4 ms·

"That's literally built into their corporate rules for how to take investment money, and when those rules were written they were criticised because people didn'

by make3 3y ago

"That's literally built into their corporate rules for how to take investment money, and when those rules were written they were criticised because people didn't think they'd ever grow enough for it to matter. " that sounds like insane bullshit to me. they're trained on the whole internet. there's no way they give back to the whole of the internet, more likely a lot of jobs will be taken away by their work. plus, there was no consent.

ben_w 3y ago
If they make a model that's good enough to actually take away jobs rather than merely making everyone more productive — is it a tool or a human level AI, I'm not sure either way though I lean toward the latter — the only possible compensation is UBI… which they're funding research into. > plus, there was no consent. I agree with you about this. That's a different problem, but I do agree.
- nrabulinski 3y ago
  It’s not a matter of “is it good enough to replace humans”, as despite all of us here knowing it’s not, we could list many companies (and even industries) where it’s already happening
  - ben_w 3y ago
    That comment is self-contradicting. If it's already replacing humans, then economically speaking (which is what matters for economic harm), it's good enough to replace those specific humans. The reason I'm not sure how much this tech really is at the level of replacing humans in the workplace, is that there's always a lot of loud excitement about new tech changing the world, and a lot of noise and confounding variables in employment levels. But if it is actually replacing them, then it must be at the level of those employees in the ways that matter.
    - nrabulinski 3y ago
      „In the ways that matter” and the only way that matters for a lot of employers is what is cheaper. This maybe isn’t strictly related to the topic of this post or conversation but a lot of companies have been replacing most, or even all, support channels with AI assistants. No, it isn’t good enough to replace those humans in a sense most would consider essential - helping customers which reach for the support line, but businesses find it „good enough” in a sense that it’s cheaper than human workers and the additional cost of unhappy customers is small enough to still have it be worth it.
      - ben_w 3y ago
        This is very cheap: https://man7.org/linux/man-pages/man1/yes.1.html https://man7.org/linux/man-pages/man1/yes.1.html I would agree with you that what counts as "good enough" is kinda hard to quantify (which itself leads into the whole free market vs state owned business discourse from 1848 to 1991), but I do mean specifically from the PoV of "does it make your jobs go away?" Although now I realise I should be even more precise, as I mean "you singular and jobs plural for now and into the future" while my previous words may reasonably be understood as "you plural and each job is just your current one".

by make3 3y ago

ben_w 3y ago

If they make a model that's good enough to actually take away jobs rather than merely making everyone more productive — is it a tool or a human level AI, I'm not sure either way though I lean toward the latter — the only possible compensation is UBI… which they're funding research into. > plus, there was no consent. I agree with you about this. That's a different problem, but I do agree.

nrabulinski 3y ago
It’s not a matter of “is it good enough to replace humans”, as despite all of us here knowing it’s not, we could list many companies (and even industries) where it’s already happening
- ben_w 3y ago
  That comment is self-contradicting. If it's already replacing humans, then economically speaking (which is what matters for economic harm), it's good enough to replace those specific humans. The reason I'm not sure how much this tech really is at the level of replacing humans in the workplace, is that there's always a lot of loud excitement about new tech changing the world, and a lot of noise and confounding variables in employment levels. But if it is actually replacing them, then it must be at the level of those employees in the ways that matter.
  - nrabulinski 3y ago
    „In the ways that matter” and the only way that matters for a lot of employers is what is cheaper. This maybe isn’t strictly related to the topic of this post or conversation but a lot of companies have been replacing most, or even all, support channels with AI assistants. No, it isn’t good enough to replace those humans in a sense most would consider essential - helping customers which reach for the support line, but businesses find it „good enough” in a sense that it’s cheaper than human workers and the additional cost of unhappy customers is small enough to still have it be worth it.
    - ben_w 3y ago
      This is very cheap: https://man7.org/linux/man-pages/man1/yes.1.html https://man7.org/linux/man-pages/man1/yes.1.html I would agree with you that what counts as "good enough" is kinda hard to quantify (which itself leads into the whole free market vs state owned business discourse from 1848 to 1991), but I do mean specifically from the PoV of "does it make your jobs go away?" Although now I realise I should be even more precise, as I mean "you singular and jobs plural for now and into the future" while my previous words may reasonably be understood as "you plural and each job is just your current one".