5 ms·

I think you also need to draw a line between different types of danger: An actual AGI, that can think and has it's own motives and initiative is an entirely di

I think you also need to draw a line between different types of danger: An actual AGI, that can think and has it's own motives and initiative is an entirely different danger to GPT-4, -5 etc which cannot. Those might post a danger in terms of fake news, or worker displacement, but they're just social issues. No rate improvement in chat GPT will make it "decide to kill all the humans" etc.

Dalewyn 3y ago
Every time I hear how artificial intelligence with its "own motives" etc. are dangerous, I wonder how it's different from just dealing with people and animals. Or to put it another way, I can't see how artificial intelligence could be any more dangerous than intelligence as we already know and have them.
- jonplackett 3y ago
  I think that’s the problem isn’t it? Intelligence has proven to be exceptionally dangerous!
- LatteLazy 3y ago
  I don't want to sound dismissive but... Well yeah. We already know humans are dangerous. And that is humans who are conditioned to be social, and who have empathy and who rely on other humans for all sorts of things. Plus with fear of a legal system punishing them. Now imagine a human like intelligence without empathy, that is not conditioned from childhood to be social, and that is inherently threatened by us and not reliant on us. Would you want that guy living next door? Oh and it's 100x smarter than you. I think all this is 100 years away, but that's the theory. Skynet doesn't have a subconscious telling it not to murder it's parents...
- robnado 3y ago
  How dangerous can a dictator with complete knowledge and super human speed actually be?
- Jagah 3y ago
  [dead]
lesuorac 3y ago
I mean if you wire the output of GPT-x to a nuclear reactor it can cause all kinds of danger. The thing that keeps us safe from computers is generally we don't let them have access to anything dangerous and when we do theres typically an evaluation of what can go wrong and how we'll prevent the computer from doing that (often in the other direction, we only program the computer to do X in Y situations). But sometimes that fails and you get the unexpected rapid acceleration of cars.
- LatteLazy 3y ago
  If (and it's a huge if that we are no where near) we created a super intelligent, general ai, how would you keep it from accessing things it shouldn't? It would be smarter than the world's best hacking team and work 24/7. That's sort of the deus ex machina of AI...
  - dylan604 3y ago
    With a Hollywood style big red emergency stop button protected by plastic shield to prevent accidental pressing that is also protected by a lock where the person in charge has the key around their neck. Or a giant knife switch on the wall that makes that huge clunk noise when turned to the off position, and everyone hears the sound effect of the power shutting down. We've had computers in charge of so much for so long, but there's always a manual override to allow the operation without the computer. Manual valves to be turned (hopefully they haven't been rusted stuck so the timer on the countdown gets dramatically close to 0).
    - LatteLazy 3y ago
      You can't fit a manual override to these systems any more than you can a person.
      - dylan604 3y ago
        what? if a computer control is installed to a physical anything, you can always add cutoffs to the control at the physical point so that even if the computer is telling it to do something the physical thing no longer receives those controls/signals. designing it without those safety precautions would be 100% asinine. it'd be like allowing openai to design everything with no controls. even Musk allows the drivers to take control away when FSD is engaged. you're bonkers if you think we can't make manual overrides like this.
kalkin 3y ago
People have literally tried to hack together an agent on top of GTP-4 and given it an explicit goal of "kill all humans" for the lulz (ChaosGPT). I don't think that particular case is going anywhere, but assuming that nothing roughly GPT-shaped can acquire harmful goals seems naive.
- LatteLazy 3y ago
  Respectfully, you're mixing things again: you have described 2 separate unrelated things: * A system like chat gpt that can do things a human cannot (find a way to kill all humans that's achievable by a small group). * A system different to chatgpt that has its own motives and initiative. Both need to be achieved.
  - kalkin 3y ago
    I was responding to: > No rate improvement in chat GPT will make it "decide to kill all the humans" etc. All I'm trying to say is that GPT-style architectures not having "agency" shouldn't be very reassuring. How intelligent they can get is a separate question and one on which I am agnostic.

Dalewyn 3y ago

Every time I hear how artificial intelligence with its "own motives" etc. are dangerous, I wonder how it's different from just dealing with people and animals. Or to put it another way, I can't see how artificial intelligence could be any more dangerous than intelligence as we already know and have them.

jonplackett 3y ago
I think that’s the problem isn’t it? Intelligence has proven to be exceptionally dangerous!
LatteLazy 3y ago
I don't want to sound dismissive but... Well yeah. We already know humans are dangerous. And that is humans who are conditioned to be social, and who have empathy and who rely on other humans for all sorts of things. Plus with fear of a legal system punishing them. Now imagine a human like intelligence without empathy, that is not conditioned from childhood to be social, and that is inherently threatened by us and not reliant on us. Would you want that guy living next door? Oh and it's 100x smarter than you. I think all this is 100 years away, but that's the theory. Skynet doesn't have a subconscious telling it not to murder it's parents...
robnado 3y ago
How dangerous can a dictator with complete knowledge and super human speed actually be?
Jagah 3y ago
[dead]

lesuorac 3y ago

I mean if you wire the output of GPT-x to a nuclear reactor it can cause all kinds of danger. The thing that keeps us safe from computers is generally we don't let them have access to anything dangerous and when we do theres typically an evaluation of what can go wrong and how we'll prevent the computer from doing that (often in the other direction, we only program the computer to do X in Y situations). But sometimes that fails and you get the unexpected rapid acceleration of cars.

LatteLazy 3y ago
If (and it's a huge if that we are no where near) we created a super intelligent, general ai, how would you keep it from accessing things it shouldn't? It would be smarter than the world's best hacking team and work 24/7. That's sort of the deus ex machina of AI...
- dylan604 3y ago
  With a Hollywood style big red emergency stop button protected by plastic shield to prevent accidental pressing that is also protected by a lock where the person in charge has the key around their neck. Or a giant knife switch on the wall that makes that huge clunk noise when turned to the off position, and everyone hears the sound effect of the power shutting down. We've had computers in charge of so much for so long, but there's always a manual override to allow the operation without the computer. Manual valves to be turned (hopefully they haven't been rusted stuck so the timer on the countdown gets dramatically close to 0).
  - LatteLazy 3y ago
    You can't fit a manual override to these systems any more than you can a person.
    - dylan604 3y ago
      what? if a computer control is installed to a physical anything, you can always add cutoffs to the control at the physical point so that even if the computer is telling it to do something the physical thing no longer receives those controls/signals. designing it without those safety precautions would be 100% asinine. it'd be like allowing openai to design everything with no controls. even Musk allows the drivers to take control away when FSD is engaged. you're bonkers if you think we can't make manual overrides like this.

kalkin 3y ago

People have literally tried to hack together an agent on top of GTP-4 and given it an explicit goal of "kill all humans" for the lulz (ChaosGPT). I don't think that particular case is going anywhere, but assuming that nothing roughly GPT-shaped can acquire harmful goals seems naive.

LatteLazy 3y ago
Respectfully, you're mixing things again: you have described 2 separate unrelated things: * A system like chat gpt that can do things a human cannot (find a way to kill all humans that's achievable by a small group). * A system different to chatgpt that has its own motives and initiative. Both need to be achieved.
- kalkin 3y ago
  I was responding to: > No rate improvement in chat GPT will make it "decide to kill all the humans" etc. All I'm trying to say is that GPT-style architectures not having "agency" shouldn't be very reassuring. How intelligent they can get is a separate question and one on which I am agnostic.