Subscribe now on: iTunes | Google Play | Stitcher | Soundcloud | RSS | or search "Ashes Ashes" on your favorite podcast app.
(Please pardon this awful machine transcription until we can manually edit this text)
I’m David Torcivia.
I'm Daniel Forkner
And this is a show about systemic issues. Collapse of the environment and if we're unlucky the end of the world.
[0:12] But if we learn from all this maybe we can stop that the world might be broken but it doesn't have to be. Google
[1:05] That was a sound of millions and millions of call center employees around the world becoming obsolete.
[1:14] Yes Daniel that's right this was actually not a human calling this restaurant but I demo of the latest technology Google just presented at their AI Conference of a robot makes a call. Understand what somebody says replies back to them in ways that make sense and it takes actions based on the information it receives. There's no human element here except the person they're calling in it considerably one day it'll be just a bot calling another bot. And this is here now this technology is already being deployed by Google. They're pushing it out for different products right now but he will very soon hit the commercial market and find itself to all of our phones and devices within a just a matter of years.
[1:52] When you said that one day we'll have Bots calling Bots and maybe think I don't even know that the restaurant owner was a human and are we looking at a world one day David Ware, millions and millions of bone hours will be spent Everyday by robots is calling each other thinking they're humans but really they're not.
[2:10] These are interesting questions.
[2:11] And what does make for a crate entertainment.
[2:15] I don't know but it is going to make for a great episode because this week we're discussing automation robots and the end of work as we know it.
[2:22] Why nobody cover some dark topics on the show but that sounds like a good thing David.
[2:27] It does at first glance but the fact of the matter is we have a system that's broken, economic cultural political system that enables something that should be as great as automation of a world where we don't need to work release work much less. And turns it into what may be when the most apocalyptic topic so we ever discussed on the show and something honestly that the world is not prepared for and is coming much sooner than anyone realizes.
[2:51] By 2030 73 million jobs could be automated in the United States alone.
[2:59] That's 50% of jobs in the US loses get that out of the way and if that doesn't drive that number in enough this more than double the unemployment rate at the very peak of the Great Depression.
[3:11] And by 2030 globally 800 million jobs could be lost to automation.
[3:17] Please give me a lot of numbers and statistics in the show what does the two figures that we're going to really come back to you and then you just need to remember so depending on the nation 30 to 50% of all jobs will be gone within 15 years.
[3:30] David when we were preparing for the show at first I thought that it might be a little strange to follow up Farmland access with a show about robots in the Sheen's. I mean it seems like a very a big gap between topics right but the more I thought about it the more I realized that this is actually the perfect. Sequence this is the perfect show to follow Farmland access.
[3:55] Okay Daniel I'll bite what was that the case.
[3:59] But you just said we have a broken system right and I think it's very simple it comes down to this first take away all the land okay that's the Farmland access show next charge people just to exist. Right in the finals. [4:23] Still own all the land.
[4:26] Well that sounds like the perfect recipe for disaster apocalypse or if you're the more optimistic type maybe Revolution.
[4:34] And that is the situation we are now on the Tipping Point of today so.
[4:40] Yeah it was good start.
[4:49] Jordan Airy challenges even for or maybe especially for some of the tasks we would consider to be the most menial basic and low skill task you can think of. And I remember watching videos when I was a kid of factories in the way assembly lines and machines work to produce anything from a toothbrush to a car. And it's fascinated me to see how these products were put together with the help of huge industrial machine. What is complex as those machines were they still relied on very predictable inputs along predetermined paths and humans had to stay close to monitor for errors conduct maintenance. Shift and sort Parts by hand or any number of complementary tax and a few years ago I was actually in Berlin. [6:06] Deuce A single standard motorcycle instead a dealer calls them up says hey I need 20 for motorcycles I want this color scheme and D specs. And it's only then that the factory starts making them and 2 to 6 weeks later those motorcycles are delivered anywhere in the world. And that's made possible by automation.
[6:27] That's true Daniel but at the end of the day it's still an assembly line. And humans are there at multiple stations along the way putting all these parts together they're doing engine work they're doing the fine-tuned customization and every single process requires a minimum of two people for inspection and quality control. Another words that humans are there to fill in the gaps. Obstacles To Total Manual Automation [6:48] Veritas to come naturally to humans that right now at least seem impossible for machines we look at a table with a bunch of clutter on it and it's very easy for us to recognize a pen or pencil automatic. Or maybe a coffee cup or flower face. Even physically pick these objects up rearrange them how we want pick them up as high put them down as hard we are very good at this is easy. At least three obstacles have prevented robots for do the same one the physical dexterity required for manipulating objects of different sizes shapes and sensitivity. To the visual perception that's actually just even having any sort of vision and then of course 3 the computation powers and software to recognize what it's actually seeing. Mathews 3D spaces and then ultimately make decisions of what to do.
[7:37] Which is why it's been easy for machines to send a 500-pound motorcycle down a line after drilling holes in the same place as every other motorcycle. But unloading a truck full of random boxes and then stacking them up against the wall well it's been Out Of Reach for the most part but that's all changing.
[7:57] Technology has advanced to the point where robots machines are now poised to invade the final frontier of manual labor. And it's because of advantage that's simultaneously address those three obstacles. Huge Leaps and Bounds were made in visual perception that was largely driven by the gaming industry actually you remember Xbox Kinect now the company that make that, as taking a technology sold at the Apple and then that's what you find in your latest iPhone that scans your face, Miss uses infrared lighting and that's through the spaces and four years Hobbies have reverse-engineered this device and found innovative ways to buy them two robots the 3D scanning. And while in the past it would have been very expensive to program a single robot for complex function. Is Barry has been all but surmounted through Innovative programming open source libraries and machine learning. [8:45] Individual robots in a factory setting no longer need expensive processing power inside their chassis but can instead be run by Central hubs add all the Computing and he's Warehouse size server rooms and connect to the robots wirelessly. Something that ass whooped discuss that's particularly large implications for the service sector in terms of manufacturing these factors combined with very precise robots. So there's a company called rethink Robotics and they have a robot which they call Baxter does to Precision arms that any worker comp program by physically moving in the pattern needed. That is it learnt once a single boxers trade Its Behavior can be transferred to any number of other Baxter's. And these advances are being spit up even more with developments in machine learning and object recognition and computer vision and many other Technologies which are finally beginning to reach maturity and spread out and become developed and ubiquitous. Of course it's not all roses yet as we seen with Elon Musk trouble at his factories trying to fully automate the Tesla production which he even has come out his admitting that he overestimated the effectiveness of automation right now and in fact. Is bringing in a lot more human workers to complement the robots that they brought in. But factory conditions production line these are very predictable simple ways to apply automation to and it's a small sector of the economy as a whole. When you go back to the beginning this episode we were talking about 70 million jobs in the US gone in little over a decade there's nothing bench factory production in the u.s. what are all these jobs being replaced.
[10:14] But David before we talk about the specific jobs at risk here. Automation Economics [10:41] Very soon. The consulting company Mackenzie found that if you wanted to fully automate jobs around the world today with current improving technology you could replace just 5% of jobs. But you don't need full animation for labor to be impacted today for most jobs about 60% of them at least 30% of the activities within the occupation could be automated. And that means that if current technology were deployed on a large scale in every sector that it could. Up to 1.2 million employees around the world would be affected.
[11:17] What is that mean practically it doesn't mean that we can be doing less work individually instead one person is going to be doing the jobs of several people. And this is where we're going to see automation playing out in the first place one person replacing several workers but still a person they're doing that work just more efficiently for whatever business owners in charge.
[11:37] And all this is going to put pressure on wages. Wages [11:50] Couple decades in fact the average American worker who is in a production or anion. [12:01] Declining real wages of about 13% for 40 years between 1973 and 2013.
[12:10] All well being more and more productive for those people paying the wages.
[12:15] And similarly that consulting company Mackenzie found that end quote advanced economies on quote around the world 66% of household wages stayed flat or spell. 2005 to 2014 incomes for workers have fallen as the share of national income around the world has shifted away from labor and towards owners of capital. And like you said David this is relevant to the cusp of our current situation because these falling wages have occurred during a time of soaring productivity. Between 1948 and 2011 productivity increased by over 250% in the us alone while wage is Pete. In the early 70s. In the past wages increased alongside productivity as tools and Technology were introduced that could increase the output of an individual's labor the value of that individuals work Road. An assembly line after all might make workers more efficient raising productivity but you can't have a functioning assembly-line without those workers. What does technology has gotten better at not just leveraging worker skills but replacing the work that people do, we've seen workers get less and less of the share of national income and instead it's going to businesses and owners of capital. Since the 70s the share of national income going to labor has fallen by 7%. But that drop in income that is going to employees and workers what it includes the salaries and bonuses that have been rising tremendously for our CEOs and other high pay professionals like bankers. [13:58] Which means that the decline and wages felt by the majority of workers is much more pronounced.
[14:21] To put it simply we work more than ever we make those that we work for more money than ever and more paid less than it anytime in the past few decades. Job Creation [14:31] Let me guess we're getting distracted for a bit talking about this I mean the topic of this show is automation right. If you ask any Economist they're going to tell you well automation well it creates jobs.
[14:42] That's the standard argument David when it comes to new technology.
[14:46] And in fact they are so convinced about this they have their own little thing the Luddite fallacy our friends to luddites which we discussed in the past and they say that anybody who thinks that automation is going to hurt jobs going to generally lower, unemployment cause mass unemployment well it's just because they don't understand economics, you don't understand supply and demand the fact that as we create more automation while the freezer people they can go follow through with whatever other jobs it creates jobs.
[15:14] David not just that I freeze them up but like you said it creates new jobs and it's actually a pretty compelling illustration so let me give you an example. Let's say that technology introduces something new like in agriculture we invent a big machine like a tractor that one person can drive. But oh no it replaces the work of 10 laborers well as the illustration goes those 10 jobs that are lost or made up by the fact. The factory that produces that tractor will it needs workers. And now new businesses need to be created to supply workers for repairing and servicing tractors and now we need sales people to sell the tractors and accountants and lawyers to help manage the industry of tractor making and selling. Oh and we probably export some of those tractors to other countries and all this creates demand for more skilled. [16:13] Something similar at least that's the argument and you know what maybe there was a lot of Truth to that in the past.
[16:19] Inasmuch as Economist like to cling to this argument I think they're missing something and that's the fact that this time well it's different. [16:27] Daniel new example here with this tractor. This replacing the laborers but we have people selling it we have people who need to service it whatever what let's look at this from an advanced automation perspective. That factory that assembly line well it's mostly robots with maybe one or two people Staffing what will be hundreds of tractors replacing hundreds and field laborer jobs. The people who are selling these tractors well as we just learn to be newest episode those sales people might themselves be Bots talking to other Bots organizing shipments that are brought together by Bots for Bots. Delivered by autonomous driving vehicles and ultimately and this exists now these tractors wants to get there there's anybody driving it it's navigated in tiredly autonomously and this whole chain the supply chain that used to create many jobs. What was jobs are now replaced by robots. Programs and the number of jobs lost greatly surpasses those that created by the new Industries but once again we're getting ahead of ourselves.
[17:25] Will David you're also talking about tractors which maybe not everyone can relate to so here's a more relatable example Blockbuster. This company used to provide brick-and-mortar stores for people to rent movies. In Chicago Blockbuster used to have dozens of locations each with around 7 employees. And Blockbuster had 60,000 employees nationally well as everyone knows Blockbuster was crushed by new services like Netflix and Redbox which I'm not complaining about. But now if we look at Chicago there are a little under 200 Redbox locations in the area and how many people do you think it takes to manage all of these David. 200 red boxes that people can they have to keep them restocked people are renting movies left and right everyday oh you know the movie just came out we going to restock it how many people do you think it takes.
[18:18] Does anybody even use Redbox anymore.
[18:21] That's the problem with technology these days David a change of self.
[18:24] I don't know Daniel I'm going to throw out like a ridiculously low number anyway 12 people a dozen people for all the Chicago.
[18:33] Take that number David and subtract or.
[18:36] Let me get my calculator app.
[18:38] The answer is eight eight people manage all the red boxes for the Chicago area. And like this tractor example that we were using we should expect peripheral jobs open up someone has to maintain these machines right and repair them when they break. Well all that's taken care of the same eight people they do everything they pick up movies from the warehouse they were stocked the boxes and when they malfunction. They can repair it from the comfort of their homes using internal Motors and arms within the RedBox machine that can be controlled over the internet.
[19:12] I'm going to go around unplugging red boxes from the internet I'm a job creator.
[19:17] It's a why is this time different David A lot of it has to do with this information technology that makes it possible to repair a machine over the Internet that has now become so integrated in every business but I think. [19:37] Change and we just aren't seeing the connection I mean a great example that people use is the invention of the ATM machine a lot of people us.
[19:47] It's ATM.
[19:48] That's what I said I said ATM machine machine.
[19:52] Yeah no you said you just said ATM machine machine which automated teller machine machine machine.
[19:58] Well the point is when this technology was invented people figured it would be the end of the bank teller but actually it created more jobs because it allowed Banks to open new branches and these bank teller now focus on other things I customer service.
[20:13] That's right Daniel once again and we like to look at the big picture but maybe we're missing it this time so instead we could look at the actual numbers. So forget trying to make connections between one job loss and another created let's just look Instead at the total job creation for the country as a whole. Now when looking at job creation by decade the US Bureau of Labor Statistics shows that the United States has been adding less and less jobs every 10 years. Increase the number of new jobs by over 30% in the 1960s and then figure fell down to just 20% during the 1990s. Obviously job creation was horrible in the late-2000s because of the financial crisis or right before the crisis occurred in 2007 when the economy was booming when the biggest bull runs the market has ever seen will new job create an increased by only 5.8%, at the crisis had never happened we would have still experience a dismal 8% increase in two jobs as it turns out when we look at the numbers now we created zero. Because we need to add up to a hundred fifty thousand jobs each month just to keep up with population growth. First decade of the 21st century resulted in a deficit of over 9 million jobs.
[21:25] So proud. [21:32] Why is this happening in a time of Rapid technological innovation and automation that should be creating new jobs as the argument goes well a lot of it has to do with information technology which is now embedded into the very infrastructure. Every business now Taps into and relies on. Which means that new innovations that take place in very labor-intensive Industries are aimed leveraging. Information technology to reduce Reliance on human labor, well at the same time any new business that is created starch from a baseline that's tapped into this information technology in a way that minimizes the need for labor. And here's another illustration. We can examine the / employee evaluation of some of the largest tech Acquisitions in recent history to get a sense of what's going on. Google acquired YouTube in 2006 for 1.7 billion dollars and at the time. [22:40] Her worker and Facebook bought Instagram six years later for 84 million dollars per worker and it acquired WhatsApp 2 years after that for a record 365 million dollars per worker.
[22:54] How much would that be workers God.
[22:56] That's good question David probably not much. And we see these high valuations in technology and energy companies because the role Information Technology plays allowing services and products to be scaled exponentially without an accompanying rise and labor, but because information technology is being infused with almost every business this will mean revenue and profits will continue to soar across the board without requiring additional labor.
[23:24] Desert unicorn examples Daniel giant companies acquired for even bigger amounts of money and it means a very small picture of the economy as a whole would most businesses are actually doing, but something that is a very good indicator of the economy as a whole above the boy mint is well unemployment. Unemployment [23:41] In the US and Europe right now there more than 285 million people who want to work but aren't. [23:48] The 45% of the population worldwide is either underemployed or completely unemployed so already we're at nearly half global population at that work. And to go back to some of those figures that we discussed earlier if automation was deployed today just the current levels of automation where we take away just a little bit of some of the work. 1.2 billion people employed around the world will be affected. De-Skill
[24:12] What does it mean David to be affected by automation what's interesting is that 1.2 billion figure. That's a large number of people that are going to be affected by this change and when we think of machines and robots replacing jobs I think it's easy to imagine total replacement. A robot instead of a human and it's not so easy to see the tangential effects that encroaching automation has on those that already have jobs. But the reality is that this automation that's encroaching is taking away the value of skill. It is de-skilling workers that have jobs right now.
[24:50] Let's talk about this what is de-skilling Daniel it's a word I've never heard before and I don't even know if it's an actual word to be honest but what is his concert.
[24:58] Let me give you an example David the fast food industry has had access to technology for a while now that can completely automate the process of creating hamburgers among other things.
[25:09] A Marvel of modern science.
[25:12] That's right David the company momentum machines debut the machine in 2012 that could produce 400 hamburgers per hour. Now that's a lot of hamburgers and this includes the grilling of the Patty slicing of the vegetables assembling it all under a pair of buns, and even wrapping it to go, but we still don't see anything like this in fast food chains around the world and a big reason for that is that other forms of automation have already eroded the need for skilled workers across the fast food. [25:50] Two interchangeable parts of a great machine. So this unemployment this descaling of the workforce lowering of the minimum wage and Automation in general it's also driving deep in equality, or labor polarization in the same way that wealth accumulation has been carving out the middle class in the US and around the world pushing people down the socioeconomic ladder and creating massive income disparities between the rich and the poor, but we see the same phenomenon occurring in jobs. The descaling of the majority labor force is pushing workers down the skill ladder and locking them in place there. There are no opportunities to learn skills on the job the translate to better opportunities when you function as an interchangeable part stripped of all the time. So ironically we are making machines more human-like in order to make humans more machine like.
[26:44] I want to interrupt you just for one second you're Daniel because I think this is a really good concept that we don't consciously think about these days because of fast food restaurant is at this point just something we take for granted, what is that Meme always about like you never graduated high school you're just going to go flip burgers or something is as disrespectful as that is the idea is that. Going and working at Wendy's fast-food restaurants is a low skill and what some people believe should be low pay because of that job. But honestly why is that the case food production is Hard Cooking is difficult at least if you want the food to be good, chef positions are highly sought-after it's it's tough work even something as simple as being a line cook it well it takes a lot of thinking and knowledge and expertise what Automation and this is something that we don't normally think of is automation but Automation in these. Fast food restaurants things that standardized to production of these Burgers isn't necessarily always just machines but also ways of thinking and doing things assembly line production carried over to this food production industry. De-skill is concept that Daniels discussing. What used to be in for many restaurants would still is a high school position requiring lots of expertise and knowledge in order to produce good tasting food.
[27:54] I think that's a great way to look at it David you're absolutely right and that's what these killing is. Taking work that otherwise requires a lot of skill automating the parts a way that you can and reducing the human to a very simple, interchangeable parts and because humans are now more interchangeable in this industry the value of their labor has plummeted. Minimum wages have fallen in a cost-benefit threshold of a fully autonomous Machine versus low-paid and low maintenance human. Just hasn't incentivize companies to go all the way and pay for these 400 hamburgers in our fully autonomous robots. Ultimately because they're already paying humans so little but they just don't even have to implement the full machine. But these workers they've already suffered as a result of encroaching automation.
[28:46] This is one of the concepts that's important to Craftsman this show that automation doesn't need to completely eliminate or replace a job for to impact the workers negatively, because fast food labor scene is low-skill as Expendable as not necessary to a complex or or skill based economy well workers are paid correspondingly something extremely low. And as automation becomes more complex by yourself into more of that software that starts replace White Collar jobs you're going to see what was traditionally seen as very safe high-skill work. Turn into the same sort of low pay low-skilled employment that cause fast food work to be such low pay and honestly contemptible unfortunate us that is. A working paper from the National Bureau of economic research which is a nonprofit organization that's not a governmental Department. [29:35] Yeah fancy title well nonetheless they argued that at the turn of the century that's 2000 the demand for skilled labor and cognitive tasks associated with higher education well it started reversing. Because of retreating demand for high-skilled labor highly educated and skilled workers have been moved to low-skilled work. As you might expect workers that were low-skilled begin with haven't pushed out of the workforce entirely. Which means I getting a college degree now less about gaining access to better opportunities then it is being more competitive with uneducated workers for low-skilled work. And we'll discuss education more soon but the authors of this paper suggests that the predicted future of law employment as a result of technology that replaces labor well it's not the future's already been occurring at least since 2000 what was mass in economic data. By the housing market bubble but maybe it's time for another example Daniel and what better example in Twinkies. Minimum Wage
[30:30] As many of you may have heard Hostess the maker of the famous Twinkie went bankrupt a few years ago.
[30:38] A tragedy for all of us that love highly preserved mostly plastic treats desserts and snacks.
[30:44] That's right David but the company was saved some investors came in, and they in the words of Economist and other Financial people they rationalized the production system they made it work they streamlined it and they brought some sense back into the process of making Twinkies. And what that basically comes down to is they purchased all the assets of Hostess and fired all the workers introducing some automation into the process so that they could spend less on labor.
[31:14] And I'm sure in the process also gutted whatever pension or other entitlements I'm sure existed for those laborers that they fired.
[31:21] And this reorganization of Hostess is used as an example by opponents of the minimum wage. [31:34] Alright cost they're not a benefit to society jobs are a cost of producing some benefits and if we want to enjoy that benefit we have to find a way to remove the cost. And I think we should question the sanity of this perspective because we're being told by. [32:01] Sacrifice for that benefit. No I don't want to suggest that having tasty food is not a benefit but who are the people now making the decisions about what benefits Society. How is it that workers themselves are not members of society anymore. We have economized everything to the point that impacts on humans themselves are divorced from consideration when it comes to what Society needs. And isn't the human consideration the point of doing anything of society to benefit humans. But I guess in the end when we reduce everything to the dollar figure we can now say that economic growth is a benefit and jobs are just a cost. White Collar Jobs
[32:41] Will Daniel if the topic is jobs if they're the cost of, growing an Economy Inn in some people's perspective maybe we should look at jobs what exactly is threatened, buy automation because so far what we talked about are these low-skill bottom of the totem pole of the hierarchy of jobs if you will fast food workers are cultural labor, really is at risk because yes factories will deploy more Automation and I will have something but we've mentioned a little bit so far white collar job so, maybe we should look at that in and see exactly what is that threat.
[33:14] That's right David a lot of automation talk is focused on machines focused on robots it's focused on manual labor jobs low-skill jobs fast food factory work. Even agriculture I mean there's a lot of machines now that are apart of industrial agriculture which in a way is the root of all this Automation in the first place.
[33:36] Where the automation story really gets interesting it's not robots it's not machines that do all these things for you but it's software that replaces the work of those of us that sit in offices all day.
[33:47] And perhaps one of the biggest signs of things to come was when in 2011 a computer won Jeopardy. IBM's Watson beat out Ken Jennings in an epic Jeopardy match the whole world watched. It was software David the enabled this robot this computer to win a game that many thought would be impossible for anyone but human because of how complex the language is involved with Jeopardy how much information there is two.
[34:15] Yeah I mean it's not just looking up facts like Wikipedia right a piece of software has to understand the hint or clue that you give in Jeopardy cuz everything is a play on words, it's a hint and it's much more complicated than just being like a what's the capital of Sweden that's easy for machines to do but understanding the Nuance of language well that's where a lot of this advancement in. Automation it is taking place and that's what enabled Watson to so handily take our jobs in win in game shows.
[34:43] And it's really the integration of artificial intelligence machine learning and cloud computing.
[34:50] It's a lot of Buzz words but what are they sending so glad Computing when you look at the marketing copies give me something like we use state-of-the-art servers to parallel process all this stuff but I mean know the real simple way for you to understand it is computer is running computations, calculations software except it's not your computer it's somebody else's computer and some Warehouse somewhere that you connect to over the internet. That's cloud computing that's all it is somebody else's computer somewhere online.
[35:16] And it allows for Extraordinary processing power to be carried out. [35:43] Cloud service for $90 an hour.
[35:45] We start talking about jobs but then we got distracted.
[35:48] I mean like if you talking about white collar jobs the whole reason why white collar jobs are at risk is because of machine learning.
[35:54] That's right and maybe we best understand it Daniel and little things that we find around us all the time the Alexa's the Google homes things were you saying to something ask him sort of command in response to an intelligent or semi intelligent the example of Siri Manor.
[36:09] What makes machine learning so powerful David is that we're no longer programming complex software to carry out a task on a single computer.
[36:17] At least the way we understood.
[36:19] We're no longer having to write complicated code what we have now. Is training data that computers can examine can play around with and can learn objectives that we're trying to achieve, and this training data can come from anywhere we talked a lot about surveillance and how facial recognition comes from providing a whole bunch of faces as training data for computers to learn to recognize faces. Information Technology as we talked about has. [36:52] Keystrokes so that we know how fast someone types but also how they type in and what tasks they're doing on their computer and all of this is being tracked and it can provide a foundation of training data, which computers can then analyze and figure out how to do what we're doing and do it better.
[37:09] Into the endgame of this process is at the very work that we do is training computers on how to do that work better than us and their entire business is built on this idea.
[37:19] There's a company that provides a service for managing projects and what they do is they provide an artificial intelligence that can assign tasks to different employees. But it goes way beyond that you can also hire freelance workers from anywhere in the world and as it hires people it monitors how well they are doing at completing certain tasks. If someone's not completing it fast enough or they don't have the requisite skill the artificial intelligence will transfer their work to someone else it might fire them. And all the while that it's doing this it's also figuring out how to do the work itself so that as time goes on, the software that this company has purchased is figuring out how to require less people to do the tasks that are needed to achieve the objectives.
[38:08] So that means this software. Is quite literally learning how to do these tasks and it was hiring out and they're only hiring Freelancers to do the bits that it hasn't learned how to complete yet and then in that process is learning how to complete those bits from those three letters doing that work.
[38:23] That's right and so this is where we have this influx of a number of factors that are creating The Perfect Storm, for automating away so many of the jobs that we consider high-skilled work like data analysis or like Consulting any job that does not require some kind of face-to-face interaction, is that risk for being automated away, so data analysis is an example of accelerated consolidation that's occurring right now as a result of this artificial intelligence combined with machine learning and and all these factors. There's a company good data it's a Silicon Valley company that offers businesses data analytics to monitor their business. And before cloud computing the CEO of this company noted that they would have had to employ at least six people or each business that they consulted and in 2012 with the integral. [39:22] And today they consult over 70,000 businesses.
[39:26] So how do we identify businesses industries that are right for this easy automation replacement. And what are the ways and Economist business strategies have figured out. This is the process where jobs are done digitally and remotely in other countries and has become a really good indicator of what will be fully automated next, so any job that deals with information and doesn't require this face-to-face interaction that Dino mentioned is very easy to fully automate. And those jobs which we think require face-to-face interactions will you know what we should actually start thinking hard about that. Just like Google has done with these call center jobs where we assume that a human had to be part of this equation you needed a human voice I can understand the complexities. He would speech well in the same way virtual reality video Bots and generated faces might be soon replacing what we thought had to be a face-to-face position. And we find as we go forward them more more jobs that we thought were good job safe jobs will these are vulnerable to automation just as much if not more so in the traditional low-skilled low-pay jobs that you normally get the brunt of automation critiques.
[40:31] Like journalism there's a company now that's combining data analysis with creative writing. And a lot of sports journalism is being replaced, biography thumbs that can turn through enormous volumes of sports-related statistics and events every week and create narratives around that that sound as if they're written by humans, what can be used in newspapers and online media to report what's happening in sports, and this is also being used for Consulting companies now that have huge volumes of data and figure out connections between the data and then create narratives around them, this is one of those jobs that would traditionally be seen as very high skilled requiring at least a bachelor's degree and maybe something like data analysis statistics or Finance, that can now be replaced by algorithms that have learned how to write the way we want them to.
[41:24] And I suppose the pressure of the modern media focus on more articles faster places like Twitter, have contributed to influence this transition to automated news in journalism even faster maybe to have a light-hearted example of automation replacing jobs, we remember of course I can we forget, currently still in the news all the questions about Twitter Bots Russian Bots Show bots of all makes models and types that put the internet social media Reddit Twitter. All this in the ways of one of automation where automation has replaced. Jobs a propagandist of PR companies of advertising and multiply the effects of these techniques these tools, making it to one person or small group of people have enough sway online using these techniques of automation to change the narrative control the media, and to as some people might try and argue even sway election.
[42:17] And another job you might not think of his being at risk for automation is attorneys during litigation that occurs between companies, all these internal documents need to be examined to find anything that may be relevant to the case and this has always been done by armies of attorneys and paralegals that sit through paper documents. And try to find connections, well now because all communication is being done digitally algorithms are being used to search millions of records for Relevant documents during litigations and humans have been taken out of the process, or simply de-skill. Like we talked about so there's actually a lot of attorneys now paralegals who have jobs sitting at a computer all day. [43:04] Relevant or not relevant in a case and it's about the most boring thing you could imagine doing especially with a law degree.
[43:12] And so now what might have been a firm once of many lawyers is reduced to a firm of just a few lawyers, and as we do skilled is Workforce as we reduce the number of lawyers required to complete a job well that has downward pressures on the wages of the law industry as a whole and that might be why we have seen. The average salary of lawyers decline over the past decade increasingly speeding up as time goes on, because as it turns out law is one of the most vulnerable Industries for Automation and this Bastion of what we used to considered one of the safest and highest paying jobs available to us, turn into eventually nothing more than somebody like Daniel mentioned who sits in front of the computer cooking yes or no all day long. And that's no way to spend your time after spending you know how many years is law school, traditionally has been the answer to automation that to replace front of jobs it says go back to school but is that going to be nicer for much longer.
[44:07] That's right David I mean this is one of the arguments, that is made when we start criticizing this automation is taking over the labor force is hey automation freeze people up to pursue higher skills it allows you to, get a better education and so we have to question this aspect of this argument say is education going to get us out of this problem. Because this is one of the main recommendations made, by institutions around the world I mean the IMF is one of these institutions that says in order to combat the unemployment that occurs throughout a nation is by increasing investment in education. So what do you think David can we improve education and prevent people from losing their job.
[44:49] Education to scab decrypting us this didn't love that will never Escape will be employed forever and have no jobs anyway because the robots all took them and they'll beat us in the acid month with whips to mine acid for the robot batteries.
[45:02] Maybe David maybe.
[45:03] You know listeners one of us is automated I'll leave it as an exercise to the audience to figure out if it's at or Daniel or me.
[45:10] Whoever seems less humans probably the one that is most human.
[45:14] Yeah that's what it robot would say.
[45:16] Look David I think when you look at education there's no doubt that it's had a huge impact on worker productivity and worker competitiveness following the Industrial Revolution but as we've discussed this time is different. Education Can Save Us? [45:29] One of the biggest things that makes this time different is the margin of change that's possible from education. The Industrial Revolution brought machines and some forms of automation to factory. Those machines were not possible without massive amount of Labor that had basic education and could conform to industrial work. And there was a huge gap between non-educated people, and those who were suited for this industrial work and so the effect that the emergent public education had on the productivity of the economy was enormous I mean at this time many people lack basic literacy, and we're coming from agricultural backgrounds so the amount of change that took place to conform labor to industry was massive. But today 9 out of 10 people with in the developed world had a basic high school education and maybe half of those have some kind of degree After High School. So the potential to educate workers more will that potential is just much much smaller.
[46:29] And you might interrupt to say will David Daniel once again you've missed the Crux of the argument in those Economist and their fallacy of the luddites well they're still right.
[46:40] Why are there right David what have I missed.
[46:42] Well you imbecile you Daniel you see you have not considered the fact that someone needs to program these robots in the first place so the future is all of us sitting inside programming the way at the day I all day long of course.
[46:57] But I don't want to do that.
[46:58] Well listener you would be wrong because those that create the automation themselves the software Engineers are in fact being replaced by their very own creations. Just a couple years ago the first automated software was writing its own automated software. That's right he's a I've giving birth to other Ai and have continued to refine this process and well right now they aren't at the scale of a good programmer or even honestly a decent or mediocre one the process in the growth in this field has been phenomenal. And most experts consider that the vast majority of programming work will in fact be fully automated in the coming decades. And so only those of the most talented expert senior-level programmers. Will be able to compete in this field because the vast majority of work and the vast majority of programming once again one of these other quote good jobs is going to be automated by the very software that they write right now. Sowing the seeds of their destruction as well as the collapse of our economy as a whole unfortunately.
[47:58] David this is going to affect everything and I don't think we should go much more death about all the types of jobs that could be automated away I mean a big one is Healthcare, right there's no software that can read medical images 10000 times faster than a human radiologist. In fact in one case two days worth of work for one radiologist was taken care of and just 46 seconds by an algorithm.
[48:24] It off in case is even more accurate than human radiologist can be.
[48:28] That's right and the University College London Hospital just announced a three-year contract to employ artificial intelligence algorithms within a system, and these machine learning algorithms will be used to diagnose diseases like cancer identify individuals that may be at risk for illness. [48:51] And past behavior and then send them friendly reminders. And he's algorithms will try to improve patient wait times in emergency clinic by prioritizing those with the highest risks. Of course there are concerns about patient privacy and cybersecurity but you know what David I think this kind of raises an important question in my mind which is, is it a lot of this automation good I mean if there's a computer out there that can read images better than a doctor and tell us faster what someone might have in terms of an illness, it's not a good thing let me take the Watson example that IBM robot that won Jeopardy it relied on 230 million. [49:41] And in the medical field we still have a system where doctors are diagnosing illness largely on subjective experience, of course they're highly educated but a lot of them have to rely on the previous cases they've seen, and whatever literature they're up-to-date on and the amount of literature that is out there is massive.
[50:01] Massive and always being updated and growing and while you might graduate from medical school with a deep knowledge of the what's current and what's known, 2030 years go by and if you're not constantly reading medical journals something and overworked doctor has no time to do. [50:16] The medicine of 30-40 years ago is very different to the medicine of today and you might at this point B peddling and not current information technology or even ideas of pathology as a whole, something that an AI That's essentially updated and constantly loaded with the most up-to-date information so it wouldn't have that problem.
[50:34] In the United States alone there are over 5,600 medical journals in each of those published anywhere from a dozen to hundreds of papers every single year that's a lot I mean it's not possible for doctors. Keep up with all of this information and so something like a Watson for the medical field that can use machine learning to, identify a patient's symptoms correlated with all the medical research and come up with probabilities for illnesses that can streamline the process of an educated and skilled doctor making a decision, well that has enormous potential to improve what is becoming an unmanageable Healthcare crisis globally.
[51:14] It's right maybe we're getting ahead of ourselves and we loaded to this in the beginning of the episode but in fact. As dire as these unemployment figures might be of a world over descaling a Workforce where future wages are uncertain and at-risk automation should be a good thing. It's taking away the shity parts of Modern Life I'm doing these boring repetitive tasks. Of enabling us to have better health care of of bringing up our time to do the things that we actually want to do but it's only because our world is dependent on work on us individually working to survive this is going to be a problem. [51:48] How do we get ourselves into this position Daniel.
[51:51] Will David I think it goes back to what we said at the beginning we took away all the. [52:05] Provide security and comfort or people of this Earth the system is setup to create dependence among us so that it is easier to exploit our labor at the cheapest price. And we don't really have a plan of transitioning those people who are going to lose their jobs as a result of automation.
[52:24] Well there are some plans for things we can do which which will address in a little while but I want to hit on some of those points yes the land is ultimately the absolute and Crux of this where we're forced to work to survive in this land but why are we forced to work so much, Jobs That Lack Meaning [52:38] you go back to the early 20th century to the mid-twentieth century and they were promising us that with the development of Technology, with the things like automation being developed at that time that in the future we would only be working 15 or 20 hours a week, what cycle early hunter-gatherer ancestors what happened. We became more productive we make more money for those employing us never before but did our wages increase that's normally the answer they say will we didn't realize that people would be so consumed with consumerism. And they decided they needed to work more in order to be able to buy all the things they wanted and it was a trade-off and that was what those are really Economist got wrong because that really true. Almost half this country can't even afford the basic necessities of life. That is to say the rent their food and their health care over 40% 43% can't even afford a $400 emergency that doesn't sound like a country driven to work in order to consume more that sounds like people who are working just to survive. But at the same time as you mention our productivity I just been increasing what kids how much of a Workforce is just employed in dead-end bullshit jobs.
[53:44] Probably a big portion of.
[53:47] This is a question that actually been asked before and defining a bullshit job is difficult and complicated there's actually a book that just came out on this call bullshit jobs. Grapples with this question and then one of the things he mentioned is a couple studies that people did that asked people straight up and said do you feel like your job is worthwhile do you contribute something to society is your job useful or simplify do you have a bullshit job. How many people do you think said my job's probably bulshit Daniel.
[54:13] 60%.
[54:16] You got a pessimistic look at Society I could see but now it is close to 40% depending on the study of people admitting that all of their job is bullshit. [54:25] And overwhelmingly the people who said this wart manual laborers they weren't janitors or hairdressers things that have conventionally been lumped into this bullshit job world but instead white collar workers lawyers. Accountants advertisers. Any of those that don't consider their jobs bulshit when asked about how much of their day-to-day work is actually useful as actual work less than half of it was anything so too productive so what the hell are we doing with our time. We're working more than ever our productivity is higher than ever and we hate every minute of it. Should we be celebrating this automation that's here to save us from this awful dystopian hellscape that we created.
[55:03] Well David I mean what are we doing with our time why why do you think so many people feel like the job they're doing doesn't contribute to society in any way.
[55:12] That's a really great question on something that that is quite confusing especially for people who point to the market as this Arbiter of what's efficient. Why an efficient economy and efficient capitalistic environment and maybe it's not pure capitalism or whatever flavor that you want to claim it. What it's supposed to be essentially efficient if a job is necessary or useful then it shouldn't exist right this who's going to pay somebody to do that. [55:37] But still these bullshit jobs persist we've all had work that we sat around wondering what the fuck are we doing I've done nothing today and they're paying me just to sit here. We have those moments where we've been working a job we wonder what am I doing at this position I'm contributing nothing and some people say what you don't realize the greater picture, the top down view you are craving something viable but you just not seeing it because you just a small piece of the greater machine maybe that's too and some scenarios. But when I'm sitting here working and I truly believe with all my heart that I'm contributing nothing that's not healthy not as satisfying way to live my life even if I'm being highly paid and I've been in room sometimes sitting there being paid $5 a minute to sit. [56:17] And I wanted nothing more than to leave and never come back. And run out in the into a field of Sun and light and lie there in this nice weather feeling the son of my back and being paid nothing. Seems like a lot better use of my time so is this not an assault on all of us. Being forced to work in these horrible conditions pulling away to create ultimately nothing the vast majority of people if not your entire job but at least large portions of the day today and the illusion of trying to look like you're busy of working. Shouldn't we be looking forward to automation excited that eventually our job will be done by some robot that we can come in and do a couple hours of work for the part that the AI the robot couldn't do and then go home and do something better with our time make a terrible podcast, Pizza Mart or go be social physical whatever it is that makes us happy.
[57:07] I think it's clear that the economy is at now ewells the world has become totally indifferent to the humans that it's. [57:31] Jobs there will we come out ahead because we have a net one so everybody's happy but like you're saying David is everyone truly happy. Are there potentially 9 people out there who have nothing now and no way to recover and live except maybe by accepting some menial tasks that, doesn't mean anything to them when we turn all societal considerations into equations we sacrifice the part that is human which in a way. Is appropriate for a world in which machines replace us. And you know one of the rebuttals too many of these concerns that we've raised in this episode about automation include the argument that says hey were lifting people out of terrible and slave-like working conditions. The company Foxconn just a mention here and the company that manufactures a lot of the iPhones that we use, that's in China and is famous for those suicide Nets that prevent people that jump out the windows from filling their goal of escaping the terrible working conditions. Well because of the need to mass produce iPhones and because production needs change according to surges and demand will the optimal setup at this company has decided for these workers is to live on site so that when they need to increase production, everyone can be woken up in the middle of the night and they can be forced to work overtime. And the argument goes hey those are terrible conditions but if we automate those jobs and then we don't have to wake people up in the middle of the night we don't have to subject people to these terrible conditions where they want to commit suicide we can just make the robots do it. [59:03] And this angle totally ignores the systemic Insanity of the system that allows slaves in the first place, and the fact that we're trying to find a solution to terrible working conditions by increasing the productivity of this work. Seems to ignore the underlying question of why is this work being carried out in the first place.
[59:24] Let's see how this plays out Daniel if we don't fix this economic system that incentivizes of these bullshit jobs all this work to survive what happens in 2030 when 70 million jobs have been in limbo.
[59:37] The places that will be most heavily impacted. David will be smaller cities towns that are less than a hundred thousand people these are the places that could be at the highest risk because a lot of the jobs in rural areas are automatable. Whereas cities attract a lot of high skilled labor a lot of Education creativity and diversity of work will a lot of towns and cities that are smaller and population can support this educated Workforce. And so a lot of these jobs are at risk for automation.
[1:00:09] Of course this is the first line to be hit as these automation tools get better and more advanced, well this will follow up and hit these big cities just as much as this small town but Rural America small towns smaller cities less than a hundred thousand people are already struggling we talked about this in the infrastructure episode we talked about this in the pension episode. Both of these things will be exacerbated by loss of jobs as the income of towns municipalities and cities are cut off by Workforce is increasingly out of work. There goes with of the income that supports these towns the taxes the money that goes to pay for all these infrastructures and Central Services the naval people to live in these places in the first place. What's more of the pensions of companies of governments that are already failing are going to be increasingly stressed as the workers that are paying into these pension to support them are cut more more replaced by automation tools. So where one worker is now doing the work of 10 workers before will that still just one worker paying into this pension system. And these systems which are already stressed the breaking point guard that much more vulnerable to collapse. We see collapse at this point coming from all across the board because as more people are out of work less money is flowing through the system the collapse comes from all ends and we going to see increasing economic inequality and disparity occurring because of distinct recent automation. At least if things continue as business as usual.
[1:01:31] The city death spiral David that we talked about in our infrastructure episode I guess there are many ways that this can come about.
[1:01:39] Some political leaders are already becoming aware of this coming problem but the solutions for this are complicated one of the areas that we're trying to see right now is people bashing up, the tide of low-income workers by increasing minimum weight something that really should be happening anyway as living and costs have increased dramatically and minimum wage has not. Who Paid For Technology?
[1:01:58] David I want to come back to this minimum wage because when you talk about minimum wage David cases like that Hostess example that we talked about earlier play into the hands of people who argue against raising the minimum wage. They say that if you raise the minimum wage company start shedding labor that can be automated and the implication is. Where at least employing you if you want us to pay you more we're just going to replace you. And this race is a question in my mind of who actually Bears the cost of much of this technology and who should benefit because a lot of the technology that enables companies to replace workers has been paid for by. [1:02:44] The Labor Center at UC Berkeley found in 2013 that half of all fast food workers in the US have families that rely on public assistance to survive costing the US taxpayer 7 billion dollars. [1:03:04] And the purchase of automation technology comes from labor cost savings made possible through subsidies. But we pay for this technology much more directly because much of the research itself is funded by taxpayer money. DARPA a taxpayer-funded u.s. department gave a lot of the early seed money for that computer network that became the internet. And are provided money to Apple to develop Siri and has helped pay for computer chips at IBM in addition, the National Science Foundation gives grants to universities for research that is used to improve the manufacturer of computer hardware that powers the machine learning algorithms replacing service-sector jobs, and the semiconductor industry Association. [1:04:04] And lower labor costs for themselves. Universal Basic Income?
[1:04:07] And so the solution to lot of this is once again all of us paying to fix these problems and it's most notably found in Universal basic income or Ubi, if you haven't already heard about this you can be hearing about it a lot over the coming years and decades, it is not already too late at that point and honestly a u b i could be its own episode but the Crux of it is this. The government pays all of us a set amount of money no matter what. No matter what you do so say it's $10,000 a year $20,000 a year and this money supposed to be enough to Mabel you to survive somewhere maybe not well but at least be able to feed yourself and Hauser. You can live without having to work.
[1:04:46] What sounds like a great concept David.
[1:04:48] Yeah honestly you know like this is a great weighted maybe we should have been talked about this a long time ago but the problems ran into this when we start talking about how to be fun this, immediately the the only way to do it is to eliminate all welfare all that is gone and this replaces all of that, and then at the same time this is going to be even with with that the reduced income that would come from a country that's half out of work. It's going to mean that much less income tax set aside to enable to pay for something like this, couple it in by 2030 will be over 30 trillion dollars in debt at the national level for the US which means almost a quarter of our budget will be going to servicing this debt and finding money to pay for UTIs can become increasingly difficult, as the economy gets tighter than when we need it more than ever.
[1:05:31] I mean isn't the idea that if you can replace labor at a lower cost than a human that itself is freeing up money that can be used to them support that worker I mean it seems like, the math would work out as long as companies are not hoarding all the benefits of this automation for themselves.
[1:05:48] Yes be magnanimous companies that have happily redistributed all the tax credits they just received from the Trump Administration. To their workers have demonstrated in the past that we can trust him 100% with incentives like automation, I'm in order to pay. What wait wait a second I just sorry that was digitec over my head that was all wrong we made up that none of those things actually happened actually hoarded all the money and will continue to do so into the future I'm sure, and demonstrates why Ubi is never going to fundamentally I think be deployed in any sort of scale that actually supports people's ability to live without work. It becomes a crutch to replace welfare with the same time still demand before to survive.
[1:06:29] I think you're right David that could and probably should be its own episode.
[1:06:33] I I mean I can literally sit here and talk for another hour on this TV I but the one concept I want to take from it is that it's an admission of two things 1. [1:06:42] If we don't do something Society State civilization as we know it is going to collapse, either by people just straight up starving to death on able to house themselves or by General discontent by Travolta people who when they can't feed themselves when they have no place to live what the hell else are they going to do except Reach Out. Grab something violent and and say this is not right I shouldn't be subjected to this this is not what humans should do to each other. And it's admission UB eyes if it was supported by people largely it in in hypothetical positions in an economic position CEOs Elon Musk is one of these. It says if we don't do this in everything collapses all this money and well that we build will be for nothing because Society civilization is gone, and at the same time it says that it also admits and this is the thing I think they did right really more than anything that we shouldn't have to work to survive. This is how we open this episode I think the real concept that we need to take away from this is that automation enables us to have all the Comforts of our civilization without the work that we put into it the busy hours every day that we waste. Like reading nothing I'm getting our productivity up there and not actually enjoying the fruits of my labor of working just because we're supposed to work or at least look like they're working. Any bi says you know what. [1:08:13] Will live with a possibility and happiness in a world where we don't have to work I think is an essential component of that. And at least where we don't have to work in this bullshit jobs there's nothing wrong with work itself in fact working on something that's hard or or difficult but being able to see the fruit of your labor. To feel proud at the end of a project and see what you need is one of the most rewarding things that humans can experience so what gives us purpose in our lives but how much was actually get to enjoy that in the work that we do day today. In a world where we aren't forced to waste our time on this busy work on things that don't matter I think a lot of people and I know I would, will find joy in doing this work that we care about it work that ultimately makes us all better that makes civilization better that makes all of us better, and if it's Ubi that we need that in order to get us there or just an admission that are economies broken our style living is broken our civilization is broke it we need to rethink our relationship with work well if that's what it takes that's what I'm prepared to do. So bring on the automation bring on the unemployment I'm ready for change. What Can We Do?
[1:09:13] Bring It On David so in the face of this coming technological change that is going to reshape the way we. [1:09:31] And do work with purpose that mean something to us.
[1:09:34] Being able to enjoy the fruits of our labor of workers being a part of the productive process and owning the things that they create is a great step for to realizing a world where we only work for what we need only work enough to satisfy the requirements of a Modern Life.
[1:09:49] Like we said at the beginning it all comes down to the land at the fact that the land is closed off from us and that were forced to work just to survive is what has created this problem in the first place this this problem of, jobs that don't mean anything to us jobs that shouldn't exist in the first place and this disconnect from anything meaningful in our lives in terms of our work. Because we've been integrated into an industrial economy that has reduced humans to interchangeable parts that simply play a small part in a large machine that is Distributing Goods across the world. We no longer have any connection to the work that we're actually doing if we're not part of something local and so. For me I think this ultimately comes down to the land we need communities that are locally adapted that can sustain themselves that can support themselves. And that's going to take all of us and maybe as consumers we need to think about the human element when we do our shopping when we go about our lives what in the same way that the economy reduces humans, two interchangeable parts $2 figures we as consumers do the same thing when we value something for it convenience or its cost, without considering the people behind this product that were longing for or purchasing and where it might have come from. You know if we want to be able to order a cheeseburger on a phone app, select a time for pick up and then drive up to a fast food chain and pick it up from an automated window if that's what we really want businesses are more likely to implement it and with it means the labor goes out the door. [1:11:28] So what can we do I think as individuals and as communities we need to support things, that are more in line with sustainable communities and at the same time start thinking about the human labor that's behind the products and services that we consume and expand our values Beyond just the dollar figures.
[1:11:46] As always at the lot to think about but we hope you will think about this and consider what your place is in the workplace, as we go forward towards his automated world and our definition of work what it means to work is challenged. You want to learn more about any of the things we talked about today we the full transcript of this episode or much more you can do all that on our website at ashes ashes. O r g.
[1:12:10] A lot of time and research goes into making these shows possible and we will never use ads. [1:12:28] Just with a friend and giving us a review also we have an email address it's contact at ashes ashes. O RG and we encourage you to send us your thoughts positive or negative will read it and we appreciate it.
[1:12:41] You can also find us on your favorite social media Network at ashes ashes cast or read it at our ashes ashes cast. [1:12:49] Next week redeem deep into debt so we hope you'll join us for that could be an educational and thought-provoking episode as always but until then this is ashes ashes.
[1:12:59] Bye.
[1:12:59] Goodbye.