4 ms·
> because it makes the rich richer and the poor poorer This is a terrible take. Whenever there is a massive technical shift it's the incumbents who struggle to
by guywithahat 1y ago
> because it makes the rich richer and the poor poorer
This is a terrible take. Whenever there is a massive technical shift it's the incumbents who struggle to adapt. We've already seen companies go from nothing to being worth tens of billions.
- esafak 1y agoThe OP is likely referring to the technology's ability to replace certain workers, which naturally leaves them poorer. The net effect is one of increased wealth inequality, the minting of new AI entrepreneurs notwithstanding. Is the data on this out yet?
- sgjohnson 1y agoWe survived industrial revolution, we’ll survive AI.
- throwaway09387 1y ago> We survived industrial revolution Did we? I mean, I guess we're surviving it for now, but climate change data doesn't tell the most optimistic story.
- cloverich 1y agoTBF the worst outcome is famine and mass death of humans which... was somewhat the standard historically (low populations / intermittent famine). I have as much existential dread as you but that's also because I overvalue human life relative to what nature has historically thought about us.
- croes 1y agoSurvivorship bias. You always survive unless you don’t.
- sgjohnson 1y agoDoesn’t matter. Our species survived. Individual casualties don’t matter in the grand scheme of things.
- croes 1y agoIt does matter because it‘s a false trust in our ability to survive. We could make an error that kills the whole species based on the assumption because we survived before
- ben_w 1y agoThe industrial revolution was so good for those living through it, that they invented communism and recognisably modern policing. And we got a few wars whose death toll exceeded the pre-industrial population of the UK. And one whose toll exceeds the current population of the UK.
- vincentpants 1y agoIsn't the jury still out on that statement? Pretty sure we've been experiencing the long-tail cultural decomposition caused by the effects the industrial revolution brought with it with the exception of a 20 year opportunity to change our trajectory as a species. I think getting out of this conundrum alive will require de-segmenting our cultural narrative around how we got here. edit: typos/grammar abound.
- guywithahat 1y ago> Isn't the jury still out on that statement? No
- archagon 1y agoContent-free comment.
- willis936 1y agoThis isn't an industrial revolution moment; it's a printing press moment.
- hayst4ck 1y ago"We" is doing an awful lot of work. We may have survived the industrial revolution, but subjective individual experience varies widely. Life would be quite different if we could subjectively experience ourselves as a whole.
- delusional 1y agoYet it's somehow the same band of people at the helm of these "new" companies.
- hn_throwaway_99 1y agoEven if there are some new entrants who become remarkably successful, it still means that for the vast majority of humans that aren't AI experts that the "spoils" will still accrue primarily to these "AI winners" while making it much, much, much harder for many, many, many more people to sell their labor. And many of the investors in these companies that are poised to make bank are the usual suspects. And even then, many of the biggest winners of AI so far (Google, Microsoft, NVidia, etc.) are already some of the biggest companies on the planet.
- bilbo0s 1y agoEven the AI experts are the usual suspects. The techies making money are the AI experts. The real AI experts. Not the TF/PyTorch monkeys. There's actually a massive and underappreciated difference between the two groups of people. It goes to the heart of why so many AI efforts in industry fail, but AI Labs keep making more and more advances. Out in industry, we mistake Tensor Flow monkeys for AI experts. And it's not even close. Worse, you look at market price to get some of the real AI experts, and you realize that you have no shot at securing any of that intellectual capital. And even if you have the resources to secure some of that talent, that talent has requirements. They're fools if they consider you at less than 10^4 H100s. So now you have another problem. I think techniques, R&D, secrets, intellectual capital, and so on are all centralizing in the major labs. Startups as we knew them 5 to 10 years ago will simply be choosing which model to build on. They have no legitimate shot at displacing any of the core LLM ecosystems. They'll all be nibbling at the edges.
- bilbo0s 1y agoWell, companies, by and large, started and funded by.. incumbents. That said, I think that's just how reality works.
- croes 1y agoYou confuse companies with people. Companies come and go but the rich people who owns them nearly stay the same
- guywithahat 1y agoYou are your own sole proprietorship, go learn an in-need skill and companies will compete for you. It can be hard to see in big companies but these market forces are very apparent in a 50-person company
- croes 1y agoHas nothing to do with my point that companies change but the people in power stay the same
- hayst4ck 1y agoYou can pay more for more access to a better model. Answer quality directly correlates to resources spent generating it. It literally and structurally offers advantage to those with more resources. That effect compounds over time and use. I am not even remotely talking about worker replacement, even assuming no job was lost, a company that is able to pay for better answers should have more profit and therefore more ability to pay for more/better answers.
- reaperducer 1y agoWe've already seen companies go from nothing to being worth tens of billions. And now the artists have to wash dishes because AI is making the art. Nope, he was right. Rich get richer, poor get poorer. A few unicorns you can count on one hand doesn't change the facts.