One of the biggest unspoken dynamics here is the shift from Software-as-a-Service to Employee-as-a-Service. When the software people use to do their jobs becomes capable of doing more and more of the actual job itself—without needing sleep, health insurance, or breaks—the economic incentive to replace people with systems becomes overwhelming.
And yet, the middle class has functioned as a kind of semi-meritocratic pseudo UBI for decades. It exists not because everyone in it is essential to production, but because their consumption and social stability are essential to the economy’s overall function.
But bots don’t pay taxes. So if automation eats into these roles and the tax base shrinks, collapse becomes a slow inevitability.
The mistake is assuming AI has to replace every job to be disruptive. It only has to displace just enough people in just enough places to tip the system. And once it can replace your best workers—not just the average—it’s over.
The correct reaction would have looked like an overreaction.
Nice overview but I’m not sure it matters when we, in the US at least, routinely vote against policy solutions to more immediate crises that impact far more people today, like single payer healthcare and progressive taxes to lower inequality. The leverage dynamic is real and only getting worse with capital valued more than labor even before labor is automated.
Yes, universal basic income, at at least $1500 per month & $500 per child. Can't swing that in the budget? Then do a UBI for kids at the very least (the parents get the money).
Then work out Universal Affordable Healthcare (I like the idea of Universal Catastrophic Care as a model with more generous tax-free HSAs on the side with employer matching options).
Sure, Universal Basic Compute should be compatible for transfer in a basic market with $$$, also bring on Data Dividends so that we can be paid for letting corporations sniff our digital bread crumbs & collect our data.
Bring back mental institutions & let's get homeless people off the street. Build AI-oriented rehabilitation centers to help people recover from debilitating mental illnesses.
Additionally, think about Time Bank Credits: the use of environmentally-friendly cryptocoins to incentivize local markets and labor ecosystems for productive, tax-free work in your community.
Get humans back on the Moon & start colonizing it. Motivate humanity around outer space expansion. Industrialize Luna & the asteroid belt, start shipping raw materials back to factories in Earth orbit. Build solar power satellite swarms.
Let's first & foremost eliminate global homelessness, poverty, modern slavery, & climate change before we move on to other projects. Also microplastics. Can't forget microplastics. Let's pick things back up in 2042 to solve the rest, can't we? Let's pick a path that looks promising & start down it, then if we come to a dead end, we'll find another path until we work our way through.
To me, the gut feeling here is that new legislation will be passed to restrict companies from using AI systems in their workforce. There will be growing pressure for this to happen - I don't think this approach is optimal (given how reductively I've described it), but applying Occam's razor, this seems like the most likely scenario. (I am additionally not sure how you police this exactly)
Ultimately, if you're building something more intelligent than humans, you've somewhat lost the battle by default, and I think lots of people trying to fight back against this kind of miss this point pretty often.
Setting my pessimism aside though, another thought I had was maybe labor disruption could be the catalyst that breaks the dam - creating widespread mainstream concern that finally pushes for serious regulation of AI labs, which in mind would be a positive.
One of the biggest unspoken dynamics here is the shift from Software-as-a-Service to Employee-as-a-Service. When the software people use to do their jobs becomes capable of doing more and more of the actual job itself—without needing sleep, health insurance, or breaks—the economic incentive to replace people with systems becomes overwhelming.
And yet, the middle class has functioned as a kind of semi-meritocratic pseudo UBI for decades. It exists not because everyone in it is essential to production, but because their consumption and social stability are essential to the economy’s overall function.
But bots don’t pay taxes. So if automation eats into these roles and the tax base shrinks, collapse becomes a slow inevitability.
The mistake is assuming AI has to replace every job to be disruptive. It only has to displace just enough people in just enough places to tip the system. And once it can replace your best workers—not just the average—it’s over.
The correct reaction would have looked like an overreaction.
Nice overview but I’m not sure it matters when we, in the US at least, routinely vote against policy solutions to more immediate crises that impact far more people today, like single payer healthcare and progressive taxes to lower inequality. The leverage dynamic is real and only getting worse with capital valued more than labor even before labor is automated.
Yes, universal basic income, at at least $1500 per month & $500 per child. Can't swing that in the budget? Then do a UBI for kids at the very least (the parents get the money).
Then work out Universal Affordable Healthcare (I like the idea of Universal Catastrophic Care as a model with more generous tax-free HSAs on the side with employer matching options).
Sure, Universal Basic Compute should be compatible for transfer in a basic market with $$$, also bring on Data Dividends so that we can be paid for letting corporations sniff our digital bread crumbs & collect our data.
Bring back mental institutions & let's get homeless people off the street. Build AI-oriented rehabilitation centers to help people recover from debilitating mental illnesses.
Additionally, think about Time Bank Credits: the use of environmentally-friendly cryptocoins to incentivize local markets and labor ecosystems for productive, tax-free work in your community.
Get humans back on the Moon & start colonizing it. Motivate humanity around outer space expansion. Industrialize Luna & the asteroid belt, start shipping raw materials back to factories in Earth orbit. Build solar power satellite swarms.
Let's first & foremost eliminate global homelessness, poverty, modern slavery, & climate change before we move on to other projects. Also microplastics. Can't forget microplastics. Let's pick things back up in 2042 to solve the rest, can't we? Let's pick a path that looks promising & start down it, then if we come to a dead end, we'll find another path until we work our way through.
To me, the gut feeling here is that new legislation will be passed to restrict companies from using AI systems in their workforce. There will be growing pressure for this to happen - I don't think this approach is optimal (given how reductively I've described it), but applying Occam's razor, this seems like the most likely scenario. (I am additionally not sure how you police this exactly)
Ultimately, if you're building something more intelligent than humans, you've somewhat lost the battle by default, and I think lots of people trying to fight back against this kind of miss this point pretty often.
Setting my pessimism aside though, another thought I had was maybe labor disruption could be the catalyst that breaks the dam - creating widespread mainstream concern that finally pushes for serious regulation of AI labs, which in mind would be a positive.
Of course, unless the USG just nationalises the labs, in which case, seems like we would never see any regulation on it.