6 ms·

I wonder if the yellow vest protests in Paris and other European cities is our first taste of the kind of revolt that will happen when this really starts happen

I wonder if the yellow vest protests in Paris and other European cities is our first taste of the kind of revolt that will happen when this really starts happening

Flavius 8y ago
Maybe, but we've been losing jobs to robots for over 300 years now.
- _i____ii_______ 8y ago
  I think we are in a different place than the Luddites were.
- maxxxxx 8y ago
  There have been quite a few revolts during that time.
- jedberg 8y ago
  Up until now, they have mostly been replacing physical labor, which allowed the workforce to shift to service and mental labor. But now the machines are replacing service and mental labor. There is no where for a human to shift to this time.
  - hutzlibu 8y ago
    "But now the machines are replacing service and mental labor. There is no where for a human to shift to this time." As far as I know, computers on their own are not very good engineers, programmers, physicians, etc. I also believe creativity in general is a kind of unsolved problem. The art/music computers made so far ... not worth mentioning. (except as awesome research projects)
    - harshreality 8y ago
      The number of jobs requiring that kind of creativity is very small. Of course there will be a need for engineers, and physicians, to interface with other humans and to integrate answers from narrow, specialized AIs into real-world solutions, but that's not the bulk of everyday work even in high-skill jobs. If AI cuts the amount of work someone has to do in half, the outcomes are on a continuum between two extremes: either effective hourly wage for work doubles and people work half the time, or the job market shrinks by 50%. But what would actually happen is closer to 33% less work and the job market shrinks by 25% too. "People are still smarter and more adaptable than narrow AI" is completely irrelevant. What matters is the workload reduction offered by the narrow AI. The OP didn't say all jobs by 2030, it said 800M. Obviously there will still be lots of jobs, but not enough. Too many people unemployed and broke, combined with not enough of a social safety net for food, housing, and medical care, will result in social unrest and, if that's not dealt with well enough, it will progress to social breakdown, which either means a revolution (to institute a new government that does implement appropriate safety nets) or a martial-law type of scenario, neither of which is pretty. Below-average doctors, engineers, and lawyers will be at risk of being out of work. On-site blue collar jobs are safer, as a percentage. Robots can prefab things in factories, but it will be a long time before automated robots can build or repair anything on-site with the efficiency of humans + power tools. ETA: I think you overestimate the number of people required to handle most of the problems you think need work. And the current free market doesn't agree that they need more work, or it would price those jobs higher and more people would jump into those labor pools. In some cases, the jobs require serious specialization (maybe not advanced degrees, but a fair amount of training) and not just anyone can do them... in which case even the optimal labor pool size would barely scratch the surface of the coming AI-caused unemployment crisis. Even a massive public works effort to employ tens of millions of workers to deal with aging infrastructure, or to build green tech, is not a permanent solution. Most of those jobs would be gone in a decade, and then you're right back to where you started, only worse because AI will have gotten better. Massive government spending (including on temp public works projects) is essentially UBI for a lot of able-bodied people. I don't object to that, but it needs to be limited to public works projects that are really necessary, and then you have to figure out where you're going to get the money to pay for them when the govt is already running trillion-dollar deficits. Is 5% debt inflation per year not good enough? Should we try 10%? 15%?
  - hoaw 8y ago
    I don't see it as much of a problem. Unless only a minority gets to enjoy the increase in productivity, which is what we are mostly seeing now.
  - 8bitsrule 8y ago
    Robot killer for hire. Will work for food.
4thaccount 8y ago
I'm not sure if we should hold onto humans doing things machine's can do better. Instead, perhaps UBI makes better sense. I know the economics part won't make sense until we're past the need for the current system, but if we don't need people to work, why have them waste their lives with suboptimal labor. My biggest issue with UBI is that you'll probably just get millions watching Netflix all day. What would be awesome is if those millions were writing books, doing physics, or creating works of art instead. You'd have machines doing the labor and humans being freed up to focus on new contributions to society. Of course, not to be elitist, but I'm not sure how many truck drivers will pick up work on the grand unified theory of physics, but if even one makes a new and important discovery that would be amazing. Also, I wonder how long until even jobs held by the highly educated (healthcare...etc) are largely automated?
- whatshisface 8y ago
  >What would be awesome is if those millions were writing books, doing physics, or creating works of art instead. You'd have machines doing the labor and humans being freed up to focus on new contributions to society. There is a really easy solution to this that I have never seen suggested. Instead of giving money to everyone for free, why not raise the budgets of research and art institutions? You can get as many people as you want doing physics and making art as you want, all you have to do is give money to CERN. Now, a pessimist might say that some people can't advance science or art because they lack the capability. I counter with the fact that anyone can do it at some speed, some are just faster than others. If you greatly increased the amount of money the Institute for Advanced Study had, they would be able to bring people on to their staff that were less efficient on a dollars per fundamental insight scale, I.E. you or me.
  - hoaw 8y ago
    The problem is that it isn't an income issue. It is a cost issue, which in turn is a power issue. As in who decides what and therefor who society is for. Taking underprivileged people, or even anyone else, and employing them at CERN isn't likely to work (Switzerland isn't exactly cheap or big on taxes). In the Nordic countries there are "folk high schools", which gives you education (paid via taxes), housing and food for ~$650 usd a month. They are mostly in the middle of nowhere though. Is it a good idea to educate people who gets squeezed out of society? Sure. But most societies are going the opposite way even now. Like many other things it isn't a technical issue, but a social one. There isn't really a need to wait for society to collapse before giving people affordable access to housing, education, health care etc.

I wonder if the yellow vest protests in Paris and other European cities is our first taste of the kind of revolt that will happen when this really starts happening

Flavius 8y ago

Maybe, but we've been losing jobs to robots for over 300 years now.

_i____ii_______ 8y ago
I think we are in a different place than the Luddites were.
maxxxxx 8y ago
There have been quite a few revolts during that time.
jedberg 8y ago
Up until now, they have mostly been replacing physical labor, which allowed the workforce to shift to service and mental labor. But now the machines are replacing service and mental labor. There is no where for a human to shift to this time.
- hutzlibu 8y ago
  "But now the machines are replacing service and mental labor. There is no where for a human to shift to this time." As far as I know, computers on their own are not very good engineers, programmers, physicians, etc. I also believe creativity in general is a kind of unsolved problem. The art/music computers made so far ... not worth mentioning. (except as awesome research projects)
  - harshreality 8y ago
    The number of jobs requiring that kind of creativity is very small. Of course there will be a need for engineers, and physicians, to interface with other humans and to integrate answers from narrow, specialized AIs into real-world solutions, but that's not the bulk of everyday work even in high-skill jobs. If AI cuts the amount of work someone has to do in half, the outcomes are on a continuum between two extremes: either effective hourly wage for work doubles and people work half the time, or the job market shrinks by 50%. But what would actually happen is closer to 33% less work and the job market shrinks by 25% too. "People are still smarter and more adaptable than narrow AI" is completely irrelevant. What matters is the workload reduction offered by the narrow AI. The OP didn't say all jobs by 2030, it said 800M. Obviously there will still be lots of jobs, but not enough. Too many people unemployed and broke, combined with not enough of a social safety net for food, housing, and medical care, will result in social unrest and, if that's not dealt with well enough, it will progress to social breakdown, which either means a revolution (to institute a new government that does implement appropriate safety nets) or a martial-law type of scenario, neither of which is pretty. Below-average doctors, engineers, and lawyers will be at risk of being out of work. On-site blue collar jobs are safer, as a percentage. Robots can prefab things in factories, but it will be a long time before automated robots can build or repair anything on-site with the efficiency of humans + power tools. ETA: I think you overestimate the number of people required to handle most of the problems you think need work. And the current free market doesn't agree that they need more work, or it would price those jobs higher and more people would jump into those labor pools. In some cases, the jobs require serious specialization (maybe not advanced degrees, but a fair amount of training) and not just anyone can do them... in which case even the optimal labor pool size would barely scratch the surface of the coming AI-caused unemployment crisis. Even a massive public works effort to employ tens of millions of workers to deal with aging infrastructure, or to build green tech, is not a permanent solution. Most of those jobs would be gone in a decade, and then you're right back to where you started, only worse because AI will have gotten better. Massive government spending (including on temp public works projects) is essentially UBI for a lot of able-bodied people. I don't object to that, but it needs to be limited to public works projects that are really necessary, and then you have to figure out where you're going to get the money to pay for them when the govt is already running trillion-dollar deficits. Is 5% debt inflation per year not good enough? Should we try 10%? 15%?
- hoaw 8y ago
  I don't see it as much of a problem. Unless only a minority gets to enjoy the increase in productivity, which is what we are mostly seeing now.
- 8bitsrule 8y ago
  Robot killer for hire. Will work for food.

4thaccount 8y ago

I'm not sure if we should hold onto humans doing things machine's can do better. Instead, perhaps UBI makes better sense. I know the economics part won't make sense until we're past the need for the current system, but if we don't need people to work, why have them waste their lives with suboptimal labor. My biggest issue with UBI is that you'll probably just get millions watching Netflix all day. What would be awesome is if those millions were writing books, doing physics, or creating works of art instead. You'd have machines doing the labor and humans being freed up to focus on new contributions to society. Of course, not to be elitist, but I'm not sure how many truck drivers will pick up work on the grand unified theory of physics, but if even one makes a new and important discovery that would be amazing. Also, I wonder how long until even jobs held by the highly educated (healthcare...etc) are largely automated?

whatshisface 8y ago
>What would be awesome is if those millions were writing books, doing physics, or creating works of art instead. You'd have machines doing the labor and humans being freed up to focus on new contributions to society. There is a really easy solution to this that I have never seen suggested. Instead of giving money to everyone for free, why not raise the budgets of research and art institutions? You can get as many people as you want doing physics and making art as you want, all you have to do is give money to CERN. Now, a pessimist might say that some people can't advance science or art because they lack the capability. I counter with the fact that anyone can do it at some speed, some are just faster than others. If you greatly increased the amount of money the Institute for Advanced Study had, they would be able to bring people on to their staff that were less efficient on a dollars per fundamental insight scale, I.E. you or me.
- hoaw 8y ago
  The problem is that it isn't an income issue. It is a cost issue, which in turn is a power issue. As in who decides what and therefor who society is for. Taking underprivileged people, or even anyone else, and employing them at CERN isn't likely to work (Switzerland isn't exactly cheap or big on taxes). In the Nordic countries there are "folk high schools", which gives you education (paid via taxes), housing and food for ~$650 usd a month. They are mostly in the middle of nowhere though. Is it a good idea to educate people who gets squeezed out of society? Sure. But most societies are going the opposite way even now. Like many other things it isn't a technical issue, but a social one. There isn't really a need to wait for society to collapse before giving people affordable access to housing, education, health care etc.