8 ms·

How do we protect people’s livelihoods without engraining regulatory capture? Interested in HN’s thoughts. <s> It won’t matter in a few years when AGI kills us

How do we protect people’s livelihoods without engraining regulatory capture? Interested in HN’s thoughts. <s> It won’t matter in a few years when AGI kills us all. </s>

pyuser583 2y ago
We wait until we have an actual, real problem. Then we try to fix it. Right now, AI-induced unemployment just isn’t a problem. So it’s extremely likely the advocates for regulation are AI corporations, rather than reams of unemployed workers.
- brookst 2y ago
  And before somebody says “we should address problems before they happen” the challenge with AI is we don’t know exactly what problems will happen or what effective regulation would be. It makes sense to say new buildings have to meet fire safety regulations that are based on learnings from past fires. AI is too new; everyone is either guessing or lobbying for personal benefit. We should be vigilant, but I don’t for a second believe the “any regulation is better than nothing” rhetoric from those who stand to gain.
  - troupo 2y ago
    > AI is too new; It's not. All the learnings we've had from ML (the mistakes, the failures etc.) are just as applicable to AI.
    - brookst 2y ago
      So what learnings do we have that inform whether it is a good idea to regulate e.g. whether a business can use AI to take customer orders?
      - troupo 2y ago
        You can read through EU AI Act, or read this excellent article about it: https://softwarecrisis.dev/letters/the-truth-about-the-eu-act/ https://softwarecrisis.dev/letters/the-truth-about-the-eu-ac...
- xkcd-sucks 2y ago
  Maybe overly cynical but "risks" of "AI" going to Congress seems mostly to be a submarine hype thing
- dbspin 2y ago
  "AI-induced unemployment just isn’t a problem." I'm not sure how you'd test this hypothesis, but it certainly looks like a tonne of the ongoing FAANG layoffs are based on assumptions about a decreased need for low-mid level programmers.
- ben_w 2y ago
  When governments make decisions in a rush, the decisions are generally not very well thought out. I therefore think it's a good idea to have a plan for AI induced unemployment before people take to the streets because they have no job, no hope of a job, and no money left to pay the bills. I have no idea how far away economic mass unemployment is, and I'm not convinced the current indicators in specific fields (much as it affects those fields) is "it", but I have seen a lot of news stories about governments getting surprised by things everyone was warning them about and they ignored.
nosefurhairdo 2y ago
Should we have protected the livelihoods of horse carriage manufacturers when cars were created?
startupsfail 2y ago
Some amount of regulation, in particular in regards to reporting biases installed into the systems is probably worth some merits. What, if, for example, we’ll have a mass information system that would start pushing something that is very beneficial for special interest communities (say community is furries and this increases their dating pool), but tends to result in higher level of depression and suicide in teenagers. And then we also find that overall the system reduces mortality, by providing high quality medical advice. Surely shutting down the system would not be a good idea. And requiring private enterprise to change their messaging infringes on free speech. But at least requiring to properly disclose biases is likely a good idea.
int_19h 2y ago
If you mean it in a sense of "what about Skynet?", then you regulate specific uses of AI which carry such danger rather than AI itself. I don't think this even needs to be AI-specific; if we, say, required formal verification of decision making algorithms to match the spec, that would already cover it. If you mean it in a sense of "what about all the lost jobs?", then we should start seriously talking about UBI, and how to fund it by taxing automation.

How do we protect people’s livelihoods without engraining regulatory capture? Interested in HN’s thoughts. <s> It won’t matter in a few years when AGI kills us all. </s>

pyuser583 2y ago

We wait until we have an actual, real problem. Then we try to fix it. Right now, AI-induced unemployment just isn’t a problem. So it’s extremely likely the advocates for regulation are AI corporations, rather than reams of unemployed workers.

brookst 2y ago
And before somebody says “we should address problems before they happen” the challenge with AI is we don’t know exactly what problems will happen or what effective regulation would be. It makes sense to say new buildings have to meet fire safety regulations that are based on learnings from past fires. AI is too new; everyone is either guessing or lobbying for personal benefit. We should be vigilant, but I don’t for a second believe the “any regulation is better than nothing” rhetoric from those who stand to gain.
- troupo 2y ago
  > AI is too new; It's not. All the learnings we've had from ML (the mistakes, the failures etc.) are just as applicable to AI.
  - brookst 2y ago
    So what learnings do we have that inform whether it is a good idea to regulate e.g. whether a business can use AI to take customer orders?
    - troupo 2y ago
      You can read through EU AI Act, or read this excellent article about it: https://softwarecrisis.dev/letters/the-truth-about-the-eu-act/ https://softwarecrisis.dev/letters/the-truth-about-the-eu-ac...
xkcd-sucks 2y ago
Maybe overly cynical but "risks" of "AI" going to Congress seems mostly to be a submarine hype thing
dbspin 2y ago
"AI-induced unemployment just isn’t a problem." I'm not sure how you'd test this hypothesis, but it certainly looks like a tonne of the ongoing FAANG layoffs are based on assumptions about a decreased need for low-mid level programmers.
ben_w 2y ago
When governments make decisions in a rush, the decisions are generally not very well thought out. I therefore think it's a good idea to have a plan for AI induced unemployment before people take to the streets because they have no job, no hope of a job, and no money left to pay the bills. I have no idea how far away economic mass unemployment is, and I'm not convinced the current indicators in specific fields (much as it affects those fields) is "it", but I have seen a lot of news stories about governments getting surprised by things everyone was warning them about and they ignored.

nosefurhairdo 2y ago

Should we have protected the livelihoods of horse carriage manufacturers when cars were created?

startupsfail 2y ago

Some amount of regulation, in particular in regards to reporting biases installed into the systems is probably worth some merits. What, if, for example, we’ll have a mass information system that would start pushing something that is very beneficial for special interest communities (say community is furries and this increases their dating pool), but tends to result in higher level of depression and suicide in teenagers. And then we also find that overall the system reduces mortality, by providing high quality medical advice. Surely shutting down the system would not be a good idea. And requiring private enterprise to change their messaging infringes on free speech. But at least requiring to properly disclose biases is likely a good idea.

int_19h 2y ago

If you mean it in a sense of "what about Skynet?", then you regulate specific uses of AI which carry such danger rather than AI itself. I don't think this even needs to be AI-specific; if we, say, required formal verification of decision making algorithms to match the spec, that would already cover it. If you mean it in a sense of "what about all the lost jobs?", then we should start seriously talking about UBI, and how to fund it by taxing automation.