The huge economic issue that no one in Washington is talking about:
Driverless trucks delivering goods to fully automated warehouses and loading docks. Drones delivering everything from pizza to furniture. Offices will become almost fully automated as work is farmed out to smart machines. There’s even speculation that AI could take the place of reporters and editors, writing copy with more speed and less bias than humans.
Most of these innovations are not far off. What’s worse, our schools are stuck in a time warp, teaching kids as if it was the 1970s, sending them to college where they major in English Lit or Environmental Management. How many of these young people would be better off going to a trade school and learning a valuable skill that would be useful in the new economy?
What’s needed is a revolution. Not rage against the machines, but a clear-eyed recognition in society from top to bottom that we can’t go back. The days when you could graduate from high school and go to work for 40 years in the local plant, earning a good middle-class wage and being able to buy into the American dream, are gone forever. Donald Trump can’t bring them back. The Democrats can’t bring them back. The unions can’t bring them back.
Nope.
The article does *not* address the 2 innovations that would allow people to adapt to the on-coming changes of AI.
1.) An educational system that will use long-term teams as their means of teaching children to look for *work*, as teams, instead of looking for jobs in a hierarchy. At present the schools suppress such teams, labeling this evolved social structure as “gangs”.
2.) The Use of these teams to teach the adoption by students of AI technology before they ever get to the markets for *work*. The combination of human consciousness with the skill sets of AIs will be far more productive in work than either will be alone.
Umm, no. Don’t claim that gangs of violent, criminal thugs are “teams” “looking for work”. That is absurd. Your idea may have merit, but you should not confuse it with the social depravity of urban gang culture.
If we were to allow productivity to soar, and if the central banks would permit deflation, then it is conceivable that a person need only work 10 to 20 hours a week to pay for the basics. After that, if one has an entrepreneurial bent, more money can be earned from a side business.
I think this is where society is going. We just need to get rid of the Keynsians who think deflation is a bad idea.
Deflation does not change the actual value of things in terms of work needed to buy them, neither does inflation.
You’re correct in the sense that deflation is defined by the contraction of the money supply, but it often leads to falling prices. Just look at the amount of money and credit thrown into the system to raise housing prices, health care and education. The two often work in tandem.
Also note that a fiat currency must automatically lead to inflation. Supposedly this has been kept low for years, but since it has a cumulative effect we are all suffering the consequences.
I should also add that you’re correct as labor and compensation are relative. If prices fall the cost of labor will fall. But that is only a one to one relationship. It does not include factors like productivity and innovation–the factors that I think are in play here. If it were only a relative scale then our standard of living would never increase.
I’ve just read “Rise of the Robots: Technology and the Threat of a Jobless Future by Martin Ford”
It’s not just driverless trucks, AI is going everywhere (with the exception of Politicians). See the recent news about wallstreet replacing 800 traders with 200 coders and also add google’s efforts for AI to write AI, even coding is going to be under threat.
Programs writing programs was a goal since the 70s. From what I see, code is getting worse. I suspect programming may be the only job left in the future.
The invention of new languages seems to only be job security for obsolete programmers on legacy systems.
Don’t forget that ‘AI’ isn’t even really code. Most of these things use neural networks, which take a bunch of inputs and generate a bunch of outputs through mechanisms that no-one ever really programmed, just trained. Give them the wrong kind of inputs, and no-one really knows what will happen.
What could possibly go wrong?
Well, humans also aren’t guaranteed to produce correct code.
A neural net-based AI that produces proofs of program correctness (for the code it writes) wouldn’t have to be perfect, since its output would be self-verifying.
More worrisome would be neural AIs applied to real world interaction problems where such intrinsic verification isn’t possible.
I wonder if mathematics is going to turn out to be surprisingly automate-able. Theorem provers have been worked on for decades; they may go through a transition like chess playing programs did.
“Give them the wrong kind of inputs, and no-one really knows what will happen.”
Haven’t you just described the entire western education system? lol
Kinda Rhetorical but is the code really getting worse or is it just there a lot more code and more programmers now. And it easier to recognize bad code than well written code. Instead of having the top 1% writing code you got 10% of people. And more so the code is being built on top of code, so a bad piece of code gets written in to a library and then that library gets reused in every program that related to it.
From personal experience, there are a) more coders and also b) those coders are ‘cheaper’. It’s pretty much a race to the bottom by most companies where IT expenditure is managed by Financials and not output/quality.
The number of times I’ve had a discussion with a ‘coder’ whose view is the SQA person (assuming there is one) is responsible for testing, not them. Also, the bully the SQA into accepting the code working as it was built by the coder and not compared to the specification/requirements.
As a programmer entering job market in late 1970s, I noticed that goal while reading ACM publications (Quote Quad, TOPLAS). I worried too, but it did not come to pass. AI got better and better but did not put any pressure on my peers and I.
The pressure came from a direction completely unexpected by me and the media: phenomenal increase not in AI but in communications. With deployment of solid Internet connections everywhere, competition came not from robots/AI but other programmers in other countries. Competition is a good thing, and we’re all sharper because of it.
I would also note that tons of new languages were being developed in ’79, just as they still are today. But I reject the idea they’re created just for “job security”. I’m not a language designer, but people with all kinds of backgrounds develop new languages, and for all kinds of purposes, just as before.
We’ve been through this before. We are on the cusp of a second industrial revolution, every bit as wondrous and disruptive as the first. Old jobs, whole industries, will be destroyed, and replaced with new ones providing far greater prosperity than can even be imagined.
No, this is the end of the Industrial Revolution. We’re heading back toward the kind of self-sufficient lifestyle our ancestors had, but at a much higher level of technology. Who’s going to give a crap about ‘jobs’ if they can fly to an asteroid, feed it into a 3D printer, and build a habitat to live in?
The problematic part is that, without Industry, there’s little need for big cities or their populations, which grew only to provide the amount of manpower required to operate industrial manufacturing. And with local manufacturing, there’ll be few resources in those cities that can be used to make the things those people need.
The world of 2050 will be very, very different from today.
there’s little need for big cities or their populations
Some people like to live in cities, many for reasons other than industry.
In the book there are example of companies with AI already replacing reporters and journalists, especially in sports and other ‘fixed arena’ type reporting.
But the AI’s are getting better.
I’ve seen some of what I think must be examples of articles written by AIs. The spelling is much better. Most blogs and newspapers nowadays are under more time pressure and money pressure than ever. So there just isn’t time or money for professional proofreaders and editors to review the copy, and it suffers.
On the flip side, the AI-written articles, while better in terms of spelling and grammar, have a bland feeling to them. They’re like baby food. This is just my own perception – anyone else notice something similar?
So what do humans do when all jobs are done by robots?
What will robots do when they get tired of supporting all those jobless humans?
Yep, that’s one of the main crux of the issue.
In the book it mentions a fable of Henry Ford and a Union Boss walking around a new automated factory. Ford to the Boss, “How are you going to get the robots to pay their dues?” Boss to Ford “How are you going to get the robots to buy you cars?”
Robot reporters will have exactly the same bias . . . as their programmers.
Nope, bias will be as per the company’s boss who pays the bills. If it doesn’t produce the ‘right’ results, then he/she will buy another one that does give the ‘right’ results.
Butlerian Jihad time!
https://www.amazon.com/Player-Piano-Novel-Kurt-Vonnegut/dp/0385333781/ref=sr_1_1?s=books&ie=UTF8&qid=1487093748&sr=1-1&keywords=player+piano
First published in 1951. Vonnegut imagined a world in which the machines were run by punched scrolls, as in a player piano, and virtually nobody had a job.
Meh. I expect things will simply evolve in ways we do not yet imagine.
“I expect things will simply evolve in ways we do not yet imagine.”
I think you are right. It’s fun to conjecture but no one knows how the new economy will shake out. people will still need money to buy things no matter if they are lots cheaper than they are now.
How will they make that money? What will the new jobs be?
Buggy whip manufacturers couldn’t answer that question before cars appeared.
Neither can we.
That’s a sell out. We may not be accurate in making a prediction, but that doesn’t mean the attempt isn’t worth while. Looking ahead down the road is how you avoid the dangers.
Human creativity has always been an important source of new wealth. The other is industrial capability. Robots are advancing on both fronts.
We see the middle class disappearing and computers may accelerate the trend. A mass die off is very likely.
I like the idea of a high tech subsistence lifestyle. There are a lot of things that need to be in place for it to be possible both materially and educationally. IMO, high tech subsistence living requires access to:
There would have to be a monetary system and economy to give people the ability to get access to these things.
Look at subsistence living today. It wouldn’t be possible without items made in a factory someplace and a lot of labor goes into gathering food or maintaining property. Having robots do those chores means you would still have to obtain a robot somehow. Manufacturing your own goods means you would have to be good at design and spend a lot of time on it.
There is a tradeoff in time but you would still have to do a lot of work. How many people would rather tend a garden than design things in autocad? And some items would still be more efficiently produced in factories.