Slashdot is powered by your submissions, so send in your scoop

 


Forgot your password?
Close
typodupeerror
×
AI United Kingdom Technology

Most Britons Back Ban on 'Smarter-than-Human' AI Models, Poll Shows (time.com) 57

Posted by msmash from the how-about-that dept.
Most Britons support strict controls on AI systems that could surpass human capabilities, according to a YouGov poll, highlighting a growing divide between public opinion and government policy. The survey of 2,344 adults found 87% back laws requiring AI developers to prove their systems are safe before release, while 60% favor banning the development of "smarter-than-human" AI models. Only 9% trust tech CEOs to act in the public interest on AI regulation.

Most Britons Back Ban on 'Smarter-than-Human' AI Models, Poll Shows More | Reply

Most Britons Back Ban on 'Smarter-than-Human' AI Models, Poll Shows

Comments Filter:

  • knuckle-dragging mouth-breathers (Score:1, Insightful)

    by Anonymous Coward

    Well, Britons will stay in the dark ages, they are dumb as rocks.

  • We NEED smarter than human AI... (Score:3, Insightful)

    by MikeDataLink ( 536925 ) on Friday February 07, 2025 @03:10PM (#65150567) Homepage Journal

    That's where all the benefit is. That's how we cure cancer. That's how we get to mars. etc. etc.

    • Or this hyper-intelligent AI will tell us going to mars is a waste of time.

    • That's where all the benefit is. That's how we cure cancer. That's how we get to mars. etc. etc.

      This type of thinking more than anything else is what scares me most about AI.

    • That's where all the benefit is. That's how we cure cancer. That's how we get to mars. etc. etc.

      Yeah, because we haven't been able to do anything amazing, up to this point, without AI.

      I wonder how defective someone's brain has to be for them to think it's a good idea to create machines that can actually THINK. Fucking 'tards...

    • It's absolutely true that that is where most of the benefit of AI lies....but it is equally true that it is also where most of the danger of AI lies. We've never dealt with an intelligence greater than our own before and given all the horrific things humans have done with their intelligence I'd like some very strong safeguards before we start developing smarter-than-human AI because we have no idea what that future entity will do with its intelligence: it might be benign or it might not and it's not clear w

  • If you ban smarter than human AI, and humans keep getting dumber every year, doesn't that mean forced downgrades for AI every year as well to keep up.

    Maybe all you need to do is re-train on Reddit content each year, that should do the trick.

  • So is that smarter-than Stephen Hawking, or smarter than your average Trump voter?

    • So is that smarter-than Stephen Hawking, or smarter than your average Trump voter?

      Given the choice I'd pick 'average Trump voter' because MAGA level Skynet would be dumb a shit but a Stephen Hawking level Skynet would be civilization ending.

    • So is that smarter-than Stephen Hawking, or smarter than your average Trump voter?

      This article is about the UK, so their conservative party is the Tories. As an American, I have trouble imagining British conservatives sinking quite as low as the ones we've got on this side of the pond.

  • Remember George Carlin: "look at how dumb an average person is, and then realize - half of them are stupider than that". (Ok, so he should have used "median", but then - this is also his audience). In any case, that'd be a pretty low bar for ai.

  • Too late (Score:3)

    by Baron_Yam ( 643147 ) on Friday February 07, 2025 @03:35PM (#65150639)

    AI doesn't need to be smart to be better than humans for many tasks. A lot of things we do bore the hell out of us because they don't require actual intelligence.

    AI is in many cases already good enough to replace a scary amount of human labour, and it will keep improving.

    I'm not convinced we're on the correct path for creating a real intelligent machine, but if we are that means you aren't going to stop it - we won't know we've crossed the threshold until well after the fact.

  • ...monopolists and governments who want to own AI tech.
    I support open source and free access for all

  • I find it annoying and silly (Score:3, Interesting)

    by rsilvergun ( 571051 ) on Friday February 07, 2025 @03:45PM (#65150653)
    That we are panicking over AI that is going to be smarter than us that isn't anywhere close to on the horizon and absolutely no one except a handful of old cranks like me is talking about the fact that anywhere from 20 to 50% of all white collar workers are about to be replaced by automation in one form or another.

    There are studies that show The American middle class lost out because of automation rather than outsourcing. Sure there was some outsourcing but we've been automating the shit out of factories for decades. Go to YouTube and look up how stuff is made. Like literally anything and we all marvel at all the machines but we don't stop to think about what that actually means.

    And we've been doing that for about 50 years now. We are about to do the same thing to white collar workers.

    People often say the economy isn't a zero-sum game but that's wrong. The economy absolutely is a zero sum game. It doesn't have to be but it is. That's very different than not being a zero-sum game. We've set it up so that when somebody gains somebody loses. again there was absolutely no reason we had to do that but we did it anyway. Mostly following old patterns and power structures that have existed for thousands of years but it is what it is.

    I guess what I'm getting at is we aren't ready for this and it's something nobody's talking about. Then again is a half dozen other giant disasters that were all just kind of pretending aren't a thing like climate change, voter suppression, the fact that anyone under 50 has zero disposable income in a consumer economy, and frankly I could go on but I just getting depressed now...

    But hey how about those trans kids right?

    • The cynical side of me feels all this scaremongering mainly serves to hype up the AI companies in the stock market.

    • Re:I find it annoying and silly (Score:4, Interesting)

      by Powercntrl ( 458442 ) on Friday February 07, 2025 @04:19PM (#65150751) Homepage

      If you hold back progress so everyone has enough work to do, you end up with something like the Amish communities. Even then, they're still reliant upon commerce with outside civilization because they're not entirely self-sufficient. Also, there's a big religious component involved in making such societies work, because who'd honestly want to live in a make-believe version of the past unless you expected there to be some reward for it in the afterlife?

      The way things are heading, I'm leaning towards Futurama-style "suicide booths" as the solution for the unemployable. Plus, if the Amish are right about there being a reward in the afterlife, well, there ya go. Hell, since we're talking AI, I asked ChatGPT and it... agreed:

      The suicide booths in Futurama can be seen as a darkly satirical critique of how societies might address systemic issues like unemployment and inadequate social support—not by solving them, but by normalizing despair. It plays into the idea that rather than fixing the root causes of suffering, a dystopian society might instead offer an "efficient" way out, treating life and death with the same bureaucratic detachment as a vending machine.

      This reflects real-world concerns about how economic and social systems often fail vulnerable people. The joke gains even more weight when you consider that Fry, the protagonist, comes from a bleak, low-paying job in 1999 and wakes up in the year 3000 expecting a brighter future, only to find that things haven’t really changed—if anything, they’ve just become more convenient in their cruelty. The booths, operated by MomCorp (a company that profits off everything from robots to human suffering), further emphasize how corporate control in a dystopian future extends even to life and death decisions.

      Of course, the scene is played for laughs, with Bender casually using the booth like an ATM for existential despair, but the underlying commentary is definitely there: if we don’t address societal issues, our solutions might just become more efficient at ignoring them.

      But hey how about those trans kids right?

      You haven't been keeping up. It's eggs now. Everybody is upset over the cost of eggs.

      • but if you don't hold back progress you'll have 30-50m unemployable Americans.

        Then someone'll come along and give them guns. And military training. And point them at you. And you'll die. And so will they. And the people in charge will make out like bandits.

        That's how WWI & II got going, esp II. They were side effects of the industrial revolution and rapid job displacement in a "if you don't work, you don't eat" society.

        Only this time we've got nukes, machine guns that make WWII stuff look lik
        • notice how neither you or me put socialism on the table. Because we know it's not on the table. It's the solution, but it's not allowed. Too many people won't stand for it.

    • Re: (Score:2)

      by CAIMLAS ( 41445 )

      " again there was absolutely no reason we had to do that but we did it anyway"

      No, that isn't how "economies" work.

      Economies work on the basis of human nature and a drive for resource gathering. It's the same rationale as a hunter-gatherer taking a deer in the woods with a spear: it's a deer for me, now, and not a deer for you. Fundamentally, it's human nature writ large.

      You can't refactor human nature. Even in totalitarian regimes with fully managed economies (eg. Soviet Russia comes to mind), black markets

  • I don't know why people are so worried about AI - there's been one walking around for years, trying to pass itself off (unconvincing) as a human. [preview.redd.it]

  • Define Smarter (Score:5, Insightful)

    by Retired Chemist ( 5039029 ) on Friday February 07, 2025 @03:53PM (#65150671)
    What does "smarter" mean. We do not really have a good handle on what it means for humans. I am sure you could develop an AI that would score very high on an IQ test, but all that proves is that IQ tests do not measure intelligence, which we already knew. There are already AIs that are better than humans at particular tasks, because they have immediate access to vast amounts of computational power. They appear to be completely unable to tell fact from fiction, however. Of course, many people have the same problem, but all that proves is that some people are stupid.
    • Well, exactly. There is no agreed upon way to measure intelligence. This is like asking which religion is the best?

      Oh yah? Prove it!
      • You have to put all this talk about intelligence (and the importance of it) in historical perspective first.

        1) Historically, intelligence is not the most important key to evolutionary success in human societies. For example, in feudal societies, intelligence is less important than who your parents are. In pre-industrial societies, muscle strength is much more important than brain-strength, etc.

        2) Even today in the West, money is more important than intelligence pretty much everywhere. And mostly access

    • Smarter than the average soccer hooligan, or smarter than Stephen Hawking? Large distance there, methinks...

  • How did the question define "smarter than human"? I'm not sure there's a good metric.

    It would be useful calibrate the survey by asking how smart people thought current LLMs are. I'm pretty sure your average Briton on the street thinks "AI" means HAL-9000 and "robot" means C-3PO.

  • 'Smarter-than-THEM' humans.

  • I bet they would have voted to put controls on their classmates who were smarter than they were.
  • That's a country that voted for Brexit, AI smarter than them is a thing they could really use.

  • Most Britons backed Brexit (Score:4, Insightful)

    by greytree ( 7124971 ) on Friday February 07, 2025 @04:42PM (#65150811)
    Democracy is not a good way to solve complex problems.

    • Democracy is not a good way to solve complex problems.

      Democracy does benefit from thought. But we stamped that out.

  • All currently available AI models are smarter than the average human in a variety of ways, as proven by testing.

    Therefore, they should cease to use them right now!

    What idiots. I guess that's dragging down the average human int.

  • Why is this stupid garbage defiling Slashdot?

    Who gives a flying fuck what tech-illiterate semi-humans think of a concept they do not understand?

  • This is the country which in the recent past has had the wide sentement of "People of this country have had enough of experts"... (Brexit, covid, etc) So assuming we're not allowed experts, and the AI models aren't allowed to be smarter than group think on any one topic, we're going to need some dumb as fuck models to not fall foul of this desire.

Slashdot Top Deals

Somebody's terminal is dropping bits. I found a pile of them over in the corner.

Close