I don't find the "AI will cause mass unemployment" argument compelling, in fact I see the opposite as being true.
The reason we are talking so much about that now is that at the current level of productivity, we just don't need as many humans to do the work, as companies can use AI as leverage to not hire as many people.
This i feel will be a short lived period where not everyone has adopted AI to the fullest and thus simply having AI gives you leverage to not hire as many people and still stay competitive. The same thing happened with computers writ large, during the early days the folks that used computers had massive leverage over those that did not.
But what happens in 5-10 years once most orgs standardize on that leverage and everyone has it. All of a sudden that advantage AI gave you over others evaporates and you're right back to how things have always been, that is more humans give you more leverage.
I think once AI is broadly adopted and standardized, there will be an insane boom period for workers.
Even before AI, there was a serious hiring problem in the United States. Companies seem hell bent on not hiring full time employees at a livable wage. I don't blame the technology, but I do blame the greed of business owners and investors.
While its easy to look at the big picture and long term and say everything will all work out, there are so many individuals who are being hurt because of layoffs, reduced hiring, etc. My team is stressed out because they feel an axe hanging over their heads because private equity has declared no more hiring US developers. It's either AI or outsourcing. We should try to be empathetic because real people are losing their livelihoods and the long term isn't going to help them one bit.
It's a globalization problem that countries have a hard time creating policy against. Countries don't have the enforcement power that they used to. The best we have come up with is using our own markets as leverage. In a hundred years I imagine wealth and income will look much more uniform across the world. Unfortunately, right now, we are stuck competing with people who can earn much less than people in the US. Now we have to worry about this AI disaster waiting to happen because countries have no real power to stop it. They say it looks like AI won't destroy jobs, but I think really we are just watching the various legs on a table be kicked out. It will still stand until it doesn't.
This makes the assumption that humans are needed to drive these AI assisted workflows. I don't think that's going to be the case at all.
The majority of office jobs are comprised of many simple tasks that current and, to an even greater extent, future, models can handle with ease.
Not all companies are looking for 10x productivity gains. I assume most of them probably just want to do exactly what they are doing now but with lower costs.
It's only a fraction of the effect (and of course there was also over financialization), but 20% of the US population was farm labor in the 20s, that fell substantially (and there was the dust bowl) up through the 40s as tractors replaced a significant fraction of all farm labor (perhaps 1% per year). Efficiency does not always drive consumption.
Sure seems to rhyme. Call it Jevon's Tractor Paradox.
Something else certainly happened in the rest of the world about that time too!
Using AI as leverage for labor is myopic. The businesses that thrive will be the ones that use it to automate and allow its employees to go deeper and innovate. The fish can rot from the head or not.
The economics of AI investment don't make sense without the promise of vast displacement of labor. There's a ceiling on how much value can be created for a company simply by them using AI to crank out a higher throughput of TPS reports.
I don't broadly disagree and see AI opening up opportunities that didn't exist before but with respect to companies hiring large numbers of people I wonder what the people will be doing?
Come to think of it, I sort of wonder what many people with office jobs now are doing.
One sickening idea I had is even more bureaucracy. You know how any big infra project now has tens of thousand of pages of impact reports, studies and so on. With AI it can be easily scaled to millions if not more. Lots of words that then will be ingested by other AIs. All happily burning electricity in the process.
So more documentation that no one will ever read... Both generated and summarised multiple times by AI...
The same thing they are already doing? Being a companion for someone who doesn't want to be alone.
The human touch, if you will, is hard to replicate. Even where AI can convincingly act human, that often isn't enough to satisfy what people feel. Just look at the HN guidelines as a perfect example. "HN is for conversation between humans." If it were about the content then any suitable content would be welcome, regardless of the origin, but people are here under the belief that they are making a connection with other people.
Maybe someday AI will replace the business owners and managers too, but so as long as humans control the resources, they will choose to surround themselves with other humans by expending some of those resources towards those other humans.
Maybe, but the friend list is already exhausted and "cool looking" people tend to not like the idea of you hiring them just because they look cool. Hence diversity hiring: It became understood that attractive women would get the "ick" if you told them that you were hiring them only because they were attractive women, so there had to be cover for as to why they were suddenly getting hired where they previously weren't as we transitioned into a companion-based economy. Its hard to say if that can be replicated again now that we're all in on the ruse. Possible, but no guarantee. We may have already reached the peak.
The article clearly implies that one of the reasons why they "talk such nonsense" is because they are male, but no explanation is given - bog standard misandry that I take umbrage with.
I'm going to assume you left it off intentionally if you can't even address it.
Read it again. They narrow it two ways in addition to being male:
"0. but the main reason is that they are, overwhelmingly, males
1. who come from narrow Stem (science, technology, engineering, and mathematics) backgrounds
2. and are trapped in a particular mindset that the technology writer Evgeny Morozov called “solutionism”.
It's not that they are male, it males who come from Stem with a particular solution oriented mindset. I think in a different context you would be less offended. We are talking about driven, solution oriented people. It's a bias. Biases are fine, we all have them, this article is just talking about it.
I could ask again, would you agree that there are differences in the sexes? A different set of biases even?
what a patronizing and absurd reply. If it's not about being male, then why mention it at all? I'm not offended in the slightest, I just don't tolerate this type of mindless twaddle.
There are places where 20% of society (or more) are unemployed. They're not nice places. They're run by gangs or warlords. They're the kind of place where teenagers sit on steps and just wait for the wrong person to walk past them.
I cannot imagine the answer is strictly just one of those two. Young men in particular who have no prospects can and nothing to domesticate them (parents, jobs, relationship prospects) can especially be a lot of trouble.
That said, I'm not particularly convinced that gangs and warlords would suppress employment (if anything, the more organized criminals tend to run protection rackets which sort of require people to have some income they can steal) however they are most certainly reducing the quality of life for people.
Unemployment causing gangs at least feels more plausible, however you seem to have plenty of areas with gang activity that have plenty of employment opportunities as well as social programs. Selling drugs almost certainly makes more money, but comes with much, much higher risks. I might guess that this is more about social mobility than it is about survival, but I don't want to try to over-simplify a complex issue.
There was a change in the world, which went largely unnoticed - sometime in 2013-2017 the global number of infants peaked, never to recover.
Those 20% would just figure that there's no need to have children when you can just focus on yourself and go largely extinct within a generation, replaced with some people from the higher tiers, but not nearly enough to make up for those who weren't born.
This actually has been happening without AI for a while now, but I imagine technology would accelerate the process.
> Those 20% would just figure that there's no need to have children when you can just focus on yourself and go largely extinct within a generation, replaced with some people from the higher tiers, but not nearly enough to make up for those who weren't born.
A person being _unemployed_ is not a hereditary trait. Also, while it's already pretty wild to effectively claim that about 20% of humanity carries some kind _doomed to be made unemployable by AI_ gene, it's another giant leap beyond that to declaring this cohort will uniformly choose not to reproduce and therefore become extinct.
It doesn't need to be hereditary. Just needs to be in the range where on one hand a person doesn't want the next generation to go through the same and on the other, where making the most out of their current situation means not starting a family.
It's what's happening right now, just with poverty being the largest factor. Gig/minimum wage workers start families considerably less often and no wonder - how they are supposed to?
I don't find the "AI will cause mass unemployment" argument compelling, in fact I see the opposite as being true.
The reason we are talking so much about that now is that at the current level of productivity, we just don't need as many humans to do the work, as companies can use AI as leverage to not hire as many people.
This i feel will be a short lived period where not everyone has adopted AI to the fullest and thus simply having AI gives you leverage to not hire as many people and still stay competitive. The same thing happened with computers writ large, during the early days the folks that used computers had massive leverage over those that did not.
But what happens in 5-10 years once most orgs standardize on that leverage and everyone has it. All of a sudden that advantage AI gave you over others evaporates and you're right back to how things have always been, that is more humans give you more leverage.
I think once AI is broadly adopted and standardized, there will be an insane boom period for workers.
Even before AI, there was a serious hiring problem in the United States. Companies seem hell bent on not hiring full time employees at a livable wage. I don't blame the technology, but I do blame the greed of business owners and investors.
While its easy to look at the big picture and long term and say everything will all work out, there are so many individuals who are being hurt because of layoffs, reduced hiring, etc. My team is stressed out because they feel an axe hanging over their heads because private equity has declared no more hiring US developers. It's either AI or outsourcing. We should try to be empathetic because real people are losing their livelihoods and the long term isn't going to help them one bit.
It's a globalization problem that countries have a hard time creating policy against. Countries don't have the enforcement power that they used to. The best we have come up with is using our own markets as leverage. In a hundred years I imagine wealth and income will look much more uniform across the world. Unfortunately, right now, we are stuck competing with people who can earn much less than people in the US. Now we have to worry about this AI disaster waiting to happen because countries have no real power to stop it. They say it looks like AI won't destroy jobs, but I think really we are just watching the various legs on a table be kicked out. It will still stand until it doesn't.
In the long term the sun will go red giant and all life on earth will die.
In the short term I still need to eat, and tech integration that takes 20 years in the economy might leave me totally fucked!
This makes the assumption that humans are needed to drive these AI assisted workflows. I don't think that's going to be the case at all.
The majority of office jobs are comprised of many simple tasks that current and, to an even greater extent, future, models can handle with ease.
Not all companies are looking for 10x productivity gains. I assume most of them probably just want to do exactly what they are doing now but with lower costs.
5 to 10 years of 20% unemployment could be verydangerous in places like the us
You think? Here's the U.S. unemployment rate during the great depression
It's only a fraction of the effect (and of course there was also over financialization), but 20% of the US population was farm labor in the 20s, that fell substantially (and there was the dust bowl) up through the 40s as tractors replaced a significant fraction of all farm labor (perhaps 1% per year). Efficiency does not always drive consumption.
Sure seems to rhyme. Call it Jevon's Tractor Paradox.
Something else certainly happened in the rest of the world about that time too!
Who's going to buy the results of that extra output while most people are laid off? Where is the growth going to actually come from?
Even before LLMs, the cost of living and finding a job that left any disposal income were major societal issues.
Using AI as leverage for labor is myopic. The businesses that thrive will be the ones that use it to automate and allow its employees to go deeper and innovate. The fish can rot from the head or not.
The economics of AI investment don't make sense without the promise of vast displacement of labor. There's a ceiling on how much value can be created for a company simply by them using AI to crank out a higher throughput of TPS reports.
I've been saying "The economics of AI investment don't make sense" for 2 years but this audience in particular doesn't (on the whole) agree.
Yes, the value of TPS reports diminishes with throughput.
That's why Jevon's Paradox is the dumbest meme we've had this year. The parent concept of marginal utility describes why AI isn't that useful.
I don't broadly disagree and see AI opening up opportunities that didn't exist before but with respect to companies hiring large numbers of people I wonder what the people will be doing?
Come to think of it, I sort of wonder what many people with office jobs now are doing.
trying to reify the proposed efficiency gains they have been told are just out of reach
One sickening idea I had is even more bureaucracy. You know how any big infra project now has tens of thousand of pages of impact reports, studies and so on. With AI it can be easily scaled to millions if not more. Lots of words that then will be ingested by other AIs. All happily burning electricity in the process.
So more documentation that no one will ever read... Both generated and summarised multiple times by AI...
The same thing they are already doing? Being a companion for someone who doesn't want to be alone.
The human touch, if you will, is hard to replicate. Even where AI can convincingly act human, that often isn't enough to satisfy what people feel. Just look at the HN guidelines as a perfect example. "HN is for conversation between humans." If it were about the content then any suitable content would be welcome, regardless of the origin, but people are here under the belief that they are making a connection with other people.
Maybe someday AI will replace the business owners and managers too, but so as long as humans control the resources, they will choose to surround themselves with other humans by expending some of those resources towards those other humans.
So the tendency to hire friends and "cool looking" people instead of smart talented people will get even worse?
Maybe, but the friend list is already exhausted and "cool looking" people tend to not like the idea of you hiring them just because they look cool. Hence diversity hiring: It became understood that attractive women would get the "ick" if you told them that you were hiring them only because they were attractive women, so there had to be cover for as to why they were suddenly getting hired where they previously weren't as we transitioned into a companion-based economy. Its hard to say if that can be replicated again now that we're all in on the ruse. Possible, but no guarantee. We may have already reached the peak.
> males who come from narrow Stem (science, technology, engineering, and mathematics) backgrounds
I see, so women from narrow Stem backgrounds would totally handle these issues differently, got it.
I think you left this off.
> the main reason is that they are, overwhelmingly, males
Was that intentional? Do you disagree that tech is overwhelmingly male?
Do you happen to be someone who believes in differences in the sexes? Just curious.
The article clearly implies that one of the reasons why they "talk such nonsense" is because they are male, but no explanation is given - bog standard misandry that I take umbrage with.
I'm going to assume you left it off intentionally if you can't even address it.
Read it again. They narrow it two ways in addition to being male:
"0. but the main reason is that they are, overwhelmingly, males 1. who come from narrow Stem (science, technology, engineering, and mathematics) backgrounds 2. and are trapped in a particular mindset that the technology writer Evgeny Morozov called “solutionism”.
It's not that they are male, it males who come from Stem with a particular solution oriented mindset. I think in a different context you would be less offended. We are talking about driven, solution oriented people. It's a bias. Biases are fine, we all have them, this article is just talking about it.
I could ask again, would you agree that there are differences in the sexes? A different set of biases even?
what a patronizing and absurd reply. If it's not about being male, then why mention it at all? I'm not offended in the slightest, I just don't tolerate this type of mindless twaddle.
> I'm not offended in the slightest
> what a patronizing and absurd reply
I had to double check my replies! The author, nor me implied it wasn't about being male, in fact it is explicitly called out. Check out this question:
"would you agree that there are differences in the sexes? A different set of biases even?"
What mindless twaddle aren't you tolerating?
Unironically, yes. It would be totally different.
I don't find the argument that advanced AI will leave 20% of society permanently unemployed particularly compelling
There are places where 20% of society (or more) are unemployed. They're not nice places. They're run by gangs or warlords. They're the kind of place where teenagers sit on steps and just wait for the wrong person to walk past them.
Do those places have gangs and warlords because they have 20%+ unemployment, or do they have 20%+ unemployment because they have gangs and warlords?
I cannot imagine the answer is strictly just one of those two. Young men in particular who have no prospects can and nothing to domesticate them (parents, jobs, relationship prospects) can especially be a lot of trouble.
That said, I'm not particularly convinced that gangs and warlords would suppress employment (if anything, the more organized criminals tend to run protection rackets which sort of require people to have some income they can steal) however they are most certainly reducing the quality of life for people.
Unemployment causing gangs at least feels more plausible, however you seem to have plenty of areas with gang activity that have plenty of employment opportunities as well as social programs. Selling drugs almost certainly makes more money, but comes with much, much higher risks. I might guess that this is more about social mobility than it is about survival, but I don't want to try to over-simplify a complex issue.
Like where?
Lots of places in the ME and Africa:
https://en.wikipedia.org/wiki/List_of_countries_by_unemploym...
In the US, ~80% of the prime working age population is employed:
https://fred.stlouisfed.org/series/LNS12300060
For every one person who is classified as unemployed, there are one or more others who are classified unable to work for some reason.
Fortunately it's extremely unlikely.
There was a change in the world, which went largely unnoticed - sometime in 2013-2017 the global number of infants peaked, never to recover.
Those 20% would just figure that there's no need to have children when you can just focus on yourself and go largely extinct within a generation, replaced with some people from the higher tiers, but not nearly enough to make up for those who weren't born.
This actually has been happening without AI for a while now, but I imagine technology would accelerate the process.
> Those 20% would just figure that there's no need to have children when you can just focus on yourself and go largely extinct within a generation, replaced with some people from the higher tiers, but not nearly enough to make up for those who weren't born.
A person being _unemployed_ is not a hereditary trait. Also, while it's already pretty wild to effectively claim that about 20% of humanity carries some kind _doomed to be made unemployable by AI_ gene, it's another giant leap beyond that to declaring this cohort will uniformly choose not to reproduce and therefore become extinct.
It doesn't need to be hereditary. Just needs to be in the range where on one hand a person doesn't want the next generation to go through the same and on the other, where making the most out of their current situation means not starting a family.
It's what's happening right now, just with poverty being the largest factor. Gig/minimum wage workers start families considerably less often and no wonder - how they are supposed to?
Wait until you're in that 20%.
i dont know anyone who wants AI
only use case i see, is ... helping disabled people
I am pretty down on the pervasive LLM psychosis - but that doesn't mean they are useless.
Over a billion unique people are using chatgpt,Gemini,Claude every month. You should try asking one of them what they're doing
Maybe you - as in, whoever says that - should go first talk to them, because I don't know any disabled people wanting AI (data set of one).
https://archive.ph/fRwM4
Why didn't you regard it as a promotion or marketing strategy? Actually it is
Full title: "Silicon Valley sees AI as the solution – for everyone else, it’s the problem"
I don't understand why OP cut that off... that's a completely different title.
Character limit..?
Even when the full title is displayed it can be interpreted in two ways.
Bait
[dead]
Rather ironically you spend time outside Silicon Valley it would appear those most at risk of “mass unemployment” are the folks full-press pushing AI.
It’s “cool” and the tech is here to stay, but we appear light years away from anything like what was promised from materializing.
The current hype cycle is clearly barreling toward the “trough of disillusionment” phase of the Gartner hype cycle.
[flagged]