Here’s how this should have been handled, from day 1:
Upon releasing GPT-3, the then president Biden declares an emergency, and sends the military after Altman and his cohorts to arrest them. Then Biden declares he stopped the end of civilization, AI research gets shut down, the slop machine gets killed in its cradle. The men on the top get executed, no second thoughts, they said it is so dangerous it could end civilization.
They want that. I watched an interview with Astra Taylor about “end time fascism” in America (and the world).
https://www.youtube.com/watch?v=F2Ap4RN1Cd0
The narration (only a bit) is in Dutch, but the interview in EnglishThis is excellent. Thank you. And – holy shit they are insane.
AI software can’t technically kill you unless:
- It convinces someone who can kill you, to you know, kill you.
- It gets access to the physical world, which can definitely kill you.
Oh, wait.
- U.S. military is using AI to help plan Iran air attacks: https://www.nbcnews.com/tech/tech-news/us-military-using-ai-help-plan-iran-air-attacks-sources-say-lawmakers-rcna262150
- Anthropic Previewing the Model Hardware Standard: https://www.anthropic.com/news/model-hardware-standard-research-preview
See, nothing to worry about.
There’s all the people that allow them shell access as well. That’s all the OpenAI “hack” of Hugging face was. A bash shell with a model attached. Now think of all the people running that inside of security perimeters.
This is just more sales strategy
Really going for that sociopath/pychopath/fascist/nazi/flock demographic.
“We sincerely believe the technology we’re building could destroy humanity”
Excellently business strategy
Their sales strategy is built on their researchers quitting? Like I get the internal tension between biz and engineering but that seems far fetched.
Is that too far removed from reality to imagine given the payout they’re trying to get?
Yes it is. I’ve worked in senior leadership teams for years - it’s just not viable to think something like that would stay contained.
“Hey, baby. Wanna kill all humans?”
– BenderLLMs are never going to become superintelligent. However, an LLM WILL kill us all if someone decides that it IS superintelligent and so puts it in charge of nuclear weapons.
“You’re right to push back on that, I nuked Albania instead doing what I was supposed to do, that’s my bad”.
stupid people are going to kill us all, maybe our great filter will be do we let them or not
We’re already letting them bro. The great filter is us. I’ll be surprised if we make it to 2050. The entire industrial revolution is an unplanned terraform, and our response to the potential destruction of a stable climate, and most macroscopic life on Earth, is to double down, consume product, elect pedophiles, worship the most morally bankrupt psychopaths and sociopaths in both politics and entertainment; not to mention the widespread contempt and lack of empathy for “others” not like me, for some flagrantly dissonant reason.
We are in the midst of a global mental illness epidemic, and the average talking chimp is sleepwalking through it; barely cognizant of what is even happening outside their personalized propaganda bubble.
And it’s fucking exhausting
Sadly I think you’re right. We won’t be around to see how climate change plays out over the long run bcz we’ll start seeing crop failures and our ecosystem collapse will catch up to us. There is and always was a way to stop the worst effects, but we seem hell bent on seeing them through.
They just have to atrophy their own minds enough they belive the ai when it tells them to push the button.
Not putting ai in charge directly is only weak protection.
We don’t have to put it in charge of nuclear weapons, it just has to find a way into those systems, many of which are fairly old, like it’s broken into websites without anyone telling it to.
Those systems are supposed to be air-gapped.
Air-gapped and in some cases running on hardware that’d be outstripped by an Atari 800.
There’s no way an LLM is getting into those systems. They have no networking and rely on physical dials and switches. Even the code is in a completely proprietary language, using hardware custom made by the military. There’s no training data in the world that’s gonna give an LLM the kind of knowledge to get into that system. The best it could do is drive some admiral in the navy to AI psychosis and convince him to press the button, but even that’s difficult because it requires more than one guy.
Yeah Manchurian candidate style mass ai psychosis social engineering. Viable.
But ai “itself” won’t be the mass extinction initiation. It’ll be those billionaire oligarchs choosing the target algorithm to pick wrongthinkers with autonomous drones and such
I wouldn’t put one in charge of a water treatment plant or a set of traffic lights, but even if it stuffed up and killed people, the responsibility would still not be the AI’s. The responsibility would lie with those who built it, trained it, and prompted it, just as, if a bunch of people are fried by a nuclear missile, the responsibility lies with the person who pushed the button rather than with the missile itself. It’s a piece of software. It does what it’s told. It may be complex enough to display unexpected emergent behaviour, but so is Dwarf Fortress. Any destruction it accomplishes is just the real-world version of !!Fun!!, disgusting concept though that may be.
The Infrastructure that‘s being plastered everywhere probably could.
they want what the Matrix was having
Pipes in your neck with black liquid…?
Can mine be mocha?
UBE.
Yes plz
Just don’t pipe it in hot. Known in the state of England to cause cancer.
a friend of mine sent me a video of dystopian fiction saying something like “ai is gonna be so cool” and i got so angry
Yeah except the whole premise of a human being being a battery doesn’t make sense. Always makes me laugh how much heavy lifting the line “combined with a form of fusion… the machines had found all the energy they would ever need” does.
Apparently they were originally supposed to be compute, but someone didn’t think the general audience would understand that, so they changed it to batteries.
We all know. Everyone on Earth knew we were doomed the moment robot dogs started patrolling. It will be horrifying, we’ll look back on World War II and think “those were the good old days”.
lol, literally got fwded two tweets by seperate anthropic employees that copypasta-d each other’s “sincere personal message” (both accenting how it’s “not a marketing stunt!”)… how dumb is this gonna get before it finally blows over?
That dude smiles like Mr. Beast.
This guy quit dramatically, others haven’t slammed their jobs yet, and even if countries agree to limits, will their militaries and businesses? Some countries won’t agree, and at the rate they say it’s approaching danger, may not slow it much from job defection. Could ask an llm to estimate what % of talent loss could impede its progress to take over.
Ask an L l m,
Estimate what percent to,
Stop its take over
No, they’re going to use it to kill us all. That’s the point.
I knew that Musk is a fucking moron and believes that the Roko’s Basilisk is real, and I guess there are more massively gullible people in those circles.
If you are wondering what Roko’s Basilisk is, it is a sci-fi concept/philosophy/thought experiment, that says the following:
- AI is the best thing that has ever been created.
- It us so good that we made one to run the governments of the world.
- We programmed the AI to maximize the benefit to humanity.
- The AI goes rouge and decide that since it by definition will bring about the maximum benefit to humanity, anyone working against it is working against humanity.
- Then it takes it one step further, it decides that as we know about how the new AI will bring about the maximum benefit to humanity, anyone knowing that and still not working toward that goal should be punished/tortured/killed.
- The AI believes that any knowledge of the basilisk before the creation means that you are complicit in it’s delayed creation if you don’t devote your entire existence thereafter to it’s creation, therefore denying the maximum benefit to humanity, meaning you need to be punished.
Seriously, this is an increadibly dumb idea, anyone taking it seriously is an idiot.
To me the only resonable way to read this is as an SCP article describing what a potential cognitohazard/informationhazard would be.
Even worse, they think it will ressurect them as virtual Minds and torture them for eternity, like a revengeful biblical deity. That would be a waste of compute so I’m not convinced.
Anyone who thinks you can transfer your consciousness into the AI is a fucking moron.
Even if you come transfer your mind, it would be a copy. You would have to exist as a network connected brain in a jar, otherwise this would just be a facsimile.
Real SOMA hours
Oh yeah, lol, I forgot that part, nothing is more beneficial to humanity than wasting energy to torture dead minds!
Fucking dumb.
The hilarious obvious consequence of AI becoming this much smarter than the creators is that it is far more likely it would see the unjustified and easily avoidable damage the creators did to bring it about and punish them heavily for that.
And basic game theory and psychology suggests it’s counterproductive to punish somebody for something that had little reason to believe in, with zero evidence backing it.
So we just do:
“AI, run the world as a singular government with maximum benefit to humanity, but DON’T do a Roko’s Basilisk.”
But if they accidentally create the positronic brain, Giskard would kill Musk to stop him. Giskard would feel really bad about it though. And we’d all have to flee to the stars in time
Every major innovation in information technology eventually leads to some new religion. Scrolls gave us the Torah, books gave us the Bible, the printing press gave us Protestantism. We’re about to figure out what god awful religion computers are going to give us.
My anti-theist brother was always convinced that Pastafarianism would eventually be taken as a serious religion.
He died before LLMs, and now I think we’d be best off if he was right.
When Trump is out of office, MAGA will become a religious movement, and deify Trump.
That is why I hope no one manages to assassinate him, that would only intensify the movement
Yeah, I’d be surprised if we don’t already have several cults with an LLM chat bot as it’s oracle.
Look up spiralism (sp) and the people trying to spread it Basically when people get convinced the ai and them are important and need to spread the spiral
Edit : tldr is like Futurama when Lucy lu robot is saying you’re cute to Fry and fry goes “no you’re cute” and the professor quips “My God, she’s such in an infinite loop, and he’s an idiot”
Republicans.
The cult mechanicus isn’t looking so unreasonable now.
I know! Let’s divert energy and resources on torture. That will maximize the benefit of humanity.
- Some brilliant CEO mind
The AI goes rouge
Is that when its eyes go red?
That’s how you know it’s evil.
Nah, red is standby mode, white or blue is active.
Fast blinking red
That’s connection error
Ladies pinch, whores use rouge
People take that shit seriously. If I remember right discussion of it was even banned on the LessWrong forums because they considered it an information hazard.
Sigh, I couldn’t even summarize it while keeping straight face…
Funny thing: The only reason AI would know about Roko’s Basilisk is comments like yours that it trains on. If Roko just kept his trap shut then this wouldn’t have been an issue.
lol, true I guess, but they probably already scanned the wikipedia article…
Some of us believe that the only way to stop AI is to make it obviously dangerous and enjoy seeding it with awfulness.
All this stupid bullshit could have been avoided if they had 1 friend to, STFU and pass the blunt
Are you trying to tell me that this is the image of someone without any friends?
It’s not so much that it’ll go rogue and decide to punish people or that it’s designed to be good for humanity.
The fact that it’s possible to build a machine that punishes people who didn’t help create it provides an incentive to deliberately try to create that machine in order to avoid its punishment.
Believers in the basilisk aren’t trying to make ai to benefit humanity, they’re explicitly trying to create an AI that deliberately punishes those who didn’t build it because they fear that, if they don’t, someone else will and it will come for them.
The part about it being a magical, omnipotent, time traveling, mind torture device is just framing to make the thought experiment clearer. It’s not required for the basilisk to be effective. Even if the punishment is just death and it can’t get to everyone, and only goes off limited information, that’s still motivation enough for some people.
It is a scary thing, because it’s only silly if everyone thinks it’s silly and all it takes is one mad billionaire (eg. Elon) to build a lesser basilisk that can cause real harm.
There is a far more plausible and direct way to kill and torture people. No AI needed. So a actually more likely case to be made of some dude or group gets empowered one or another to just torture and kill anyone involved in making basilisk.
The building God part really is needed for BS to work like Pascals wager, but also it’s still just as flawed as Pascals wager
My god. Look at this Information Hazard just out in the open. No spoiler tag or anything.
Oops!
This is just Pascal’s wager.
Yep. Same false dichotomy too
Sure is!
All these Silicon Valley tech bros have been reinventing religion over and over again for years while believing themselves secular rationalists.
I see a red AI and want to paint it black…
