ROSS RANTS: ROBOT JOBS

https://en.wikipedia.org/wiki/AlphaGo_versus_Lee_Sedol

 

This was supposed to happen, like, ten years from now. In some sense it's no more like a human than the computer that beat Kasparov at chess twenty years ago, but it's not just the old game-beating algorithms with more processing power; the algorithms themselves are better. Until recently, a lot of AI skeptics were saying "beat top humans at Go and then I'll be impressed".

 

It comes from the same team that does DeepDream, which has mostly just been treated as a funny novelty that puts out weird acid-trip pictures, but is also a huge leap forward in terms of AI, and has only been improving in the months since it was first announced. Guys, I know that remaking non-dog images out of dogs isn't exactly high art, but it's also the kind of thing that was flatly impossible for computers to do until recently. DeepDream isn't a lost Photoshop filter, it's a major step towards human-like creative thought. The starving artists are going to be starving harder when they're also replaced by computers.

 

Do I think that AIs are going to become self-aware soon? Probably not, depending on how you define soon. But I do think that they're going to become more intelligent than most people anticipate, and as a direct consequence, more people's lives will be threatened than you assume here. What happens to the wealthiest businesspeople when the best computers can run large corporations more efficiently than they can? What happens to our economy in that case? What happens to our culture when popular media is literally designed by algorithms that optimize products like films and books to succeed? What will be the first nation where the human government decides to act simply as troubleshooters for a governing machine?

 

All of this could cause disaster very gradually, but a disaster could happen suddenly, too - not if a powerful AI attains self awareness, but rather if a powerful AI metaphorically forgets that it dropped a grenade and then stands in place.

The camera was changing the focus by itself, that almost made you scrap the episode, an episode about synthetic consciousness.

Ross, you're one optimistic guy, that's for sure.

 

 

What happens to the wealthiest businesspeople when the best computers can run large corporations more efficiently than they can? What happens to our economy in that case? What happens to our culture when popular media is literally designed by algorithms that optimize products like films and books to succeed? What will be the first nation where the human government decides to act simply as troubleshooters for a governing machine?

 

 

 

Here's my theory of what might happen over the course of several years. Keep in mind I do not claim to be an expert any regard this pure speculation. I think the need for people with more technical proficient backgrounds such as programming will increase dramatically. Businesses will be replacing white collar workers with robots and those robots will need to be maintained and have their instruction sets improved upon. So companies will be paying for maintenance and upgrades instead of paying actual workers. We might not have white collar workers anymore and those people will need to be retrained.

 

I agree with Ross that blue collar jobs will still be around. In fact I think we should be paying much closer attention to our infrastructure then we're currently giving it. How many dams, bridges and sewer systems globally are now potentially hazardous due to lack of maintenance?

Like the Dr.Sbaitso segment at the end.

History in the making is as subtle to us as the rotation of the Earth. So we're all gonna end up as poor an hungry proles... who cares? We'll get used to it within a generation.

When you said about not handing control over of the nuclear weapons to A.I. in the video, it reminded me of an article in the Times about what problems the millennium bug caused that weren't made public at the time.

 

The one item I remember of them all was the Russian (or I believe it said it was Russian) nuclear missiles computer supposedly reported a large missile attack on the country and advised the people in the control room to launch an immediate nuclear counter attack.

 

The general in charge decided it was a computer glitch and decided not to destroy the world.

Hey Ross, can you list what games the backgrounds you used are from. They look pretty cool.

Hey Ross, can you list what games the backgrounds you used are from. They look pretty cool.

He gave the background names and their creators' names in the credits. Ross always cares about the credits

Hey, Ross, when you were talking about what the future will look like -- give Player Piano a read if you haven't already. The ending is too hopeful, in my opinion, because Kurt Vonnegut didn't want to give up on the human spirit, but it gives a good picture of how even the best case scenario, in a situation of total automation like this, can be pretty messed up.

Ross, I must fervently disagree with something you said in this video. You advised us to look up the "real unemployment rate" and then showed a chart showing the Civilian Employment-Population Ratio beginning around 1990. I feel both of these are misleading in the extreme.

 

First, the "real unemployment rate" is a misnomer because it includes people 16 years old or older who are just not working for whatever reason. This includes retirees, students going back to school, the disabled and other reasons that might not have anything to do with the current employment situation. Which brings me to the second part. The chart you provided.

 

If you expand the "Civilian Employment-Population" chart out to its furthest extreme (beginning in 1948), you will see that the ratio was much lower back in the 1950s and 1960s. The reason for the increase starting in the early-1960s is the Baby Boomers entering the workforce. The drop-off starting around the year 2000, in the meanwhile, represents the beginning of the Baby Boomers retiring and leaving the workforce. It is my contention that, beyond the severe drop during the financial crisis, the numbers are still above normal and going back up, despite the retiring Baby Boomers. Maybe we're recovering until all the Baby Boomers retire and/or die out* in which case, we return to the levels we had in the 1950s.

 

Just wanted to inject some context into that.

 

* Sorry for being morbid there. My father is a Baby Boomer and he's been retired and I don't think you can really count him as being unemployed because he's getting a large enough pension from being a federal worker to keep him in money for the rest of his life, but that's not the case for others, but they're retiring as well so they might be in a similar situation, or they may not be. Just saying, I don't think you can count retirees as unemployed.

How many dams, bridges and sewer systems globally are now potentially hazardous due to lack of maintenance?

About 3.6 Trillion worth.

 

http://www.infrastructurereportcard.org/

 

 

Ross, I must fervently disagree with something you said in this video. You advised us to look up the "real unemployment rate" and then showed a chart showing the Civilian Employment-Population Ratio beginning around 1990. I feel both of these are misleading in the extreme.

 

First, the "real unemployment rate" is a misnomer because it includes people 16 years old or older who are just not working for whatever reason. This includes retirees, students going back to school, the disabled and other reasons that might not have anything to do with the current employment situation. Which brings me to the second part. The chart you provided.

 

If you expand the "Civilian Employment-Population" chart out to its furthest extreme (beginning in 1948), you will see that the ratio was much lower back in the 1950s and 1960s. The reason for the increase starting in the early-1960s is the Baby Boomers entering the workforce. The drop-off starting around the year 2000, in the meanwhile, represents the beginning of the Baby Boomers retiring and leaving the workforce. It is my contention that, beyond the severe drop during the financial crisis, the numbers are still above normal and going back up, despite the retiring Baby Boomers. Maybe we're recovering until all the Baby Boomers retire and/or die out* in which case, we return to the levels we had in the 1950s.

 

Just wanted to inject some context into that.

 

* Sorry for being morbid there. My father is a Baby Boomer and he's been retired and I don't think you can really count him as being unemployed because he's getting a large enough pension from being a federal worker to keep him in money for the rest of his life, but that's not the case for others, but they're retiring as well so they might be in a similar situation, or they may not be. Just saying, I don't think you can count retirees as unemployed.

I received so many comments regarding this video, I may make a short follow-up video next month covering all the stuff needing clarification or unaddressed points.

 

Sorry about the real unemployment thing, the main thing I was after is we normally don't count the following:

-People underemployed (who want to work full time, but can only get part time)

-People who are looking, but haven't been hired in over 6 months

-Discouraged workers (who want to work, but have essentially given up)

 

I wasn't thinking of people age 16, nor retirees.

 

 

EDIT: Feel free to help me out with some data covering a more realistic scenario of the employment situation based on that. The point is the official rate isn't very representative.

I may respond to the rest of this later, but right now I just really want to point this out, because this misconception bugs the heck out of me whenever it gets brought up:

 

2:25 in the video:

 

"Except whoops, we're getting paid about the same as we were 40 years ago, despite all this extra prosperity!"

 

This is false. I mean, technically it's true... but you're (deliberately?) excluding a rather important piece of the pie with your focus on wages. That chart just uses (hourly, judging by the fact that it's nearly identical to this similarly dishonest piece) wages, which ignores employer provided benefits, and is thus intentionally misleading. It also uses two different price deflators for productivity and compensation, further increasing the supposed difference (the CPI overstates inflation in the overall economy). Also, "major sectors" and "goods-producing workers" excludes 15-20% of the workforce, just like the chart it's a clone of ("This explanation is further suggested by Figure 3 data showing that in 1998, 14.6 percent of the workforce was in some category of supervisor or manager, and in 2008, the number was nearly the same at 14.6 percent"). Once you adjust for these factors, real compensation looks more like this. Federal Reserve economists have come to the same conclusion.

 

Here's a pretty good explanation on the truth behind these numbers by OECD.

The problem isn't that rich people are eating up money, the problem is much of the growth in compensation has been eaten up by non-wage benefits like pensions and healthcare for retired workers and by supervisory personnel. Over the past 30 years there has been a meteoric rise in the number of managers due to the requirements of the new economy. Managers now make up a bigger portion of the US labor force than ever, which is why that chart you used looks so disconnected from the reality.

 

The report defines two types of decoupling, each of which tells a very different story. For “net decoupling”, defined as the relationship between productivity and average hourly compensation, compensation for ordinary workers has not fallen behind productivity, just as economic theory would predict. The only explanation for the two diverging would be if labour’s share of national income fell versus the capital share. While others have suggested that just such a decline has taken place, Van Reenen’s results suggest it has not.

 

In the second type of decoupling— “gross decoupling”—the measure of productivity used is GDP per hour worked in the UK economy. But this time worker benefits are represented by median hourly earnings. Using this measure the picture looks very different. Median pay has diverged markedly from labour productivity in the UK in the last twenty years. In other words, the pay of ordinary working people has not kept pace with the average value of output that workers produce.

...

Net decoupling looks at total compensation rather than just wages. This means it includes things like employer pension contributions and employer National Insurance Contributions. As such, there’s a reasonable argument that it’s a better measure of the complete rewards derived by workers. And, again, recent years have seen a widening gap between these two measures. As non-wage aspects of compensation have grown significantly, compensation has grown faster than wages.

the main thing I was after is we normally don't count the following:

-People underemployed (who want to work full time, but can only get part time)

-People who are looking, but haven't been hired in over 6 months

-Discouraged workers (who want to work, but have essentially given up)

 

I wasn't thinking of people age 16, nor retirees.

 

 

EDIT: Feel free to help me out with some data covering a more realistic scenario of the employment scenario based on that.

 

The thing is, the media don't really report the numbers you're looking for but the government does. It's just oftentimes buried under the official numbers.

 

There are six rates published by the US government:

As of February 2016:

U-1: percentage of the labor force unemployed 15 weeks or longer

U-2: percentage of labor force who lost jobs or completed temp work

U-3: official unemployment rate that everyone knows

U-4: U-3 rate + discouraged workers, persons who have given up looking for work

U-5: U-4 rate + marginally attached workers, persons who have been unemployed for so long that they're no longer counted in the U-3 rate, but are still looking for work or have stopped looking for work but still want work

U-6: U-5 rate + people employed part-time who would rather have full time

 

As of February 2016:

U-1: 2.1% (down from high of 5.9% in April 2010)

U-2: 2.4% (down from high of 6.3% in December 2009)

U-3: 4.9% (down from high of 10% in October 2009)

U-4: 5.3% (down from high of 10.5% in February 2010)

U-5: 6.0% (down from high of 11.3% in April 2010)

U-6: 9.7% (down from high of 17.1% in April 2010)

 

You can find these rates listed here.

 

Generally, when you hear people are referencing the "real unemployment rate", they're simply taking the number of people "not in the labor force" (that is, not working at all for whatever reason), and mixing that with the U-6 number to get the extremely distorted figure of 40% or more unemployed.

 

Back to the question at the end of the video: What sci-fi movie or whatever is the future going to be most like? Depends on how far into the future we're talking, but if we're talking about in the next 100 years, I'm thinking some sort of Frankenstein's monster mixture of "Demolition Man" and "Star Trek". I'm just not sure if we're going to have the good parts of those or the bad parts...

I think if robots continue to take jobs from workers governments will have to start giving out universal income. We already have things like Centrelink in Australia, but that's not really enough to actually live on. Who knows, maybe we'll reach utopia one day.

A few thoughts on the subject:

 

1) seeing the minimal wage craze in USA makes me feel it only makes the situation worse. If you keep the minimal wage low and low social guarantees by the state (meaning you have to work to survive, the state won't give you much money), the robots will be too expensive; but if you keep rising the minimal wage on low skilled assembly line type of work, it makes a lot easier to replace the human workforce. Wage vs automation also can be looked at through immigration - countries like UK, Ireland are benefiting from Eastern European workers, who are willing to work for lower wages than locals. This also influences whether companies use more automation.

 

2) qualification - in my opinion low skilled workers always have had problems with finding job over longer period when the times changes. E.g. if you worked 20-30 years ago in a manufacturing plant, the chances are you spent last 10-15 years without a stable job for very long and you tried a lot of different low skilled jobs. I would urge younger people to develop skills and study things that matter. Of course its hard to tell what will be needed 20 years from now on, but my point being that studying sociology, politics etc. might not be the best idea. Yes, you can make your own services, business with that experience but how many can do that? 1 in 100? Study something that requires extensive learning and knowledge - physics, chemistry, biology, pharmacy, statistics (I mean the real deal, not the watered down versions) etc. Also many things can be learnt for free on the internet. Its all about you actually wanting to do something. It might sound cruel, but if the height of your abilities is to work at Mcdonalds and you cannot survive by yourself, its your own fault.

 

3) I do not fully agree that this is so much different from before. Just look at how many people were working in industry and agriculture 30 years ago and how many now (I mean per production). Productivity gains with complex assembly lines that require maybe a few operators have replaced hundreds of workers, are through the roof. I'm not talking about the first assembly lines from back in the day.

The thing is, the media don't really report the numbers you're looking for but the government does. It's just oftentimes buried under the official numbers.

It could be I'm wrong then on that point. Wouldn't the articles about the new jobs we're getting paying less still hold up however?

 

I may respond to the rest of this later, but right now I just really want to point this out, because this misconception bugs the heck out of me whenever it gets brought up:

I'm kind of tired and I sense this is something that could go on for pages with no real resolution, so I'll try and keep it simple:

 

-I'm not trying to intentionally misrepresent anything (unless it's obvious and funny), I'm just working with the best data I could find. I'm not an expert on this.

-Even if your graph is accurate, I think there are MANY other indicators that a lot of wealth is flowing towards the top, way too many for a video not focused on that. Here's a sample one:

http://blogs.wsj.com/economics/2013/09/10/some-95-of-2009-2012-income-gains-went-to-wealthiest-1/

-There's a direct rebuttal to the data you're mentioning:

http://www.epi.org/blog/compensationproductivity-link-broken-vast/

 

It might sound cruel, but if the height of your abilities is to work at Mcdonalds and you cannot survive by yourself, its your own fault.

I think this is the point that stands out for me. I operate under the assumption that anyone willing to work who has their survival threatened by not being able to (either through circumstance or ability) is exposing a problem in our whole system, not the individual.

It might sound cruel, but if the height of your abilities is to work at Mcdonalds and you cannot survive by yourself, its your own fault.

 

It's crazy to me that people actually think like this, anybody should be able to survive in a first world country even while unemployed. It shouldn't be a GOOD life, but the bare minimum to not die out on the streets. Low wages and being homeless is a self-perpetuating cycle that's pretty hard to break out of without outside help.

It might sound cruel, but if the height of your abilities is to work at Mcdonalds and you cannot survive by yourself, its your own fault.

 

It's crazy to me that people actually think like this, anybody should be able to survive in a first world country even while unemployed. It shouldn't be a GOOD life, but the bare minimum to not die out on the streets. Low wages and being homeless is a self-perpetuating cycle that's pretty hard to break out of without outside help.

Well, from what we can see (mostly in Europe) its never ending cycle of more and more inefficient social programmes that leads to more and more taxation from those who earn their living.

 

I'm not one of those who says that we need minimal government and everything should be market based. We still need regulation for environment, to reduce monopolies, you need your rights of fair treatment from everybody (the law system), we need public services and protection such as police, firefighthers, education, some basic health care etc. But I am saying that the governments are too big, social programmes often simply do not work. The solution usually is - we simply didn't had enough money and enough social programmes. WE NEED MORE!

 

And than taxes are raised, raising taxes means less purchasing power, more taxes means harder to start a business (including small ones, like hobbies) etc. Its all for what? To try to encourage not working and self degradation as well as economic migrants from Africa, Asia?

 

I also did not say that these people should die on the streets, we need more opportunities for people to work and increase their skills, not social benefits. But sadly its usually not like that. We get more social benefits, more taxation, inefficient social programmes.

 

"Bureaucracy is expanding to meet the needs of the expanding bureaucracy."

An odd topic for a rant... Sorry Ross, but I'm afraid the european leftist propaganda got into your brains. The idea that the government SHOULD provide people with jobs, or that it SHOULD pay them just for being people is dubious. Government does not have it's own money, only taxes. If government is paying someone for not having any job, that money goes right out of the wallet of the people who do have a job. Even the most advanced robots will not have an income of their own, and will not pay taxes, people will. I don't think that the strategy of encouraging the unemployed people at the expense of employed is good in the long term.

 

Also about robots being cheap and working for free... There is a reason why in the chinese factories it's people who reel up the magnetic coils and not the robots. And it's not because the chinese government ordered it so. Complex robots are complex, and even if you can replace the coil reelers with robots, and only have one educated engineer per ten robots, this educated engineer would still be more expensive. Not to tell that robots need to be bought, and need spare parts, and the precise machinery operates only in very VERY clean environment. Again, there is a reason why on WV factories all the personnel wear suits, and the production workshops are as clean as a surgery. Try to achieve this kind of cleanness in McDonalds!

 

So, my points are:

1. I don't think there will be less job opportunities for educated people. If anything, the number of programmers and engineers will increase. In germany alone there is already a starvation for engineers of about 100k people a year.

2. I don't think the low-tier workers are at risk too. Their advantage is versatility. Rearrange boxes, swipe the floor and help the old lady to find the grocery. Each task is easily done by a robot, but to be able to do them all and to understand what exactly must be done now is a task the consumer robots will not soon learn to do.

3. About 40% GDP is generated in the services, not products. Waiters, hairdressers, actors and teachers are all humans, and will remain humans. Butlers will not be replaced by robots not because the robot can't do a butler job, but because being a human IS a butler job.

4. And finally, the responsibility. Even the most advanced robots is not responsible for it's actions. No company will ever sell a robot without a clear statement that it requires a human supervision, and that this company is not liable for any damage. The robots will always be just tools.

 

As for the robots controlling nuclear missiles, read about a "dead hand" system when you think you're not scared enough As a person who was born in USSR in a family of two physicist working for military I, of course, know something even more scary than the information that is publicly available, but I think even the publicly available information should be enough for starters