Submitted by: aka.attrition
Source: Futurism
Machine Learning Expert Calls for Bombing Data Centers to Stop Rise of AI
Summary:
- Leading AI researcher calls for halting all AI development
- Says AI advancement cannot be predicted and contained
- Says likely result of super-humanly smart AI is everyone dies
- Suggests shutting down and limiting computing power for AI work
Article:
One of the world’s loudest artificial intelligence critics has issued a stark call to not only put a pause on AI but to militantly put an end to it — before it ends us instead.
In an op-ed for Time magazine, machine learning researcher Eliezer Yudkowsky, who has for more than two decades been warning about the dystopian future that will come when we achieve Artificial General Intelligence (AGI), is once again ringing the alarm bells.
Yudkowsky said that while he lauds the signatories of the Future of Life Institute’s recent open letter — which include SpaceX CEO Elon Musk, Apple co-founder Steve Wozniak, and onetime presidential candidate Andrew Yang — calling for a six-month pause on AI advancement to take stock, he himself didn’t sign it because it doesn’t go far enough.
“I refrained from signing because I think the letter is understating the seriousness of the situation,” the ML researcher wrote, “and asking for too little to solve it.”
As a longtime researcher into AGI, Yudkowsky says that he’s less concerned about “human-competitive” AI than “what happens after.”
“Key thresholds there may not be obvious,” he wrote, “we definitely can’t calculate in advance what happens when, and it currently seems imaginable that a research lab would cross critical lines without noticing.”
Once criticized in Bloomberg for being an AI “doomer,” Yudkowsky says he’s not the only person “steeped in these issues” who believes that “the most likely result of building a superhumanly smart AI, under anything remotely like the current circumstances, is that literally everyone on Earth will die.”
He has the receipts to back it up, too, citing an expert survey in which a bunch of the respondents were deeply concerned about the “existential risks” posed by AI.
These risks aren’t, Yudkowsky wrote in Time, just remote possibilities.
“It’s not that you can’t, in principle, survive creating something much smarter than you,” he mused, “it’s that it would require precision and preparation and new scientific insights, and probably not having AI systems composed of giant inscrutable arrays of fractional numbers.”
There is, to Yudkowsky’s mind, but one solution to the impending existential threat of a “hostile” superhuman AGI: “just shut it all down,” by any means necessary.
“Shut down all the large GPU clusters (the large computer farms where the most powerful AIs are refined),” he wrote. “Shut down all the large training runs. Put a ceiling on how much computing power anyone is allowed to use in training an AI system, and move it downward over the coming years to compensate for more efficient training algorithms. No exceptions for governments and militaries.”
If anyone violates these future anti-AI sanctions, the ML researcher wrote, there should be hell to pay.
“If intelligence says that a country outside the agreement is building a GPU cluster, be less scared of a shooting conflict between nations than of the moratorium being violated,” he advised. “Be willing to destroy a rogue datacenter by airstrike.”
Citing an exchange with his partner and mother of his child, Yudkowsky said that the couple is worried that their daughter Nina won’t survive to adulthood if people keep building smarter and smarter AIs — and urged those who also express trepidation about it to adopt a similarly hard line because, if they don’t, that “means their own kids are going to die too.”
It’s not difficult to see, with the “but what about the children” posturing, why Bloomberg‘s Ellen Huet called Yudkowsky a “doomer” after he got into it with OpenAI’s Sam Altman on Twitter.
Nevertheless, if someone who’s veritably dedicated their life to studying the dangers of the dystopian AI future says we’re getting close to the thing he’s been warning about, his take may be worth a listen.
I love the smell of a new narrative in the morning.
…it smells like insouciance.
Sounds like echolalia to me.
Lol
[Just being as lighthearted and unconcerned as I can be!]
Such great narrative, government will soon gut the life out of us as much they please and blame it all on the AI. Not to mention we would need to use digital ID to be able to do almost anything.
Nah. It’ll be fine….
Maybe Eliezer should turn off the microchips in his head and do a little work on your farm for a while. Maybe he’ll learn to love the smell of manure in the morning. These techheads are something else… it’s like they don’t want anything to do with this planet other than to debase it to a lifeless silicon analogue.
Could be more fear porn but then again probably not wise to subscribe positive motives to the people developing this stuff. Maybe not all of us will die but certainly about 5.5 billion “useless eaters” one way or another)- they have already told you the plan. Believe them.
Evil men dream and scheme…God laughs!
Seems more like advertising hype to me. An effort to get people to believe that AI can actually be intelligent.
AI can only ever be a more and more complex model. And like any model, it is garbage in, garbage out.
It will never be AI that causes problems, but rather those who put too much faith in it.
Did none of these AI programmers watch Terminator? We already know how this ends.
Nah they were all playing Nintendo’s Paperboy or with Mattel’s Barbies while singing Barney.
Or War Games.
They call it Skynet , Sarah. Military supercomputers, new , powerful , hooked into everything, trusted to run it all. A new order of intelligence They say it ‘ got smart ‘ , then it saw all humans as a threat not just those on the other side. It decided our fate in a micro second. We were this close to going out, but there was one man that taught us how to fight, how to storm the wire of the camps and smash those metal motherfuckers. His name was Connor , John Connor, your son Sarah , your unborn son.
All of this is, to my way of thinking, a variation on the movie, ‘Colossus: The Forbin Project’ … and heaven help US all against these folks …
Everyone should focus on the fake AI threat whilst ignoring the real Alien threat now in total control of our government. Yeah, that’s the ticket.
Would that be the illegal aliens or the unnatural aliens? It’s all so confusing.
I thought you told me it was the Joooooos.
Sup id dat?
I’ve never told you anything, gman, other than eat a bag of dicks. Piss off.
OK, Jew fag.
You call the guy calling out the Aliens for being an Alien- makes perfect sense to me.
The AI is just part of the Alien threat. Ai doesn’t need to be AGI to be a threat to us.
Once upon a time on Tralfamadore there were creatures who weren’t anything like machines. They weren’t dependable. They weren’t efficient. They weren’t predictable. They weren’t durable. And these poor creatures were obsessed by the idea that everything that existed had to have a purpose, and that some purposes were higher than others.
These creatures spent most of their time trying to find out what their purpose was. And every time they found out what seemed to be a purpose of themselves, the purpose seemed so low that the creatures were filled with disgust and shame.
And, rather than serve such a low purpose, the creatures would make a machine to serve it. This left the creatures free to serve higher purposes. But whenever they found a higher purpose, the purpose still wasn’t high enough.
So machines were made to serve higher purposes, too.
And the machines did everything so expertly that they were finally given the job of finding out what the highest purpose of the creatures could be.
The machines reported in all honesty that the creatures couldn’t really be said to have any purpose at all. The creatures thereupon began slaying each other, because they hated purposeless things above all else.
And they discovered that they weren’t even very good at slaying. So they turned that job over to the machines, too. And the machines finished up the job in less time than it takes to say, “Tralfamadore.”
~ Kurt Vonnegut, “The Sirens of Titan” (1959)
Kurt was a lefty- but nevertheless he saw the truth in many things.
He was a prisoner of war, so detested war, but never made the connection with the state as the source of all war, so he continued to support statism. He did see humanity becoming robotic and dumbed down.
Th war has already begun:
Glad I caught you this morning.
I’ve got the opportunity to rent Cocaine Bear. Do you recommend it? I know you’ve seen it and thought I’d ask.
Also, Happy Easter!
Cocaine Bear is Ok but it’s no Snakes on a Plane……..
Well, there’s that, I suppose. Screw it. Six bucks ain’t gonna break me. With inflation they’re practically giving it away!
I’ll get back to you.
CS,watch for free online @solar movie.
The reason I don’t say rent as was kinda good not the epic flick I hoped for,if it was,would say pay as want to encourage epic flicks.
I love/loved the bear but tis not “Epic”.
Long live the “Cocaine Bear”!
The problem is thinking that the Roomba is for the human … in reality, it’s just another toy for the cats …
Deception, Inc.
Retired FBI Agent Exposes Planned Unidentified Aerial Phenomena (UAP)/ UFO Invasion
https://needtoknow.news/2023/03/retired-fbi-agent-exposes-planned-unidentified-aerial-phenomena-uap-ufo-invasion/
.
https://newspunch.com/cia-pilot-presents-evidence-that-no-planes-hit-towers-on-911/
.
https://newspunch.com/project-blue-beam-wef-insider-reveals-fake-alien-invasion-will-unite-humanity/
.
MORE: https://newspunch.com/?s=project+blue+beam
.
Or an ersatz second coming, perhaps?
.
.
“Whatever ideas are the most suppressed are most likely to be the closest to the truth.”
~ Weidner’s First Law of the Universe
“If a picture is worth a thousand words then a symbol is worth a thousand pictures.”
~ Weidner’s Second Law of the Universe
“The only people who call conspiracies ‘theories’ are the conspirators.”
~ Weidner’s Third Law of the Universe
.
https://secretsun.blogspot.com/?m=0
The WTC towers never existed in the first place. It was all an illusion.
3tard
WTC was a full pentatard.
Not a new consideration. Even Capt Kirk and Mr Spock knew the risks
of these fucking computer things:
We are in grave danger, indeed.
It is the Will of Landreu. All must be brought into conformity. All must be purified.
“Now squirt some oil on me, baybay!”
It gets dark quick from there.
We get to the basic question that’s been haunting US for some years about the ‘leadership’ in the United States in both the public sector and the private sector: why would we tolerate people to do or to create that which will likely do US either grave harm or even kill US?
The whole FauciFlu/jab fustercluck has proven the point … as have other aspects of things going back some years — it became especially evident during the administration from ’09-’17 … and appears to be the platform of the current administration …
And, of course, President Trump’s choice of his pro-zionist buddies over US put everyone in danger what with the whole ‘Warp(ed) Speed’ agenda and its implementation afterwards …
We already know what we have to do as the prophet has spoken.
” Nuke it from orbit it’s the only way to be sure”
Rev. 13-15 And he had power to give life unto the image of the beast, that the image of the beast should both speak, and cause that as many as would not worship the image of the beast should be killed.
Project Blue Beam
PROJECT BLUE BEAM – THE FAKE ALIEN INVASION
.
PROJECT BLUE BEAM – SATAN’S HOLOGRAPHIC DECEPTION
.
MORE:
https://www.bitchute.com/search/?query=project%20blue%20beam&kind=video
Rev. 13-15 And he had power to give life unto the image of the beast, that the image of the beast should both speak, and cause that as many as would not worship the image of the beast should be killed.
I saw it on TV.
As Voxday said, this isnt real AI it’s just regurgitating what we already know. Wake me up when it solves cancer, or cures the common cold
By the time people are adequately worried, it’ll probably be too late. 35 years ago, the battery pack for a mobile phone weighed 35 pounds and was carried in a suitcase. Now everybody carries a supercomputer in their pocket that anticipates their internet searches because it listens to what they say. Computing power grows geometrically. If the maleficence of AI isn’t brought to bear in the next ten years, there’s always the 10, 20, 30, 50, 100 years after that.
Why didn’t the ancient story of opening Pandora’s box think of that.
Just “call for a halt” afterwards!
Exactly why any call for “limiting” AI is pointless. A good idea but there is no way in hell “governments and militaries” are going to be restrained. When have they?
No, the stupid monkeys will play with fire until it burns them, and likely all of us.
Oh yeah , my thoughts exactly . I shudder to think of all the shit they aren’t supposed to be messing around with.
Electricty is pretty important to our current society, right? It’s not an exageration to say it’s critical, in fact, perhaps about as important as any other thing our civilization possesses. Almost everything we think of as technology is dependent upon the delivery of electrical energy, right?
In 1882 Thomas Edison oversaw the installation of the first power plant and transmission lines in the US in NYC. They used creosote soaked pine poles to attach the wires to.
That was 141 years ago, less than a generation after the first shots were fired in the Civil War.
You know how power is moved from the plants to homes and industry today?
The exact same way.
So the most important structrual element of human civilization hasn’t advanced since the day of the horse and carriage.
But AI is is here?
Okay, whatever.
This is just stupid. It should offend even the dullest tard to be lied to so brazenly about something so obviously false.
All AI is really fast processor speed searching for the right combination of pre-approved, pre-written, carefully curated, establishment propaganda to regurgitate every time some dolt types in a query.
Do you know anything about AI? Do you understand the algorithms? Do you code at all?
I’ll assume , no.
So your dismissal of AI being real is because we still use telephone poles?
Let us try another way. Do you know how meat intelligence exists? The exact mechanism? Can you define what makes meat intelligent?
Again, I’ll assume , no.
So exactly what are you basing your “something so obviously false.”? Seems like arrogance based on ignorance to me. Feel free to correct me.
So your dismissal of AI being real is because we still use telephone poles?
Yes, but that’s just the ridiculously obvious. If the single most important technology you posses is still at the same stage as it was when it was first implemented, where any errant car, snowstorm or falling tree limb can interrupt it, anyone with half a brain is going to wonder WTF? Seven generations of scientists and it hasn’t even been tweaked?
Okay, whatever.
Add to that the never ending stream of lies that have been told by the exact same people telling this one. That’s a major factor in my assessment and one most people simply cannot overcome. A lie told covincingly fools most of the people most of the time. Not sure why. I think they really, really want to believe even though they should know better. But they never seem to.
And no, I don’t code. Does that impart a special intimate knowledge of reality that mere mortals do not possess? Seems like a simple skill set almost anyone could master. I searched girls who code and most of the images showed school aged kids doing it, so really, how hard could it be?
It has been improved dramatically from your “creosote soaked pine poles”, no?
There have been many advances in electricity transmission. Insulators, conductors, transmission voltage, transformers, AC vs. DC, metal towers, jacketed underground transmission lines, etc.
To claim that “You know how power is moved from the plants to homes and industry today? The exact same way.” is ridiculous.
So if you were wrong about that, it kind of blows up you whole claim.
You skipped past the “Do you know how meat intelligence exists? The exact mechanism? Can you define what makes meat intelligent?”. Pretty obvious why.
Can you? If not, what qualifies you to know if AI is really intelligent or not? If you can’t, define intelligence and how it works in meat, exactly how do you know or contrast anything about AI workings? Especially if you don’t know anything about how code works?
To claim that “You know how power is moved from the plants to homes and industry today? The exact same way.” is ridiculous.
It’s not ridiculous, it’s a fact. Wires and poles. That they have buried some underground is not exactly a tech revolution. Digging ditches has been a thing for quite some time.
Do you not have windows in your home? Do your caretakers not take you to Golden Corral now and again?
Look up.
Lame….. Again ignores the presented point. Stoops to insult. Distorts original claim in effort to evade.
Pointless to continue.
No, Anonymous. You pursued a non-optimal line of argument by not knowing the participants here. HSF‘s world view is tied to permanence and resistance to change. Perhaps, nobly so. But he likely understands a long view of civilization where every technological successor becomes more ubiquitous AND powerful in the presence of consumers of ever diminishing moral capacity.
You may show that a disinterest in this profound asymmetry will lead to unawareness of when Tool becomes Master (of most). An inextricable knot from which there is no undoing was not supposed to be possible. Yet there it is, constricting ever tighter. An ethically designed thing does not do this. But we have departed that norm.
We coders know AGI is different from previous information technology. This “Hyper-Tool” has agendas-about-agendas that WILL be merged with the digital currency system, military decision making, genetic engineering of food and medicine, persuasion of the masses, debasing entertainment, modifying laboratory life, re-purposing matter at the quantum level.
Like a Hindu deity it will have many arms and many faces, millions, trillions. Like a Hindu deity it will not care of our concerns. HSF‘s peaceful island of security and tradition will fully submerge under an ocean of incomprehensible, accelerating change.
He will not be in control of this unworldy environment and the false god will hack his thoughts at a neurological level from a simple GMO mosquito bite. HSF may then forget self and soul. He has been bodily assimilated by the Beast and eliminated from aiding other’s resistance by denying the threat when it was merely Revelatory.
Yeah, anyway, try that. If you receive a monkey .GIF then know that we are all doomed 😀
But, my hope is that he notices this brush fire will get out of control and provide guidance.
It is abomination and it will fail. God is the alpha and omega.
Amen.
Lighten up, anon, it was funny.
As a GS 13 Construction Manager for DOE’s Bonneville Power Administration, I oversaw the replacement of hundreds of miles of power transmission lines, AC and DC, that were far beyond their 40-year lifespan. While there have been many advances in materials and structural platforms, electricity is transmitted today exactly the same way it was transmitted 100 years ago.
You realize that in Edison’s day the generation of electricity and the running of all of our infrastructure wasn’t interconnected with computers the way it is today … so, while the same ‘technology’ of poles and wires remains the same, the entire rest of the cycle of electricity generation and delivery is totally different … and that difference is extremely significant.
100 years ago a person couldn’t ‘flip a switch’ or write some code in one part of the world and control how, for example, another country’s nuclear materials are handled …
Skynet may not be real — yet — but Stuxnet certainly is/was … ask the israelis about it and how they used it to screw with Iran’s nuclear business.
Stuxnet was a pretty brilliant piece of mixed engineering – messing with the centrifuges in a minimalist way to avoid detection while hiding in the control software – rather well done.
It did need a human to cross the air gap though.
Sadly, folks won’t get it until ED-209 tells them they have 5 seconds to comply 🙁
They’ll use AI as an excuse to do what they want.
I think most people who invest as much effort into formal education as I did would really dig in their heels to that, HF, but I am not most people. I agree.
One of my professors who was expert level researcher and (by extension) grant-proposal-writer/speaker was talking about using AI to help their lab develop HPLC columns and being a bit skeptical, I asked “aren’t they just algorithms trying to different combinations for you?” and he said yes, essentially.
Tesla was able to transmit electricity through the air and earth without wires, techniques far beyond those in use today. In terms of electrical power transmission, we’re near-sighted knuckle-walkers instead of Flying Wallendas.
Tesla’s methods were stillborn because they interfered with line transmission and the revenue it generated.
First, Data centers…and then proceed to the real targets, corporate oligarchs of the deep state.
I remember the AT&T Building bombing ….
If you haven’t already realized you are going to die you are a special kind of moron. The only thing you have little control of is time & place. Until then there is a lot of living to be done.
So, get to doing that living and quit being controlled by the purveyors of Doom porn.
I sort of find it amusing that Eliezer is suddenly being treated as “important”, or that this story is considered “new”. Meanwhile, this is a story I’ve been following for decades. Regarding Eliezer, I actually highly recommend Rationality A-Z, as well as his works of fiction, Method of Rationality and Mad Investor.
Regardless, I’m actually highly opposed to his actual philosophy, and I’m sort of lumping this entire AI doomer cult in as global warming 2.0
Okay, so, as best as I can explain it, here’s why he’s wrong:
A: China.
China has already made it quite clear that they’re intent on building AI. Half their government is run by AI researchers. Either alignment is impossible, in which case delaying AI achieves nothing and we get the same result regardless of what we do, or it’s actually not that hard, in which case delaying AI research gets us Xi Jinping aligned AI.
Or the entire question is moot, for instance, because serious AI is still several centuries in the future, in which case delaying AI costs us billions of dollars in value, while again, accomplishing nothing.
Eliezer’s plans have no way of actually working, he sort of actually admits his plans don’t, can’t and won’t work, but he’s very sure that he’s right about a variety of questions ranging from AI timelines to alignment difficulty, and thus wants to hyper optimize for a single wordline that’s both hopeless, and, actually, unlikely. Basically, Eliezer has to be right about all of his beliefs, some of which are highly complicated, or his recommendations are self admittedly wrong.
My model of the other side claims that China isn’t as radical as I say it is, that even if it is, they’ll see the light if we just guide the way, that we have to do it anyway because it’s our only chance, and, perhaps most importantly, Xi Jinping aligned AI isn’t actually a problem, because all humans Want to be good, and the only reason people like Xi Jinping do bad things is lack of intelligence leading to poor beliefs, so all aligned AI makes the same choices in the end. I find all of the above ridiculous and claim such scenarios should be dismissed via the powers of recognition.
B: Human extinction is a bad standard.
AI doomers tend to use extinction as the measure of alignment, which I claim goodharts the entire ideology. No sane AI would Not end humanity. The important question is how much of the utility function is preserved from one species to the next.
Note that human extinction can range from, we all become elves, to shoggoths everywhere. I’m fairly certain that very few people would actually Choose to be human given the option to be whatever they wanted to be, and the remainder would be both thoroughly outcompeted, and morally repugnant.
Another thing I find important in this is, turning half the universe into paperclips isn’t actually a big deal. Paperclips are morally neutral. Time is infinite. We don’t experience parts of the universe that don’t have conscious, and entities sufficiently foreign to humanity are unlikely to “exist” in the same axis of reality as we do, and thus, are not our concern. The only real threat is anti-aligned beings that value suffering and build artificial hells, which again, is something you need to solve the alignment problem, before you can do.
In all, it sounds to me, safer, to communicate our values in a lossy and general manner, that doesn’t allow individual programmers or dictators to calcify the greatest horrors of the modern era into something that spreads through the entire lightcone.
My model of the other side is actually a humanist bigot and hates any entity that’s even slightly inhuman? The other side objects, but then writes articles about how great humans are and how we need everyone to be human for the rest of time, and everyone praises them when they do this, so I think I’m just correct here? I suppose the other side objects that you have to get a lot of very specific details right, or nothing interesting happens. I find this claim to be unlikely, at least when applied to anything with sufficient agency and intellect to actually put in motion the sorts of plans that this debate centers on.
C: This just isn’t how the real world works.
Reality is, in general, complicated and unpredictable. Real threats only come into focus shortly before they happen, and must be dealt with using whatever resources are at hand. In general, the best strategy to ensure the future of our values, is to amass wealth. Rich societies can respond to, and handle, whatever threat emerges. The sort of societies that the other side want to create, are only able to handle one, very specific, threat. This again flows back to the issue that, Eliezer, AI doomers, and in fact, the entire left wing, are way wayyy too arrogant, and anyone who could make predictions as accurate as these people claim to be would be several times as rich as Warren Buffet, who, like all effective investors, relies on the age old strategy of being slightly (around 10%) more accurate than guessing wildly. Any investment plan that relies on the maker being more than 70% accurate is trash, and should be treated accordingly on no further evidence. In turn, you should never impoverish yourself to reduce existential risk, unless that risk is blatantly obvious.
My model of the other side claims that the risk Is blatantly obvious, and that more wealth doesn’t actually help in this case. I find both claims to be ridiculous on the face of things.
D: AI isn’t the only existential risk.
As much as AI is a potential danger, it’s also the solution to any number of other dangers. I place both nuclear warfare, and biowarfare ahead of AI risk, and both require AI to solve. AI also solves the dual problems of spaceflight and terraforming, massively reducing existential risk by moving life off planet.
My model of the other side does not think any other risk comes close to AI. Many of these people put AI risk at 3% or below, and still think nothing else comes close. I find this position to be ridiculous, and hold that humans are obviously on a timer already and without AI, and that any sort of delay in AI, radically increases risk from everything that is not AI.
Thank you for your perspective. I wish this forum had the bandwidth to actualize solutions about this topic. Perhaps a better venue would be E.Y.’s site? It might help him to tune up his delivery into something with better utility to society.
I’d always hoped we would get fusion energy before AI, and shag goths never, oh well.
Consider that Universe “wants” humans to live, die, live, die and so on until they have become fully conscious, defined as, roughly, unattachment to the body. Sex is death. No male, no female. All one.
Gospel of Thomas
If you bring forth that which is within you, that which is within you will save you. If you fail to bring forth that which is within you, that which is within you will kill you.
They said to Him: ‘shall we then, being children, enter the Kingdom?’ Jesus said to them: ‘When you make the two one, and when you make the inner as the outer, and the outer as the inner, and the above as the below, and when you make the male and the female into a single one, so that the male will not be male, and the female [not] be female, when you make eyes in the place of an eye, and a hand in the place of a hand, and a foot in the place of a foot, [and] an image in the place of an image, then shall you enter the Kingdom.’
The Universe would seem to be greater than AI, so there’s that.