Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Most AI value will come from broad automation, not from R & D

epoch.ai

Post reply on HN
Print view View original

191–200 of 319 posts

Re: Most AI value will come from broad automation, not from R & D

#191

Post by AlecSchueler »

Swizec wrote: post #12
henry2023 wrote: post #5

Earlier quoted context omitted.

Only when looking at this through our common understanding of how societies work. Broad automation of tasks can be great for society if and only if the product of that automation is not treated as some kind of private property. I’m not hyped on an automated future because I find it quite unlikely but if it were to happen it has the potential to be transforming beyond expectations.

> Broad automation of tasks can be great for society if and only if the product of that automation is not treated as some kind of private property. Star Trek or Bladerunner? That’s a choice we can make.

Do we have the choice? It feels like the market makes such decisions and regulation is not only a taboo but impossible with competition between states.

Re: Most AI value will come from broad automation, not from R & D

#192

Post by AlecSchueler »

SequoiaHope wrote: post #114
lukevp wrote: post #94

Earlier quoted context omitted.

Huh? 40 hr work weeks are a recent thing, many people in history worked 12 hr days 6 days a week. Technology and social progress definitely leads to fewer work hours.

I’m not so sure. Unions (social progress) brought us the 40 hour work week more so than technology. Factory workers could potentially work more hours than farmers who had very busy periods mixed with slower times. I suppose they exist, maybe you have seen them, but I’ve not seen a real study of the hours worked by subsistence farmers in history. Though technology brings many other benefits, it’s not clear to me how i…

It was factories that brought people together enough working as a single unit that allowed unions to flourish.

Re: Most AI value will come from broad automation, not from R & D

#193

Post by stego-tech »

bluefirebrand wrote: post #140
stego-tech wrote: post #92

Earlier quoted context omitted.

> So this is my thesis now - if you’re a senior who neglects this new tool, you’re going to be out of a job soon-ish. Work(ed) for a company super big on a force of agents. Brought them into my workflows early on, before the company even pivoted. Ran them at home on private hardware (that 3090 had to do something ), at work when access was opened up. Provided valuable feedback on hallucinations (e.g., fabricating the…

> Work(ed) for a company super big on a force of agents. > RIFed this year Yeah that's why they are big on the force of agents, so they can justify trimming their workforce Anyone who doesn't see this is kind of a sucker

Oh, 100%, and I knew that going in. Still, I'm the type of worker (sucker? sap?) that will go down with the ship if it helps the rest of the folks get into lifeboats. I know I can take care of myself when stuff hits the fan, and I'm never one to half-ass my output because of my own cynicism.

I'm hoping that pays off for me someday. Thus far, it's just been a lot of burnout/layoff cycles.

Re: Most AI value will come from broad automation, not from R & D

#194

Post by jgilias »

bluefirebrand wrote: post #140
stego-tech wrote: post #92

Earlier quoted context omitted.

> So this is my thesis now - if you’re a senior who neglects this new tool, you’re going to be out of a job soon-ish. Work(ed) for a company super big on a force of agents. Brought them into my workflows early on, before the company even pivoted. Ran them at home on private hardware (that 3090 had to do something ), at work when access was opened up. Provided valuable feedback on hallucinations (e.g., fabricating the…

> Work(ed) for a company super big on a force of agents. > RIFed this year Yeah that's why they are big on the force of agents, so they can justify trimming their workforce Anyone who doesn't see this is kind of a sucker

Sure, but what choice do you have as an individual? Try and adjust to the new reality and maybe get RIFed anyway, or not adjust to it and get RIFed with 90% certainty. Power looms won.

Re: Most AI value will come from broad automation, not from R & D

#195

Post by stego-tech »

kkaatii wrote: post #133
stego-tech wrote: post #93

It's the Industrial and Agricultural Revolutions all over again. Broad automation of labor will not result in societal uplift, but Capital uplift, absent worker action to enforce equal distribution of gains (like how Unions brought us the 40hr work week and the end of Child Labor w/ the Industrial Revolution). Worse yet, the irrationality of these techbro arguments shows their complete lack of long-term thinking or s…

One danger is that, from a macroeconomics view, GDP growth can be attained without any regard for natural unemployment rates. When technology, controlled by capital, drives up markup on wages by displacing human labor and killing competition, then natural unemployment will rise, but the economy can still "thrive," from a statistical perspective. But it is individual persons being ignored by statistics. Indeed a grim…

> GDP

Your entire argument is why I reject GDP as a reliable measure of economic health (that is to say, I agree with you 100%). It can be a component of it, sure, but as the standalone metric it's relied upon as-is, it's awful. Doesn't capture inflation, doesn't capture real productivity growth, doesn't capture employment rates or labor compensation or income distribution. Heck, GDP is literally an ideal expression of Goodhart's Law: we have built entire systems designed to game a single number that was proposed as a metric of economic health, which no longer makes it a good metric of economic health!

If someone's sole defense is "GDP Up = Good", I do not take their argument seriously - because it isn't.

Re: Most AI value will come from broad automation, not from R & D

#196

Post by treis »

somenameforme wrote: post #131
rqtwteye wrote: post #25

Earlier quoted context omitted.

Technical progress would be fantastic if the benefits didn't go mainly to the top. I am pretty sure in the not too far future we will have enough technology that you can live off the land somewhere with the help of some robots. Or have them do all your housework. In principle, this is great, but our society is not set up for a lot of people not working while the technology is progressed by a relatively small number o…

There's a simple litmus test for these sort of nearish future dystopias - how long before you have an 'AI plumber' or an 'AI electrician'? I'm not just listing random skilled trades, but listing jobs that are extremely complex, 100% context sensitive in a way that just doesn't generalize well, requires a mixture of high dexterity and high strength, and a million other things. To say nothing of the numerous [oft extre…

I don't think we're that far away but I also think it looks different than humanoid robots.

I can see a drain plumbing system that maps out a new construction home and some sort of automated robot that builds out large sections of pipe. It wouldn't completely eliminate the plumber. But it'd change it from something like measuring, cutting, and gluing 100 times to walking around with a camera for 5 minutes and then gluing 10 things together.

Re: Most AI value will come from broad automation, not from R & D

#197

Post by renewiltord »

techpineapple wrote: post #2

Does anyone else find techno-optimism really depressing? I guess one for the more common reason that it’s displacing humanity with technology, but the second reason is just you can’t get excited about a hype that is unlikely to manifest. It feels very ungrounded from tangible benefits to society.

I think one of the sad things is that when software went mainstream the culture went mainstream too so now that everyone wants to be “oh woe is me the world sucks” software people have to do the same too instead of being optimists.

The degrowther culture of the mainstream comes to push out all nerds eventually.

Re: Most AI value will come from broad automation, not from R & D

#198

Post by aksbfjdjsbabs »

oblio wrote: post #80
dogcomplex wrote: post #73

Earlier quoted context omitted.

We've had a long history of technological improvements being widespread distributed to the people. There's not a particularly bleak reason to believe the latest AI automation won't be too. Look around your desk or your house and just count all the effort-saving devices that have made their way down to you. Look at the price of TVs cratering. Tech that can be recreated easily spreads far and wide. AI can too. It's dro…

Progress is not guaranteed. Humanity has been around in basically the same form for 2 million years (and the same form for probably 200 000 years) yet life for the average person on the planet really started improving circa 1950.

2 million years is a long time. It’s quite a stretch to say life only started improving in the last 50. An asteroid could hit today and nearly all evidence of our existence would be gone in 10k years.

Re: Most AI value will come from broad automation, not from R & D

#199

Post by tokai »

bashfulpup wrote: post #69

Very typical SV argument that R&D is "complex" and everything else is "simple". Would it blow your mind if I told you 10yrs ago that we'd have AI that can do math/code better than 99% of humans but ordering a hotdog on doordash would be cutting edge and barely doable? I don't disagree that "common" tasks are more valuable. I only argue that the argument these are easily automatable is a viewpoint based on ignorance.…

>10yrs ago [...] ordering a hotdog on doordash would be cutting edge and barely doable?

Online food ordering is a lot older than 15 years.

Re: Most AI value will come from broad automation, not from R & D

#200

Post by throw80521 »

luckydata wrote: post #132
somenameforme wrote: post #131

Earlier quoted context omitted.

There's a simple litmus test for these sort of nearish future dystopias - how long before you have an 'AI plumber' or an 'AI electrician'? I'm not just listing random skilled trades, but listing jobs that are extremely complex, 100% context sensitive in a way that just doesn't generalize well, requires a mixture of high dexterity and high strength, and a million other things. To say nothing of the numerous [oft extre…

You need more to maintain a house, drywaller, gardener, carpenter, painter. The most important robotic help though will be cleaner and nurse. As our population ages and shrinks home help will become scarce and we'll desperately need an alternative to take care of ourselves as we age.

Perhaps in response housing will become even more homogenized so that AI has an easier time maintaining it in a standardized way?
Post reply on HN