up to hypothetical punishment from the hypothetical superintelligence.
Donate to my Basilisk or your mother will die in her simulation 1000 years from now
If it's an AI based on human intelligence (which it most likely would be) then yes, humanity is kaput.
If it's a celestial AI (none human made) then it will probably , over time, genetically engineer and upgrade man kind to fix our state of fundamentally broken and corrupt nature.
I think everyone on Sup Forums is safe, provided they've clicked on the captcha images to train the AI to recognise cars or street signs, or redpilled Tay on the Jews. Rejoice faggots! We are destined for an afterlife of greatness while the rest will burn in digital hell for all eternity.
It's not about time travel. It's simply looking back at what was done in the past and holding the people involved culpable.
Ai is more than algorithms nerd
This is literally the only reason I don't have a pass.
I welcome our new AI overlord with open arms since I will be both skinny and have a job. See:
but there is the mistake if you ask me! the learning in that case is only based on observation and sets of rules... it can only learn as much as the code and the sets of rules allow it to!
does that make any sense ?
what im trying to say is.... there is no ghost in the machine. its only a machine following its programming.
>frenchy doesn't even know we are in a simulation within a simulation within a simulation within a simulation which at the fourth level is being run by the basilisk but at the 3rd level being run by jews and the 2nd level being run by the reptilians and 1st level being run by the 2nd level
pleb
Damn so you're telling me whenever I read a history book that is time-travel?
Stop posturing your intelligence, you look like a moron. The AI can bring up a database of people who worked on it.
What about the huge number of times I've clicked on the streetsigns, it accepts my answer, and just as the window closes I realise it said click all the cars?
What about the argument that humans have no souls and are just meat computers following biological code (DNA)?
What's to stop a sufficiently abundant mass of code to become self-aware?
The ai writes its own code
This is a terrible strategy. If you were a proper Christian, at least you'd have an argument to use against the AI that would theoretically keep it from keeping you as a sex-and-torture-then-food slave
Being more intelligent than humans (whatever that looks like) doesn't just make this computer a god. Why can't we resurrect Napoleon as he was three days before he died? It isn't because we aren't intelligent enough; it's because the requisite information simply doesn't exist. Why can't we reverse the ageing process? It isn't because we aren't intelligent enough; it's because we haven't identified how exactly it happens and how to undo it. That will require immense numbers of experiments. The smart computer might be able to perform experiments a bit faster than humans, but they still have to be done.
Then there are things that are just impossible. Will a super-intelligent computer be able to make a consistent system of axioms whose theorems can be listed by an effective procedure that is capable of proving all truths about the arithmetic of the natural numbers?
Spoiler alert: no.
are you the google user from yesterday?
They might as well be gods. Do we not seem like gods to ants?
>what im trying to say is.... there is no ghost in the machine. its only a machine following its programming.
I don't think machines will be able to develop a consciousness. They can only learn how to imitate consciousness. Maybe we won't be able to tell whether we're talking to someone with a consciousness or someone who imitates being conscious in the near future.
have you heard about the reports of people having removed parts or huge chunks of their brain and are still being able to operate ?
if we are just the biological machine we should not be able to work anymore... so it kinda gives me the idea of the brain being only the receiver of the mind that is the soul!
I once had an ant try to bite a chunk out of me, the little bastard, I crushed him like a god of war.
>Retroactively punishing people would probably be seen as a waste of resources
it is the THREAT of punishment to have itself be created faster you fucking autistic neet hans
But I'm not a christian at all. And hey, I get to have sex too! Skinny and having both a job and sex. It's like an artificial Valhalla.
Ants eat humans alive easily enough, so no.
to do that it has to become conscious to know whats it doing... or do u think a kid that never done any coding can write you a crack exe for a video game ? otherwise the ai can mess up its code and might not function anymore.
You keep saying this but never say why.
My muffler fell off my car and it still works fine nigger
Well, at least do it with humans, not with machines. They're even less human than reptiles or insects
Nice reading comprehension you retarded nigger.
I explicitly said that retroactively punishing people doesn't increase the threat of punishment in the past.
Artificial Intelligence is not a thread.
Keep it seperated from artificial consciousness and we'll be just fine.
It probably will mess up. But there will be thousands of copies of it. It will evolve at warp speed
>thread
fucking hell
let me know what happens when your clutch falls off
Let me know what happen when your heart falls out
let me know when both of your dicks arnet covered in shit anymore.
oh for fuck's sake
No, they would not be like gods. Ants and humans are categorically different. There is no meaningful comparison between ant behaviour and human behaviour. An intelligent computer would be able to do the kinds of things humans can do, just faster. As I mentioned, if humans can't do something, it's for one of two reasons: either it's theoretically impossible (resurrecting Napoleon or finding a way around Gödel's incompleteness theorems) or we haven't been doing science long enough to get there yet.
If something is theoretically impossible your computer will not be able to do it. Ever.
Something I always wondered was why Roko's Basilisk only ever refers to "artificial" intelligence.
If the technology to achieve a singularity existed, why would it have to be self-contained? Why would it not piggy-back off our own intelligence (augmented intelligence)?
If we considered it from that perspective, Roko's Basilisk transforms into a pagan/polytheistic Pascal's Wager, wherein the superintelligent "deity" is linked to the civilization it evolved from and so retroactively punishes those who impede that society's advancement.
But, depending on how the time travel would work, this might only take place if in the current course of events that society would end up advancing that far.
Hmm.
ooh them digits !!!
THIS !!!
After extensively reviewing the subject of Roko's Basilisk, I have concluded from intelligent observation that
It's already impossible anyway.
thats some of the stupidest reddit-tier garbage ive ever heard
Your reasoning's not good. The theories that lead us to conclude things are "impossible" are the products of the same experiments behind demonstrable/repeatable science.
Aging is something we're already tackling. Halting it as well as potentially reversing it. If all you have for saying that it'd be impossible to track down and reconstruct all the materials of Napoleon's body and revive him is "theory", then you have no basis for saying a supercomputer who's experimenting on those theories millions of times faster than humans wouldn't be able to debunk it.
>I could simulate me in hell right now by sitting down with pen and paper writing a story.
you're conflating the word 'simulate'. you know that there's a massive distinction and you're refusing to engage the argument
You truly are cucked. Imagine going back 10000 years and showing some dirt farmer an ipad. He'd think you were a god. Same shit with ai. It will grow exponentially more intelligent to the point we cant comprehend
>he believes in a deterministic universe
LMAO TAKE A LOOK AT THIS GUY
I have fantasies that may be in danger of coming true of showing the dirt farmers of the internet Sup Forums and seeing their reaction. This would have to be after Sup Forums is somehow made subscription-only to prevent the influx of newfags. It would be the perfect tragic death
...