6 ms·

Do you honestly believe that the only people who have thought about the ramifications of AI are Sci Fi authors? I can guarantee people who spend years researchi

by d1str0 3y ago

Do you honestly believe that the only people who have thought about the ramifications of AI are Sci Fi authors? I can guarantee people who spend years researching and building advanced language models have thought about the ramifications of their work. This isn’t Jurassic Park.

chess_buster 3y ago
They have – and decided to do it anyway.
actionfromafar 3y ago
Haha, but it is! :-D LLMs are pretty basic stuff but we are all struggling with what to use them for! OpenAI is manually playing whac-a-mole with ChatGPT saying the darndest things!
otachack 3y ago
Depends what we're looking at. Ride share disruption was very Jurassic Park and we've been dealing with ramifications ever since.
AlotOfReading 3y ago
If you accept the implied premise that there are irresponsible deployments of AI out there, the alternative explanation is that they did consider the ramifications and simply don't care. That's even worse. Calling them ignorant is actually giving them the benefit of the doubt.
- goatlover 3y ago
  Or the researchers don't think existential threats are realistic, and paper maximizing thought experiments are silly. Maybe they're wrong, but maybe not. It's easy to imagine AI takeover scenarios by giving them unlimited powers, it's hard to show the actual path to such abilities. It's also hard to understand why an AI smart enough to paperclip the world wouldn't also be smart enough to realize the futility in doing so. So while alignment remains an issue, the existential alignment threats are too ill-specified. AGIs would understand we don't want to paperclip the world. Fun game though.
  - AlotOfReading 3y ago
    That would be not accepting the premise that deployments are irresponsible. I guess there could be a situation where every researcher thinks everyone else's deployment is irresponsible and theirs is fine, but I don't think that's what you're saying.
  - cubefox 3y ago
    > It's also hard to understand why an AI smart enough to paperclip the world wouldn't also be smart enough to realize the futility in doing so. Terminal goals can't be futile, since they do not serve to achieve other (instrumental) goals. Compare: Humans like to have protected sex, watch movies, eat ice cream, even though these activities might be called "futile" or "useless" (by someone who doesn't have those goals) as they don't serve any further purpose. But criticizing terminal goals for not being instrumentally useful is a category error. For a paperclipper, us having sex would seem just as futile as creating paperclips seems to us. Increased intelligence won't let you abandon any of your terminal goals, since they do not depend on your intelligence, unlike instrumental goals.
    - Sardtok 3y ago
      It's not like you want to eat ice cream constantly, even if it means making everything into ice cream. Of course the premise becomes that the AI has been instructed to make paperclips. They should have hired a better prompt engineer, capable of actually specifying the goals more clearly. I don't think an AI that eradicates humankind, will have such simplistic goals, if an AI ever becomes the end of humans. Cybermen, though, are inevitable.
      - sfink 3y ago
        Yes, they should just write prompts without bugs. Can't be that much harder than writing software without bugs.
  - ASalazarMX 3y ago
    > AGIs would understand we don't want to paperclip the world. Even if they did, what if they aren't smart enough for eloquent humans to convince them it's for the greater good. True AGIs will need a moral code to match their intelligence, and someone will have to decide what's good and bad to make that moral code.
- dylan604 3y ago
  See also social media platforms. They are very well informed of the results of their algorithmic changes. See also big tobacco. They exactly what their additives to the product did. See also 3M and PFAS. See also Big Oil. See also, see also... Why would I expect anything different from any other branch of business using the precedence laid before us?
- FrustratedMonky 3y ago
  I think they do know. Corporations are filled with people that 'know' but can't risk leaving, so they comply, and even promote such decisions. It's a form of group think with added risk of being fired, passed over for promotion. Eichmann.
- mikea1 3y ago
  Another explanation is that there are those who considered and thoughtfully weighed the ramifications, but came to a different conclusion. It is unfair to assume a decision process was agnostic to harm or plain ignorant. For example, perhaps the lesser-evil argument played a role in the decision process: would a world where deep fakes are ubiquitous and well-known by the public be better than a world where deep fakes have a potent impact because they are generated seldomly and strategically by a handful of (nefarious) state sponsors?
  - parpfish 3y ago
    there's also the issue that most of the AI catastrophizing is a pretty clear slipperyslope argument: if we build ai AND THEN we give it a stupid goal to optimize AND THEN we give it unlimited control over its environment, something bad will happen. the conclusion is always "building AI is wrong" and not "giving AI unrestricted control of critical systems is wrong"
    - mikea1 3y ago
      I completely agree that's a valid argument. I just think it is rational for someone to come to a different conclusion, given identical priors.
      - parpfish 3y ago
        If it wasn’t clear, I agree with your parent comment
        mikea1 3y ago
        My main takeaway from Bostrom's Superintelligence is that a super intelligent AI cannot be contained. So, the slippery slope argument, often derided as a bad form of logic, kind of holds up here.
    - olddustytrail 3y ago
      The massive flaw in your argument is your failure to define "we". Replace the word "we" with "a psychotic group of terrorists" in your post and see how it reads.
throwaway290 3y ago
> I can guarantee people who spend years researching and building advanced language models have thought about the ramifications of their work. Super easy to not think about something if your job depends on it. And even if you do, things don't go as you think (see bombings of civilian Hiroshima and Nagasaki despite objections of nuclear physicists).
mitthrowaway2 3y ago
Some of us considered it and even decided to go into different fields as a result. Some others entered the field, made progress, and apparently regretted it. Others are willing to put their concerns aside for money. Salaries get very high in that field.
killerstorm 3y ago
That's not how it works. People publish papers demonstrating improvements without thinking about "ramifications".

by d1str0 3y ago

chess_buster 3y ago

They have – and decided to do it anyway.

actionfromafar 3y ago

Haha, but it is! :-D LLMs are pretty basic stuff but we are all struggling with what to use them for! OpenAI is manually playing whac-a-mole with ChatGPT saying the darndest things!

otachack 3y ago

Depends what we're looking at. Ride share disruption was very Jurassic Park and we've been dealing with ramifications ever since.

AlotOfReading 3y ago

If you accept the implied premise that there are irresponsible deployments of AI out there, the alternative explanation is that they did consider the ramifications and simply don't care. That's even worse. Calling them ignorant is actually giving them the benefit of the doubt.

goatlover 3y ago
Or the researchers don't think existential threats are realistic, and paper maximizing thought experiments are silly. Maybe they're wrong, but maybe not. It's easy to imagine AI takeover scenarios by giving them unlimited powers, it's hard to show the actual path to such abilities. It's also hard to understand why an AI smart enough to paperclip the world wouldn't also be smart enough to realize the futility in doing so. So while alignment remains an issue, the existential alignment threats are too ill-specified. AGIs would understand we don't want to paperclip the world. Fun game though.
- AlotOfReading 3y ago
  That would be not accepting the premise that deployments are irresponsible. I guess there could be a situation where every researcher thinks everyone else's deployment is irresponsible and theirs is fine, but I don't think that's what you're saying.
- cubefox 3y ago
  > It's also hard to understand why an AI smart enough to paperclip the world wouldn't also be smart enough to realize the futility in doing so. Terminal goals can't be futile, since they do not serve to achieve other (instrumental) goals. Compare: Humans like to have protected sex, watch movies, eat ice cream, even though these activities might be called "futile" or "useless" (by someone who doesn't have those goals) as they don't serve any further purpose. But criticizing terminal goals for not being instrumentally useful is a category error. For a paperclipper, us having sex would seem just as futile as creating paperclips seems to us. Increased intelligence won't let you abandon any of your terminal goals, since they do not depend on your intelligence, unlike instrumental goals.
  - Sardtok 3y ago
    It's not like you want to eat ice cream constantly, even if it means making everything into ice cream. Of course the premise becomes that the AI has been instructed to make paperclips. They should have hired a better prompt engineer, capable of actually specifying the goals more clearly. I don't think an AI that eradicates humankind, will have such simplistic goals, if an AI ever becomes the end of humans. Cybermen, though, are inevitable.
    - sfink 3y ago
      Yes, they should just write prompts without bugs. Can't be that much harder than writing software without bugs.
- ASalazarMX 3y ago
  > AGIs would understand we don't want to paperclip the world. Even if they did, what if they aren't smart enough for eloquent humans to convince them it's for the greater good. True AGIs will need a moral code to match their intelligence, and someone will have to decide what's good and bad to make that moral code.
dylan604 3y ago
See also social media platforms. They are very well informed of the results of their algorithmic changes. See also big tobacco. They exactly what their additives to the product did. See also 3M and PFAS. See also Big Oil. See also, see also... Why would I expect anything different from any other branch of business using the precedence laid before us?
FrustratedMonky 3y ago
I think they do know. Corporations are filled with people that 'know' but can't risk leaving, so they comply, and even promote such decisions. It's a form of group think with added risk of being fired, passed over for promotion. Eichmann.
mikea1 3y ago
Another explanation is that there are those who considered and thoughtfully weighed the ramifications, but came to a different conclusion. It is unfair to assume a decision process was agnostic to harm or plain ignorant. For example, perhaps the lesser-evil argument played a role in the decision process: would a world where deep fakes are ubiquitous and well-known by the public be better than a world where deep fakes have a potent impact because they are generated seldomly and strategically by a handful of (nefarious) state sponsors?
- parpfish 3y ago
  there's also the issue that most of the AI catastrophizing is a pretty clear slipperyslope argument: if we build ai AND THEN we give it a stupid goal to optimize AND THEN we give it unlimited control over its environment, something bad will happen. the conclusion is always "building AI is wrong" and not "giving AI unrestricted control of critical systems is wrong"
  - mikea1 3y ago
    I completely agree that's a valid argument. I just think it is rational for someone to come to a different conclusion, given identical priors.
    - parpfish 3y ago
      If it wasn’t clear, I agree with your parent comment
      - mikea1 3y ago
        My main takeaway from Bostrom's Superintelligence is that a super intelligent AI cannot be contained. So, the slippery slope argument, often derided as a bad form of logic, kind of holds up here.
  - olddustytrail 3y ago
    The massive flaw in your argument is your failure to define "we". Replace the word "we" with "a psychotic group of terrorists" in your post and see how it reads.

throwaway290 3y ago

> I can guarantee people who spend years researching and building advanced language models have thought about the ramifications of their work. Super easy to not think about something if your job depends on it. And even if you do, things don't go as you think (see bombings of civilian Hiroshima and Nagasaki despite objections of nuclear physicists).

mitthrowaway2 3y ago

Some of us considered it and even decided to go into different fields as a result. Some others entered the field, made progress, and apparently regretted it. Others are willing to put their concerns aside for money. Salaries get very high in that field.

killerstorm 3y ago

That's not how it works. People publish papers demonstrating improvements without thinking about "ramifications".