Foxconn replaces '60,000 factory workers with robots'
bbc.com
bbc.com
Of course, two major problems here. First, the government, for obvious reasons, and second, the corporations that have a stronghold on the government. If the government did it's job, it would simply be re-funneling money where it needs to go; to pay for education, social services, safety, and so on. And if corporations didn't have their way, they wouldn't be able to twist laws and regulations in ways to maximize profit and avoid taxes, because that's precisely what taxes are designed to do -- cut profits.
But if history can teach us anything, it's that even when manufacturing jobs were shipped overseas and call center jobs were shipped and agricultural labor jobs were swept by undocumented immigrants, the government here didn't do much of anything, and the businesses got their way with increased profits. In turn, our labor force got displaced, but we survived because ultimately people find a way to make money.
Every non-government job is created by an entrepreneur, either directly, indirectly, or down the line. Entrepreneurship is really what has driven America. Many will take that over government and taxation (solving the two major problems outlined above), and this is the free market argument in a nutshell.
The takeaway is, if we tax robotic labor, they'd be working for us, the people. Just as anyone paying 15% in income tax is working 2 months out of the year for the USA. Or we could hustle it out and have everyone desperately seek work to save their lives. Our choice. One leads to greater inequality.
https://www.washingtonpost.com/world/asia_pacific/china-lift...
These are, fundamentally, political-economic questions. Not merely one or the other.
More-over, I hardly think a tariff on the output of robotic labor applied only to expats who left the country for tax reasons is going to cause a trade war...
My robot submarines will run your coast guard surveillance and deliver goods tariff free.
Yet somehow most countries are able to tax labor, income, land, and so on, without losing their entire populations.
Not to say that people won't leave because of taxes. Just that there are also reasons that make people stay.
- Look at your cell phone location information for last several months.
- Use Police's license plate trackers to track your car (Probably they do not even need court order for this one).
- They can get your car's location information (I assume high income people have one of modern expensive cars and most of them have tracker in case it is stolen).
- They can talk to your made(s) (I assume you have one, since you make 10+ mln) and pressure them to tell the true story.
Sounds really dangerous to lie about your location now days when it is so easy to check it, without even leaving the room.
That's because most of the population aren't free to give everything up because a government gradually makes their life slightly worse.
Just because the frog doesn't skip out of the saucepan, it doesn't magically mean that it's perfectly acceptable to keep turning up the gas.
Also, what cannot be ignored is the true drop in production costs because savings do get passed on to the consumer. This makes the politics difficult. If kittens died change would be much easier.
Not just corporations. All tax payers try to pay the least they can.
There are professional accountants who make a living by filing personal tax returns, and people pay them to advise them on tax deductions and investment schemes that lowers their tax bill.
Other way to avoid paying VAT would be to hide the sale altogether which might be hard for company that makes and sells tons of stuff made by robots.
sure robots and factories move to China with no taxes at first..but than Chinese gov figures out to pay for other infrastructure improvements to have things like the USA they have to tax something and end up taxing robots anyway..in fact its already happening
I mean, the Fed can conjure trillions of dollars of investment capital out of thin air, I think 350 million people have a fair chance of choosing to operate a magic building that makes their lives better.
Which is ultimately what we are talking about. Should we operate highly automated factories and how should we divvy up the benefits from them?
automated robots = free labor = us not having to do it + receive their products
... it just isn't that straight forward. Currently it's heading towards:
automated robots = free labor for business owners = us not having those jobs + them still getting what we would have produced
Or more accurately, we just stopped caring about the people who can't.
And to treat our fears about inevitably joining them, the denigration starts. "well, they're just red state rednecks" "Poor folks should just learn more and work harder and pull those bootstraps" "if there's 5 jobs for 10 people, then the 10 people need to go back to school and get degrees, then they'll have 5 jobs for 10 people with degrees, which will be much better"
"Utopians on the coasts occasionally feel obliged to dream up some scheme whereby the unnecessariat become useful again, but its crap and nobody ever holds them to it. If you even think about it for a minute, it becomes obvious: what if Sanders (or your political savior of choice) had won? Would that fix the Ohio river valley? Would it bring back Youngstown Sheet and Tube, or something comparable that could pay off a mortgage? Would it end the drug game in Appalachia, New England, and the Great Plains? Would it call back the economic viability of small farms in Illinois, of ranching in Oklahoma and Kansas? Would it make a hardware store viable again in Iowa, or a bookstore in Nevada? Who even bothers to pretend anymore?"
Though that may get messy as traditional human labour costs change. Maybe a straight-up imputed rent on the rental value of the machines?
This is a total non sequitur. Machines have been taking human jobs since the late 1800s and adding robots to take/process orders is no different. You discount the notion that new jobs will appear as "the free market argument," but new jobs will be created regardless of any tax on robot labor (negating the utility of such a tax) -- I suspect the driving factor here is the size of the unemployed workforce, not the need for money, ceteris paribus.
Moreover, the two problems you note (government and corporations influence on it) aren't policy or cultural problems (implementation details, if you will), they're problems with the fundamental nature of a powerful state and human society. Never has there existed a society where powerful institutions were not able to influence other, more powerful institutions (i.e. the state) with that said power, and as long as those institutions are run by men that cannot change. Individuals are generally corrupt (to some extent or another, if you give me $5 I'll shut up), and no amount of oversight or transparency will stop corruption so long as a critical amount of power supports that corruption. In simple terms, this is the tyranny of the majority, where "the majority" is measured in terms of power, not votes. This is the fundamental political dilemma, and limiting the scope of the state to only limiting the power of society's other powerful institutions is the only solution that solves this problem empirically, as opposed to the notion that corruption is ok as long as it promotes a certain set of ideals with which I agree.
I would add that humans have been paid to do repetitive, automated, routine work for ever, and they do so inefficiently. It's not human nature to stand in a production line for 8 to 12 hours straight placing screws somewhere.
There is something profoundly wrong with the life expectations of some if they are actually defending drone work.
I don't.
> I suspect the driving factor here is the size of the unemployed workforce, not the need for money
And the difference is? Fundamentally people work to pay bills.
> aren't policy or cultural problems
State is policy and human society is culture. You seem to think the latter terms imply some innate quality absent in the former, but they both change, and that is because they are one in the same. Both have changed rapidly, and it's called history.
> "the majority" is measured in terms of power, not votes.
You're describing corrupt nations. This is blatantly anti-democratic, and there are too many people working to change this in America who have already succeeded to bring us to where we are today, and who will continue to succeed to help build a better tomorrow.
Limiting power is not ideal. Power does not need to be contaminated or broken, and presuming that part of it will never change is discrediting everyone that got us here. It's okay not to be satisfied because that is what will take us even further. But we're not Brazil. And their problems won't go away just by limiting power of government.
A single tractor is the sole responsible for eliminating the need for hiding a small brigade of farmhands or construction workers.
Efficiency shouldn't be taxed.
However, if robots ever became fully autonomous and sentient, then owning them could be equivalent to slaveownership, and fundamentally immoral.
Slaveowners can be said to collect rent, because just like with owners of land titles, their profit comes not from production but from the government giving them privileges.
Not according to neoclassical economics! Land is just another form of capital. I still don't understand why though. Some say it's political but that feels like too much of a conspiracy theory to me. There must be a good explanation but I haven't found it yet.
The idea of taxing automation sounds like discouragement of increasing the efficiency. If anything we should subsidise it!
The rent houses produce is very different from the money robots help you make. Rents (in the economic sense) are good to tax, can't be passed on to the tenant, and has no bad distortionary effects. If you tax robots, and don't do anything else, all that does is raise the prices of the goods produced by robots. It harms the consumer the most. If you tax robots enough, you can incentivize the owners to not use robots and hire humans instead. But then we're missing out on all the benefits of having a society where robots do the hard work, there's got to be a better way.
One solution would be to tax robots, and redistribute the taxed money back to the consumers. Everyone should benefit from the fruits of automation. This would probably work pretty well. The biggest issue here is that the government isn't known for being very good at redistributing tax revenue. So there are proposals like basic income to keep the process as simple as possible. There are probably even better options, like taxing the rents only, and taxing them very highly, and then paying that out as basic income. Theoretically this might target the people you want to be targeting, without harming the consumers. Obviously, theory only goes so far with this kind of thing. I think the point here is that taxation is tricky, and the burden of a tax is not necessarily on the person/company you're taking the money from. So "tax the robots" might not be the best thing, even though it has populist appeal.
Real estate taxes are in principle very similar to taxing a robot. It will be passed on to the consumer.
If you are referring to rental income being taxed (after all costs are deducted) then it is very similar to profits from robot assisted manufacturing being taxed. This will happen in the current tax regime in the US and here both are similarly situated as well.
Better, smarter robots are a threat to Chinese workers, not so much to say Dutch or Canadian workers.
The west can afford and is willing to pay more even for less, just look at how many people are willing to pay 10 times the price for "Organic Fare Traded Tomatoes" or any other "artisan" goods.
The west is also not as much dependent on manufacturing and the manufacturing it does do is usually high tech, low(er) volume and highly specialized one where it both might not make sense to automate everything and where people are more than willing to pay the extra to know that something was "hand crafted".
But overall we've developed a consumer culture especially in the past decade that is pretty much focused on retro/old school craftsmanship, small scale production and "artisan" touches, it almost makes one wonder if the whole "hipster revolution" wasn't hand crafted by some cabal of shadowy figures that foresaw the "rise of the machines", because our willingness to pay 250$ for a hand crafted hemp bag on etsy is just might be the thing that would protect us from AI automation.
Already there are a lot of junior lawyers being replaced [0].
Check out what a former CEO of McDonald's said about robots and the fast food industry [1].
> “I was at the National Restaurant Show yesterday and if you look at the robotic devices that are coming into the restaurant industry – it’s cheaper to buy a $35,000 robotic arm than it is to hire an employee who’s inefficient making $15 an hour bagging french fries,” the former US chief executive Ed Rensi told Fox Business’ Maria Bartiromo.
[0] http://www.nytimes.com/2011/03/05/science/05legal.html [1] http://www.theguardian.com/us-news/2016/may/25/former-mcdona...
[edit] then again, automation of software development tasks held back the software industry at all...
I think a lot of the service sector automation that will happen in the next 20+ years is analogous. Anything customer-facing is bound to stick around as long as people have spending money, while anything that's not customer-facing will slowly become automated.
E.g. The person at the counter in a McDonalds is the easiest person to replace -- all you really need is an iPad and a couple accessories. But if I were a betting man, I would bet that McDonalds will replace food prep jobs before replacing humans with kiosks in most locations.
Go into any "artisan" coffee shop and the barista would pretty much look identical, many of them will engage and have their own "shtick" to make the experience of drinking their coffee "unique". The coffee it self is also going to be heavily branded and sold to you as some fantastic story with many images which again will be heavily focused on the people who were involved in the entire process from growing the beams to your cup.
This is what we are buying now, 20$> an ounce "fair traded organic beans" with some nice retro photographs of Bolivian coffee farmers combined with some 20 year old millennials inspecting the beans to make sure you get the "best" ones in your cup.
This is what you can't replace, brewing the coffee and even foaming milk is something that a machine can do easily, and arguably better than most people, heck even latte drawing can probably be better done with a 2-axis drip nozzle than a barista with a bamboo stick but when you pay 7$ for a cup of coffee you aren't paying for coffee you are paying for the experience that no machine can actually give you simply because you don't want it. You want the human contact because at least in my personal belief that is a coping mechanism we have adopted to make for the isolationist life style that many of us live today in which we spend more time with our phones than with people, and so we care now more about how our coffee and burritos are being made and by whom than ever before.
Is this true?
I'm mostly convinced about the brewing part (except that even the most meticulous machine will make mistakes, and a good barista will throw out a bum shot whereas a machine probably won't know the difference).
But I'm not convinced about the milk part. Do you have links?
(FWIW I totally agree with the tone and overall thesis of your post. And I really wish someone would deploy freshly ground-or-vacuum-sealed/fresh milk machines in airports/cafeterias/hotels/anywhere there's only a starbucks. But I also feel like better-than-average humans are still better than state-of-the-art machines.)
There's still room for the human touch but sometimes you just want a decent double espresso and to be left alone.
For human contact, just have a nice coffee shop with comfy chairs where people can sit around and read, chat, etc. You don't need humans making the food for your human contact, you just need to be in an environment with other human customers.
You can also have a human server who brings the drink you ordered to you. One human can run the whole establishment, filling in in all the ways the robots can't, but letting the robots do the more skilled work. When I go someplace for food or drink, I want to sit down and relax, not talk to the guy prepping the food. And since the food prep necessarily has to be done someplace away from the seating area (because that's where the foaming-milk machine is), I don't see any value in having a human back there when a machine can do it better. Now I do see some value maybe in being able to chat up the girl who brings the drink to me where I'm sitting. But the person behind a counter making the drink, who I can barely even see because there's a giant milk-steaming machine in the way? No, I don't see the value there.
* A good barista will notice when they screw up the shot and pull pull another.
* A good barista will not burn or under-steam milk.
* A good barista will make sure you have the proper amount of milk and foam. (Getting a perfectly consistent result when steaming milk isn't really possible using a dead-reckoning controller or when using simple sensors like temperature AFAICT.)
* A good barista will make sure the drink looks as good as it smells and tastes.
* A good barista will have no problem customizing drinks.
Totally. But the current generation of dead-reckoning-and-maybe-a-bit-of-very-simple-sensing style of control for grinding/tamping/steaming/pouring won't be sufficient. I would be unsurprised if substantially improved sensors (perhaps even vision) are necessary before we see computers over-taking the best baristas in an abssolute sense.
And unfortunately I don't think there's much demand for that sort of investment for the reasons outlined in other posts on this thread (even if robotic baristas could smash the human competition in coffee making, they still wouldn't displace the neighborhood coffee shops.)
But it is just interesting, from a technical/hacker perspective, that good barista work is a pretty hard robotics problem!
While I have bought alcohol from a bar, the ratio of my non-bar alcohol still dwarfs bar-purchased alcohol. And frankly, even with the bar-purchased alcohol, it was social events with friends, the bartender could just as easily been a serve-yourself machine.
Really fucking cool, and the coffee is excellent, too.
Except that most people are REALLY BAD at operating a kiosk. You want a person at the kiosk so that they can gently guide the SUV driving suburban wife with the high maintenance order gently out of the way of the rest of the customers.
The backend jobs will be the first to get automated.
It uses pictures and text. I will have to watch next time I'm in there to get coffee and see how well people use it.
McDonalds has to compete with other chains and prices are already bottom barrel. In McDonal's case, taking human interaction out of the equation seems like a losing proposition.
Even if that's the case, you could still plausibly turn three to four people manning a register into four kiosks and one person to assist.
My success rate for getting my order correctly at fast food restaurants is somewhere south of 50%.
It's currently optional, but I could see it becoming standard fairly soon.
Panera Bread proves you wrong. Go into any corporate-owned store (not the franchise ones, they're behind), and they have iPad kiosks for taking your order. The food is still prepared by hand. They frequently even give you a bonus, like a free cookie, if you order with the kiosk instead of going to the human. And the kiosk does a far better job too: there's little or no line, you can easily look up your old orders or favorites and re-order them, you can easily make changes to your order (more or less salt/pepper, more or less lettuce, different bread selection, etc.) which you can't easily do with a human because the human won't present dozens of different options to you when you order, they just ask "what do you want?". Visual communication systems are far more information-dense than verbal ones.
for essentially two reasons; a) mcdonalds treats their employees like shit. no job is better than that job ( i live in a country with social welfare) b) the person bagging the food scratched his balls just before doing so, and he sure as hell isn't bothering to use gloves now.
Too much salt on the other hand...
Bartender on a cruise ship.
The problem isn't so much east vs west. The problem is technology is cheaper than people so unskilled labour is easily replaceable.
Not this year. Not next year. But they will.
Won't displaced truck drivers become motivated to be owners themselves? It doesn't seem as bleak as everyone makes out. I don't really think most big companies want to own and maintain their own fleets of trucks.
The idea that everything will work out because working people can just become capital owners ignores the obvious fact that working people work precisely because they don't have capital.
Car manufacturing: We still build cars here, but we don't employ nearly as many people.
Logging: contrary to popular belief, the spotted owl didn't destroy logging jobs. Machines that let 1 person replace 20 did.
Manufacturing is the easiest, first place to replace workers, but AI will replace lots of people in America. It looks like trucking and taxis are on the short list - that's a few million workers - but fast food restaurants are up next (and those innovations will move up the chain to fancier restaurants).
Over the last 200 years, there was almost always a new role for people when their job was automated, but over the last 20 years a lot of those new roles were lower paying service jobs.
No way is our consumer culture actually based on small scale production and artisan touches - and not that many of those artisan touches are made by people or machines in China. That may be what you notice, but most of America's paycheck goes to the regular old normal stuff.
The manufacturing in western nations in many cases is already as automated as it can get (yes there is always a room for more optimization but industries that have been automated since the 70's will and are already hitting diminishing returns), what has not been automated are mostly tasks that we choose not to automate for various reasons, many of which are cultural.
And yes most of the stuff we buy is mass produced, generic, and already highly automated but it's also not made in the US, the US consumer doesn't care about Chinese workers or unemployment in China, and it won't buy a 250$ hand crafted anything in China unless they are caught drunk in a tourist trap. And that's exactly the point artisan crafting and manufacturing has been developing in the US and it is pretty much immune to threats from automation and the AI/Machine Learning revolution because we as consumers are making a conscious decision to buy it. And even when we do buy mass produced generic crap just look at how many products are marketed today by selling you the human story from shoes to coffee beans it's all the same.
[0] http://www.cambridge-news.co.uk/dozen-modern-day-car-wash-sl...
Most the stuff we buy isn't stuff - it's services, lawyers, finance, insurance, health care. AI will dramatically reduce employment in every single one of those industries. (https://en.wikipedia.org/wiki/Economy_of_the_United_States#G...)
Self serve shops where one staff member supervises 20 registers with shoppers doing the shop work!
These were the jobs that young people used two learn how to work at a job. I worked the register when I was a teenager and it taught me well and "can be trusted with money" is an important resume tickbox.
The main exception is the few states that don't let customers pump their own gas and the service stations that still exist.
No more lumberjacks.
[1] https://www.youtube.com/watch?v=luygMuoFnXs [2] http://www.ponsse.com
[0] http://www.nbcnews.com/business/business-news/adidas-pulls-b...
Like we've got a ton of truck drivers, who'll surely disappear within a few decades, which is just one at most two generations. In terms of policy planning this is a short-term timeframe.
Same for most people who work in retail, from clothes shops to supermarkets.
But even higher-level jobs will change massively. I highly doubt education will stay as labour-intensive. And I've got family members who are attorneys, and when I speak to them they'll tell me a ton of their workload is simply going through legal documents to look up stuff, and then preparing mostly template legal documents. A lot of this is manual work that'll disappear sooner than you'd think. We've already seen examples in this field with the automated bot that helps void parking tickets.
The amount of people that work high-skilled labour is really small if you ask me. I think a decent first order approximation is to compare the amount of people that work at the Walmart or Amazon headquarters, versus their stores and warehouses. For Amazon I've seen numbers that it's roughly 24k vs 240k, and consider that this is a tech-oriented company that's already automated a lot. For walmart it'd likely be 1:1000. Take a Facebook with 12k employees, a leader in a high-skill industry, and it's the margin of error of the employee count of a leader in a low-skill industry, not in China, but right here in the west, like say Walmart (> 1m employees) or say a franchise like McDonalds. These are also Dutch or Canadian workers.
An anesthesiologist friend of mine went back and got his MBA because he's certain that machines will dramatically reduce the amount of jobs for anesthesiologists; you'll have one anesthesiologist supervising X machines instead of X anesthesiologists.
Artificial intelligence was now out-performing the capabilities of junior
lawyers and paralegals in due diligence, discovery and document drafting.
Upcoming predictive analytics would be able to assess the outcome of court
cases while legal diagnostics systems analysed legal problems and provided
advice.
I bet there will be a lot of refinancing of mortgages if the closing costs were $500 and the process was one click.
I presume that this is the case with a lot of professional services. In that sense service based economies are better situated than manufacturing based economies.
etsy: $2.3B [0]
target: $73B [1]
williams sonoma: $5B [2]
bed bath and beyond: $12B [3]
[0] https://investors.etsy.com/phoenix.zhtml?c=253952&p=irol-new...
[1] https://www.google.com/finance?q=NYSE%3ATGT&fstype=ii&ei=ruN...
[2] https://www.google.com/finance?q=NYSE%3AWSM&fstype=ii&ei=EOR...
[3] https://www.google.com/finance?q=NASDAQ%3ABBBY&fstype=ii&ei=...
I hope you like playing with Play Doh!
Just as mass industrialization necessitated huge social change in response. The emergence of a middle class, the growth of cities, organized labor and work hours, etc.
As automation and AI continue to replace jobs and more and more people find it difficult to compete at any price. I predict you will see similar societal changes in response. I don't know what those will be, and they probably will not come easy. We could see a utopian future similar to Star Trek, or a dystopian one instead.
We live in interesting times. :)
Interestingly, someone has to purchase the goods that these systems create. Thus, giving people the funds to do so, regardless of where they come from, seems like a necessity.
The connection between mounted warfare and feudalism is controversial [1] but it's common enough that societies adopt social structures optimized for defense. There were almost certainly multiple reasons for the prevalence of feudal relations.
60,000 jobs were just destroyed by automation, this isnt hypothesizing anymore its reality.
Where do these 60,000 people go now? What are the new jobs that these robots have created?
This is what you are proposing as the solution to the problem of job shortages?
His belief was that on one hand capitalism was/is absolutely necessary to create sufficient wealth to make redistribution viable (he made the point in The German Ideology that a redistribution before that point would do what we saw e.g. in the Soviet Union: simply spread poverty around, and restart the class struggle it tried to erase as people tried to et out of poverty), on the other hand it would in its end-stages lead to "overproduction" combined with mass unemployment as it would fail to reconcile the need to try to win more of the market with the need to cut employment costs.
The ways society could prevent this happening are many, including things like basic minimum income as you suggest. But Marx argument was that societal changes of this magnitude tends to instead result in increasing tensions until revolutions erupt for the reason that ruling classes rarely see the gravity of the situation until it is too late.
There are many ways he could be wrong about this, but I find it very interesting to observe the increase in frequency with which this issue of automation and its consequencs is now coming up again.
He also foresaw the shift from manual labor to intellectual property as the essential component in production.
What Marx saw as an inexorable trend towards socialism may have in fact just been a temporary consequence of the industrial revolution, wherein labor was especially important and the power of an individual worker was large in historical terms. It's not impossible to imagine a sort of "Neo Feudalism" where a small minority of elites find it cheaper to maintain control via technological force-multipliers than to share their earnings such that everyone is actually happy or nearly so.
His message was less about how important labor was, but how labor got squeezed between the industrial capitalists owning the factories, and the rentier capitalists who owned the land, housing and issued loans.
We don't have to. There are many complicated jobs that aren't white collar but still require a highly skilled individual and are not currently achievable by robotics (millwrights for example).
The OP is referring to removing monotonous, simplistic, repetitive jobs.
Do you have a citation for that? I would be interested in seeing that. According to the Chinese government (which has been accused, in fairness, of fudging the numbers) their unemployment rate is highly stable.
Historically, as countries transition from developing to developed their workforce will shift from primarily non-skilled labour to primarily skilled labour [1]. China appears to be following this line naturally, if you trust their reported unemployment rates.
[1]: http://images.slideplayer.com/21/6263323/slides/slide_12.jpg
The generation preceding the current generation will be better educated and take-on the more skill demanding jobs.
Basically hiring people to dig and fill up white-collared holes.
Sure it will, because it'll force large social changes and such changes are interesting regardless of whether they're bad or good.
World War III breaking out would most certainly be considered "interesting" news, although there is absolutely nothing positive about it.
Says who? Addidas is moving its robot factory into Germany and the US. The perceive benefit you see from China doesn't exist once we have proper automation.
http://fortune.com/2016/05/25/adidas-robot-speedfactories/
http://www.theguardian.com/world/2016/may/25/adidas-to-sell-...
>Try sourcing a new chip in the states and then try over there.
Fabs are different because western nations have strict environmental laws. China's big benefit with fabs is its willingness to pollute its lands and poison its people for us. This is a negative all around and the sooner we get fabs out of China and into environmentally responsible countries and using environmentally responsible methods the better.
1: http://www.npr.org/series/248799434/planet-moneys-t-shirt-pr...
2: http://www.npr.org/sections/money/2013/12/04/247360787/our-i...
There are those that are seriously concerned about China's ability to compete, including the Chinese Government.
People who say it is impossible to move production from one country to another don't understand what the word 'impossible' means.
That's also the answer to 'can we move production under the sea?'
Also it is not limited to electronics - as others have pointed out, Addidas moved automation back to Germany.
What I think most people mean is that moving production is impractical, e.g., that there is no realistic incentive that justifies making the investment and so for that reason nobody will make the investment, and production will not move, etc.
It would be more effective to argue that the required investment is smaller than expected, and some particular entity does have an incentive to make the investment after all, etc.
once that happens it's a 'snowball', if the price is right the supply chain will catch up.
PS: see the link posted in another comment: http://www.theguardian.com/world/2016/may/25/adidas-to-sell-...
Energy costs also become increasingly relevant as labor costs drop to a smaller fraction. Why not set up in iceland where geothermal electricity is free.
I guess you could always pay off the local warlord, then pay for your own roads, and all the rest... But perhaps better to have a government take care of it for you?
Getting smart people to come on contract to live in luxury and service your robots won't be hard. Why on earth would you live in a dingy NYC or San Francisco apartment if you could go somewhere and as merely an engineer be waited on hand and foot in a luxury condo while saving up bank. Go golfing after work every day for free.
1. Security
2. Friends
3. Family
4. Security
I've lived overseas in Cambodia and Taiwan. The lower cost of living, warm weather, and challenges of living in a different culture can make life pretty enjoyable.
I chose both places carefully and visited before deciding to move. I still visit the US once a year and my vacations now take me to places I would never go if I lived in the US. A weekend trip to the Philippines, Vietnam, South Korea, China, or Japan is feasible from Taiwan. From Cambodia, you're right next to Vietnam, Thailand, and Laos, all safe places if you read up on what to watch out for.
Taiwan is much more developed than Cambodia, and yes, are nothing like Somalia. That's not a place I would choose to live, myself.
Rationale: you now just need to bulk order the raw materials / components and "feed the machine".
We aren't there yet, but jeez we seem to be racing there quick?
The supply chain to support making shoes seems like it would be MUCH smaller.
Germany, in fact, is in the business of producing the machinery that is then used in other countries to produce your phones, tanks, ice cream cones.
It would take a relatively small tweak to import duties to make most types of outsourcing uneconomical.
https://erdbeerjoghurt150g.files.wordpress.com/2008/06/joghu...
Example: Zara produces in Spain as high fashion items have a very short shelf life, whereas a container with skirts takes a bit to come from China to Europe.
If I gain an advantage by producing closer to my buyer, I'll do it.
Just-In-Time production becomes even tighter. That Amazon Prime 24h delivery item could come hot of the press.
Their Bangladesh suppliers have ~300,000 workers vs. ~6,000 in Spain [0]. To counter your point, it seems that a piece of clothing manufactured in China will be shipped to Spain, quality controlled, tagged, fitted with RF-ID and then sent back to China for selling [1]. So it doesn't seem to follow the model you suggest where proximity matters.
[0] http://www.elespanol.com/economia/20160222/104239851_0.html [1] http://economia.elpais.com/economia/2015/03/17/actualidad/14...
The solution going forward is specialization. We handed manufacturing to China while our service sector exploded. Let each country produce those goods it does best and then trade with one another. Why bring production back to the US? Only if it makes sense to do so.
Interesting.
The supply chains are based around china, india being the primary manufactoring.
e.g. There is a statistic that peak manufacturing output for the UK was actually fairly recently:
"A 2009 report from PricewaterhouseCoopers, citing data from the UK Office for National Statistics, stated that manufacturing output (gross value added at 2007 prices) has increased in 35 of the 50 years between 1958 and 2007, and output in 2007 was at record levels, approximately double that in 1958"
https://en.wikipedia.org/wiki/Manufacturing_in_the_United_Ki...
Another factor in this matter is that the uncomfortable reality is that "we" need to stave off the influx of uneducated and unskilled people. Reality is that people will lose ever more value, especially as, rather oddly, the liberal autonomist types advocate for things the they don't realize will only commodify the vast majority of people.
I hope that the spokesperson was speaking the truth when they said that the increased automation would not lead to long-term loss of employment.
But I can't imagine reeducating that much (previously) low(?) skilled employees, rather than hiring a new graduate (of which there are reportedly many) for R&D.
Increasingly to me a universal income guaranteed by the government seems like sensible policy to ease in sooner than later. I see no other option, lobbying to allow lower minimum wages or eradication of minimum wage to keep humans competitive against mass produced robotics and AI is quite frankly idiotic.
`High` is subjective. $x minimum wage did not prompt automaton, a rising minimum wage catalyses R&D into automaton with better incentives. It's inevitable, and Mr CEO was always going to fire everyone and replace them with robots the moment it became cost effective.
> How UBI will be funded is another matter
Larger profit margins for big corporations turning to automaton, let's tax them more.
Raising minimum wage is for workers who are not going to be fired because of automation in the next 5+ years.
Mind you, the robots-are-taking-our-jobs predictors have been wrong before. But if they aren't, the issue I see is that a big spoke of the modern capitalism wheel revolves around employment. Employment provides income that supports a wider range of people than mere rentiers can. These employees can in turn spend this income at businesses. If there is no consumer income, any business that depends on consumer demand... well, they have a problem.
I guess it's easy to envision things being pushed a lot further, where automation has eliminated huge amounts of work. The main economy would thus largely be revolving around rentiers, who are unaffected by a lack of jobs, people that support the automation (which so far has been a relatively small group), and any other employment which can't be automated (in an extreme scenario this would be very little). From my perspective, such an economy would hardly look like modern capitalism at all, it would potentially look more like previous rentier-dominated economies. Because jobs are such an important pillar in 20th century life, there will probably also be a culture shock on top of that. Without some changes, it's possible that this transition will be rather rough (even if the automation doesn't quite go so far as some predict).
There are other factors that come into play as well like overtime pay rules, medical costs/regulations for workers etc.
Not only should we not try because we can't win on economic grounds (I think). It'll probably force the people that must compete into inhumane working conditions for negligent compensation before we admit that we lost. Banning automation is also not ethical. Because increased efficiency also indirectly translates into increased quality of life, and directly translates into better health care and lower mortality rates.
You're probably right about the basic income, though I wish we could skip that part. My personal crystal boll says that we'll end up with a system more aching to that in the star trek universe anyway. Where _things_ are free, instead of giving people free money.
Even in post-scarcity world of things, there are non-divisible goods, like land, access to people (think, concert) and so on. So some form of allocation of resources would need to happen. Money is one way to do that. Others exist too like lottery, or government ownership are options too.
I'm all for post-scarcity, and think we can go a long way toward just having things be allocated and free, but you'll never get away from the miserable business of allocating scarce resources.
The economics of Star Trek are the incoherent and inconsistent mass of contradictions that one would expect of a system in the background of fiction where the focus is elsewhere and no one was doing detail socio-economic worldbuilding in constructing the setting.
I mean, on the most basic economic element: the Federation both explicitly does not have money, and explicitly has a defined unit of money (the Federation credit) to which there are several references to its use in trade.
That is at least theoretically debatable.
Demand is a function of price. Imagine a human driven factory producing 10 parts for $100 in wages, and a competitor using humans and robots that can produce 11 parts for $100 in wages. The competitor can bid prices down, and not only capture the human only factory's orders, but create new orders as it becomes cheap enough to use the part in new ways.
The empirical question is whether that increase of orders makes up for the decrease in labor demanded per unit. It's not immediately obvious which way that goes, and there's multiple anecdotes provided law professor James Bessen of cases where increases in labor demand occur http://www.theatlantic.com/business/archive/2016/01/automati...
Nothing I've read of his suggests a conclusive 'more bots always means more jobs,' and perhaps his anecdotes are carefully chosen, but a blanket assumption that more computers implicitly means fewer people working is unwarranted.
Hopefully many countries will get an unconditional basic income and only a few become new Nazi Germany :\
Shelter, well that's a huge variable: building your own house is almost impossible with many regulations that require a certain expertise to navigate. But once you have food, shelter—and friends/family—you're in a decent spot from a survival standpoint.
We don't have to live like the Amish, but perhaps a happy medium can be found, by being less dependent on the economy. Instead of buying 100% of your survival needs, perhaps it's possible to buy 50% of them and make the rest yourself. With the better telecommunications we have today, I often wonder about a society that's only 50% plugged in to the economy—and 50% self-reliant.
Disclaimer: I'm no macro-economist.
Say growing your own food while limiting inputs from the broader economy would take 30% of your available time. You can compare that to how much time you are currently spending earning money to pay for food (probably less than that).
Is the greater certainty that comes with the self reliance a good trade?
Contrast that to someone who is 100% self reliant. They don't have friends or family to lean on. Every day they have to figure out where their next meal comes from, where they will sleep. Basic safety is not even guaranteed. (You can't be 100% on guard, without break.) The constant demand to take care of one's basic needs slowly wears them down. They have to struggle just to stay clean and well dressed. If you live in a city, you see people like this all the time. They're not part of social support system that benefits economically successful people – and the results are tragic.
So yes, there's a bias in my proposition.
If one nation taxes the robot overlords and gives the money to unskilled humans... then the robot overlords will move to a friendlier nation, while also seeing inflows of unskilled humans immigrating.
The immigration issue is a difficult one, because the nations most likely to embrace basic income are also the least likely to embrace draconian immigration controls. The movement of capital issue is more difficult still, and perhaps outright impossible. How do you break the race-to-the-bottom cycle if some nations opt-out of the new order?
Basic income is an interesting idea, but I haven't heard any plausible proposals for it that wouldn't depend upon nations giving up sovereignty and moving closer to global governance. Back here in reality, Europe is struggling to keep a single continent loosely tied together.
This is a bad assumption, the robot overlords can't just move to another nation, markets don't work that way; they're here because they're selling in this market, you don't just say oh well I'll stop selling stuff in the US and only sell stuff in Germany because they're going to tax me less as it's a much smaller market and a culturally different one as well. What sells in the U.S. won't be what sells in another country, and other markets are much smaller and/or would require complete culturally relevant re-branding to a large degree that might not be possible. The U.S. is a huge basically mono culture market, the E.U. is not, you can't just move shop, you have to change the entire way you do business.
tldr; markets are sticky, the rich people don't/won't/can't just leave when tax policy doesn't swing their way because their money machines are tied to those markets.
This is of course untrue (its technically true as foxconn hires a lot of temps/seasonal/demand workers it doesnt consider employees). I think China is in some deep water when it comes to automation. Their main productivity is based on having lots of cheap human labor. If they're not careful, they'll fall into an India-like slump where over-population means lots of servants and such, but no real industry, growth, or innovations.
Addidas is moving back to Germany soon also. They are opening a fully robotic factory. This stuff is moving fast. Jobs are definitely going to evaporate here. China is probably headed for another peasant revolution sooner than later if this moves too fast. Building your economy on making other people's things isn't a smart move.
http://www.theguardian.com/world/2016/may/25/adidas-to-sell-...
When automation occurs, the wages previously paid to the workers doesn't just evaporate into thin air, but it moves around into other areas.
I'm sure there are lots of studies out there that might show something different in reality, but the world is richer as a whole than it's ever been. It may fluctuate in allocation, but it's still richer overall. Now I suppose overall richness and unemployment are two different things, but the poverty level is lower than ever, which is a good sign. I suppose the challenge is for poor/middle class learning how to better separate the wealthy from their wealth at a faster rate, in a productive way?
Of course, there is always friction in the redistribution process of the labor, but in the long run, I can't help but be optimistic about the increased human potential that can result from moving people away from menial tasks. Unless of course, you believe that some people are better off with tasks that require low creativity and human initiative.
It's not that I don't want a world where automation makes shit-work go away, it's that I want that future to be empathetic to, and understand that it needs to incorporate a solution for those who don't have the power to change their situation. Otherwise, what's the point?
- those people were not well educated so that kind of work was something they could do and for a decent salary
- that was a job and they had pride in being employed and useful
- that job was not mentally exhausting so they could think about something else or talk while working; many found that quite enjoyable, socially interesting if you will
- they were doing different things, going from one "simple" task to another and so they all shared a knowledge ofthe build process they were involved in. They were proud of that as well.
So well, those "not very interesting" tasks were actually taken care of very seriously and with pride. So in the end the "how it was done" was just as important than "what was done".
That's why if robots take over, I can't help but see that these people will miss something they loved. You can say that's the price to pay but, go and say that in front of them; you'll understand what the "it's-logical-unescapable" kind of argument really means...
Thank you for saying this. Not many people on HN mention that.
However, if people are laid off and they don't get other jobs then yes that will indirectly lower GDP too.
If you have 11 pounds of flour to put in a 10 pound bag, no matter what you do with applying nice looking labels, its not gonna fit.
The only way to reduce GDP is by lowering the price of your goods (assuming demand stays constant), or implicitly by letting your excess profits sit as cash instead of consuming or reinvesting (which both create and increase value).
If you are selling in a competitive market, the price for per gadget will drop with the price of production, reducing GDP before considering other effects (while the purchasers will probably spend their additional cash on some other good, there is no guarantee that good will come from your country, so while world product may not decrease, your domestic product may.)
If that's not the case and you have pure monopoly pricing power, then the effect on GDP depends on how you spend the excess profit: if you spend it in ways that have the same velocity in your domestic economy as your old employees would have (and if the maintenance expenditures for the robots likewise are spent in such a way), the effect is neutral -- if the net of these two things is spending that has less domestic velocity, then GDP could still drop as a fairly direct result of the decision to bring in the robots.
However, big business are often corrupt so wouldn't surprise me equally if there was some sort of cartel. It's strongly in their collective interests to not drop the prices.
If the country had 100 people making gadgets, and now it only requires 50 to make the same number of gadgets, the country can now double the amount of gadgets it makes, or the 50 people can go make something else.
Therefore, a doubling in productivity will not decrease GDP (total production of an economy), but increase it.
The alternative demand constrained world with excess supply would look very different
Create an economy where the vast majority are unable to consume very much while aggressively expanding productivity and you'll create an economy with a demand shortfall.
This vicious circle has to stop, or the planet will stop it for us. We should think about how to work less and consume less.
Alternatively, imagine that someone makes a universal bot-slave that provides for all of your needs, manufacturing as necessary from raw inputs. It just buys the natural resources for pennies and makes them for you. The only inter-human commerce is for the natural resources (delegated through the bot-slave).
In that case, almost none of the production shows up in GDP accounting (because the bot just makes it and gives it to you), yet people are much better off.
But it is possible, and does happen, that technological change messes up GDP measurement. It was definitely a metric formulated without changes in the production process or even consumption process in mind.
But even if those don't happen and you have lowered GDP, a lower GDP is not a bad thing; it's perfectly acceptable, and desirable, for people to spend a smaller fraction of their income.
As for your company, if you can reduce costs such that instead of making $50 on each $1000 gadget, you make $100 on each $500 gadget, both you and the consumer of your product win. Arguably, GDP seems broken as a metric for not taking that into account.
Lower real GDP doesn't mean lower individual spending, it means lower domestic production of value relative the cost of living.
Lower nominal GDP also doesn't mean lower individual spending, it means lower domestic production in terms of some particular unit of currency, ignoring cost of living.
Both spending and capital investment within the domestic economy (that is, excluding imports), as well as all exports (whether consumer or capital goods) are included in GDP.
> As for your company, if you can reduce costs such that instead of making $50 on each $1000 gadget, you make $100 on each $500 gadget, both you and the consumer of your product win.
But the suppliers of the inputs to the product lose.
> Arguably, GDP seems broken as a metric for not taking that into account.
GDP does take all of that into account, including the parts you ignore.
The point I was trying to make is that production/GDP alone does not seem like an entirely appropriate metric, if you don't necessarily consider the increased production and consumption (of "gadgets" or otherwise) a universally desirable property. Be careful what you measure and reduce to a numeric metric.
> But the suppliers of the inputs to the product lose.
GDP doesn't include that. GDP includes "final goods and services", not intermediate parts or business-to-business transactions. There are other metrics that include the intermediate steps, if that's desirable.
Basically, $ used to cost 0.001 gadgets, but now it costs 0.002 gadgets. But your country's Product, measured in gadgets, does not change.
Competition means the price you charge the importer has to drop by "something" otherwise your competitors (assuming you haven't used the .gov to outlaw competition) will undercut you. Unless you use .gov or "something" to prevent competitors from installing the same, or even newer better robots.
Surely as per efficient market hypothesis the employees working there were already at their best possible job over a long term average over a large group of people, like an entire factory. So an external force eliminating those jobs mean in an efficient market they'll by definition be working less well matched to their abilities jobs. This ripples thru the economy see anecdotes about aerospace engineers driving taxi and so on. Whoops taxi driving jobs are going away too. So society as a whole will be less productive, have less revenue, have an inferior match of their abilities to their jobs. This will further depress the GDP.
Traditional income inequality models indicate the marginal use of a dollar / yuan is much lower for small higher income brackets. Not just the wealthy CEO but say, engineers. So the impact on the greater economy of taking away 1000 line jobs is much higher than taking away 1000 line jobs equivalent out of the CEO's bonus. This is not seen as overly controversial, obviously the velocity of any given yuan is much faster for a poor dude than a rich dude.
Finally there's the multiplier effect where productive jobs that generated 1 yuan or whatever also generated as a side effect service transactions of 1.5 or 3 or 10 yuan per productively generated yuan. Obviously the real estate agent won't be selling homes to the now unemployed people which has downstream effects like the commissioned mortgage broker and commissioned bank reps all the way down to title registration won't get those service dollars, er, yuans.
There seems to be no reason to expect the GDP to increase or stay the same. There is a tragedy of the commons aspect where every individual benefits individually, but society suffers. Its much like dumping industrial waste into parks, for example.
If we are comparing economies that are, in a suitable sense, "the same shape" then more GDP is better. But adopting new technologies in the way that your question imagines, changes the shape in a way that breaks the proxy. GDP goes down and that is a good thing. No surprises or paradoxes here; GDP is well known to be a poor proxy for wealth.
http://lesswrong.com/lw/jmm/true_numbers_and_fake_numbers/ai... works through the details, but the other way round - the robot breaks down, prices go up, GDP goes up and that is a bad thing.
Their FoxBot robots are conventional robot arms on fixed bases. There are four standard models, differing in size and precision. Foxconn is making them in-house, for about $20,000 to $25,000. Not clear what sensing they have; the ones they've shown in public just had simple suction end effectors.
There's nothing exotic here except the scale on which Foxconn is operating.
[1] http://www.scmp.com/tech/innovation/article/1829834/foxconns... [2] http://www.4erevolution.com/en/foxconn-foxbot-apple/
It must be driven by either population or productivity per capita increase, period.
Industrial economies cannot continue to grow in this unregulated way forever, due to it increasing both the environmental impacts and wealth disparity. That said, information economies can grow much more aggressively without the negative impacts- but we don't have a system for adequately valuing information GDP, i.e. rewarding people for creating content.
If our society is becoming entertainment-centric (which is great, and inevitable) then we need to figure out how every person can ascertain a quality-of-life standard through providing entertainment.
Eventually natural resources and industrial production will be completely automated. Obviously the trickle-down economic theory does not hold up in that situation.
I'd agree that ever-increasing GDP is not reality.
The thing we all disagree about is when a downturn will occur, and to what degree.
When you talk about a forthcoming downturn, it sounds like you think it'll happen this year. It's natural that some people will disagree.
Adidas comes back to Germany but with a heavily automated factory http://www.theguardian.com/world/2016/may/25/adidas-to-sell-... and ex McDonalds executive pointing out robots are cheaper than 15 bucks an hour employees http://arstechnica.com/business/2016/05/mcdonalds-ex-ceo-15h...
Automated at the bottom and outsourced at the top, going to be an interesting squeeze and I am more worried about all the screams for protectionism in trade which pretty much shut down the world the last time it was tried
We are already witnessing this revolt/discontent in the US in this presidential cycle in my view. People are very angry and I fear it will only become worse in the coming years/decades as promised solutions by the new crop of leaders elected by angry people don't work out as expected.
The dominant left narrative is to see automation as "job loss" but in reality automation is bring manufacturing back to USA. Helping Americans live in a more safer and better workplaces and rapidly increasing technological progress.
Also, people whose jobs are replaced don't just go and jump into river. They simply find some other jobs possible at lower wages thus helping other businesses get even more competitive.
Charlie's father was a factory worker who struggled to make ends meet, the irony being that his job was to screw toothpaste caps on toothpaste tubes.
After the events of the book, Charlie's father went back to the factory but as an engineer to work on the machines that do his old job. Not everyone at the factory could have received that kind of promotion. I wonder what happened to them?
Thousands, literally thousands of professions have been made redundant by mechanisation and automation over the last few hundred years, but we're still all just as busy as ever.
Look at all the professions that have been created in just the last 20 years thanks to the internet. Sure automation changes the nature of work, and the type of jobs that need doing, but I've yet to see a shred of evidence that continued advances in automation are going to put us all out of a job, other than vague "this time it's different".
With the death of those thousands of jobs, we now have tens of thousands of new jobs, all filled by the billions of people, and boy is it great. Our work is more fulfilling as we get more done in less time in ways that improve the world while earning money that buys magnitudes more than what it used to. People are healthier with mental health at an all time high since the great shift of 2008 when the majority of humanity reached the top of Maslow's hierarchy of needs.
I work 10hrs/wk at double the total pay what it took my parent a 60hr week to do. My commute is 10 minutes where his was 40. My healthcare is 3000x better and costs 1/100th. My house is three times bigger and 5x as energy efficient. All my friends live nearby and have similarly awesome lives with growing families across the board. Nobody has to rent anymore either.
/s
If robots can farm, mine, refine, and manufacture everything, what's left for people to do? Secretary? Replaced by AI. Customer service? AI. Fast food? AI. Bartender? AI. Doctor? AI. Lawyer? AI. Accountant? AI.
Programmer?
AI.
Let's say you have an expert blacksmith. He comes from a long line of blacksmiths, each a renowned expert in their trade. One day, he starts teaching an AI (which has human dexterity) how to do his job, as his forefathers taught their children. Eventually the AI matches his skills, maybe even surpasses. The AI will never die. It can copy its instances infinitely. We don't need human blacksmiths anymore.
When humans began to take over the earth, the single best survival trait among other organisms was being useful to humans. How useful are we to an intelligence that never gets tired, that can manage all the resources and means of production on earth?
What if humans are seen as a net drain on the earth and its resources? What if 1% of the population doesn't need the 99% for anything anymore?
We have no idea how a computer can be designed to come up with its own goals. We can teach it English, French, and Chinese, or get near perfect photo identification, but the computer cannot decide to learn to write a novel on its own. We could teach it to write novels, but there would still be other tasks it can't do. Even if a computer could self-direct 99% of known possible things in the universe, if there were one thing we could prove it would never learn, then it's not AI.
It may not even be possible and we could be stuck with humans around forever. Ug.
Most "basic" AI stuff like voice and image recognition was "a long way off" for like, forever. Every "advancement" in machine learning since the 50s ended up as a dud.
Then BAM! Voice recognition is here. Not perfect but definitely good enough. Image recognition is way, way, way better than it was 10 years ago (which was only marginally better than it was 10 years previous). The current trend of machine learning is accelerating progress, not a plateau.
Just think of the Geth from Mass Effect: their creators made them to perform basic tasks and share data. As their scope widened they were tweaked to handle more work. Eventually they became self-aware before they or even their creators really understood what was going on.
Besides, why use novels as a use case? Why not, I don't know, organized crime drama? They all follow the same arc anyway: someone does something evil or greedy, rivals descend into civil war, leader(s) assassinated, vague hope for better tomorrow.
It'll be like Winston Smith in 1984, writing the same stories over and over again and changing little more than names.
That was the general public's perception. Certain people in the field of ML knew such advances in application-specific domains such as voice and imagery would come at some point.
The difference between the application-specific solutions and general AI is top researchers don't have any real theories about general AI.
Even the OpenAI folks have toned down the rhetoric on their about page [1]. Their goal is no longer general AI, as Musk and Altman defined it at the outset. It's to contribute to existing developments. They write,
"In the short term, we're building on recent advances in AI research and working towards the next set of breakthroughs."
I can't explain my view any better in text. I've studied ML, ML researchers are nearest to AI, and they're all saying it isn't around the corner. They don't even have theories on how to make it truly general. Society will continue to see really great advances as a result of ML developments, but general AI isn't going to be one of them.
The domain is also limited, Siri can't play Go, and AlphaGo can't do voice recognition.
I believe the ML researchers if they are saying AI is not close.
Great point. This was part of it. There were other elements, all of which did not come together until a few years ago, yet their result has been expected since at least the 90s if not the 50s by some very visionary and optimistic people. The other elements were (1) having enough labeled data, (2) fast enough hardware, not just to solve the problems, but also to be able to test repeatedly with different initialization weights to identify good values for these parameters (3) good enough libraries for programming the hardware, enabling more developers to easily make use of GPUs for matrix operations, and distributed computing for problems that benefit from parallelization (4) various mathematical advances whose details I couldn't do justice, but things like rewriting the math so problems are more easily parallelizable helped a bunch, (5) sharing of research. ML researchers shared their advances openly via places like arxiv.org and this led to faster advances (6) the use of patents for defense rather than offense - I believe google led the way here
Now, and going forward, yes, it's believed that those with the most data have the biggest advantage. Even so, researchers are making advances and specializing in all kinds of areas of ML, like ones requiring little data, no data or lots of data, varying degrees of labeled data, different classes of algorithms, etc. The field is growing rapidly
That's the gist. I'm definitely missing a few things
Yann LeCun [1] is a good person to follow if you're interested in a top ML researcher's opinion about AI and the media's perception of it. He frequently posts articles with which he agrees or disagrees
For example, if this trend continues Apple will no longer get bad PR for supporting sweatshops. In the US there were also years (might still be true) where a company fully could write off capex purchases in the first year and not a multi-year depreciation schedule. Perhaps China is doing something similar.
A few years ago I made a vow to myself that I wouldn't work on software that puts people out of jobs. When I worked on job automation projects I would get very depressed watching people lose jobs because of me. I can't imagine the damage of something at this scale.
how can you tell? what is the heuristic here?
That's not to say that there aren't ethical issues here. My boss wants to automate customer service operations with customer service bots. So I'm not even safe there.
The problem happens when it happens so quickly that we can't re-allocate workforce or if it happens at such a scale that we can't find use for masses of low-eductation/low-skill workers.
i think the larger point in this entire discussion is that nobody is safe anywhere, both on the receiving and delivering end of the job-killing.
Arguably the whole point of software is to automate tasks that would have been done by humans. How do you avoid this when making useful software?
That person playing the video game is choosing NOT to go to the movies, or the sports game, or the bar.
Book authors, most likely.
> I think you could argue that software like Adobe Creative Suite and Microsoft Office are net positive on job growth.
Yea, I can see that. It's still replacing earlier jobs typesetting, editing, printing, distributing, etc etc.
> Software can empower people to do entirely new kinds of jobs.
I agree with this too, though I don't know if it's a positive thing. This computer revolution was supposed to increase quality of life, not employment. I'd rather have sustainable income for everyone rather than shitty jobs for everyone, and it seems like automation encourages the latter.
The early Jeffersonian model of the United States pictured Americans as living largely self-sufficient lives on small farms, making everything they need for themselves like food, furniture, barns for livestock.
Except it also included slaves.
Are computers that much different than slaves? Perhaps only in that they are not regarded as a form of life, but they still serve us as we desire, they can be bought, sold, or smashed with a baseball bat by disgruntled office workers.
If most transportation can be done with autonomous cars, if most goods can be produced by 3D printers which source material from mines and facilities owned by robots, if there is no need for human labor on a massive scale, then it might become more of a cost than a benefit for people to be working. Most of us do not like to work, but we have to work. I like my job, but I would prefer a world in which I did not need to do it and could do other things instead.
Well that problem is fixed now! These workers are no longer being taken advantage of because they don't work there anymore.
http://www.reuters.com/article/us-global-economy-employment-... http://2.bp.blogspot.com/-gworxUlu27o/VPmu_-G-ykI/AAAAAAAAik... https://upload.wikimedia.org/wikipedia/commons/9/9c/US_Unemp... etc...
That implies you can't do anything to make people more productive. If you produce software that makes people more productive, some companies will do more with the same people, but other companies will employ less people. Either way, though, you'll put some people out of jobs, and not all of those people will get the new jobs you create; even if you create more jobs than you destroy, you can't guarantee that everyone who had a job before will have one after. Improving productivity via some new tool has a similar effect to replacing low-skilled jobs with high-skilled jobs, just with a smaller scale and skill differential.
I don't think you can reasonably use "number of people put out of a job == 0" as a sole value metric, to the exclusion of all else. It's certainly worth including "number of people put out of a job" one of many factors you value, but not to the exclusion of all other values.
It's one of those thoughts that feels good, but really isn't practical.
Don't most software fall in this category though?
We detached this subthread from https://news.ycombinator.com/item?id=11770928 and marked it off-topic.
But you're SURE that these jobs will be created.... eventually?
The worker must get wages to buy the product. The company must sell product to get the money needed to pay workers their wages.
When you put robots in the mix you break this cycle. When workers/customers don't have money to be customers of employers, everybody loses. It doesn't matter if products are cheaper.
The tragedy of the commons is that corporations in seeking to become more efficient by eliminating employees are actually shooting themselves in the foot.
The change the OP suggests is drastic, but it is a valid methodology to maintain the money cycle. The only thing wrong with it, is that it is not fair. Why should someone who produces no economic output get paid?
But of course you may be right. In many cases, the endless greed of big corporations lead their owners to act without thinking the consequences.
That's the essential element that the rich folk concentrating capital and eliminating labor are forgetting. When the people have no bread, they do not eat cake. They start chopping off heads, without being very discriminating about whether they deserve to remain attached to their necks or not.