I completely disagree with the article's assertion that 'knowledge professions like ... lawyers' will be relatively unaffected. 'DoNotPay' (the robot lawyer) has already helped overturn millions of dollars of parking fines. NPR article on the subject: https://www.npr.org/2017/01/16/510096767/robot-lawyer-makes-...
The thing about lawyers is that they're often able to create work for themselves. They can lobby for laws that create opportunity if automation affects them too much. They can create class actions out of nothing and recruit their clients. They're also the ones that write the laws, and you can always expect that the rules of the "game" will always favor those with the most input into those rules.
Except the real trend has been the exact opposite. We're now in the fourth decade of lawyers (legislators and judges) narrowing the laws you're talking about: class actions, discrimination claims, environmental claims, and securities claims. As a result, the legal sector as a percentage of the economy has been shrinking since the late 1980s: https://lawschooltuitionbubble.files.wordpress.com/2012/12/r....
I personally think that's a good thing because the economy was over-regulated in the 1960s and 1970s. But in some areas, it's arguably gone too far. Judges and legislators are so self-concscious about excessive litigation that they've significantly hampered the ability of claimants to recover when they should be able to do so. E.g.Wal-Mart v. Dukes.
Agreed, this goes for other professions as well such as accountants and real estate agents. There always seems to be some new complexity that requires the profession. That being said I do think it will put a dent in the number of professionals. Sites like homie.com for real estate already seem to he effecting real estate personnel.
Some of these automation companies are doubtless being run by people with legal backgrounds. I can't imagine all lawyers are interested in preserving the status quo.
I happen to be watching tree trimmers work on a tree from my office window, so I entered that into their form. Very high automation risk (77%)... but I'm looking at this work, and it's absurd. Automating it would be extremely difficult, and serve no purpose since there's nothing particularly expensive or wasteful about the current process. To be fair, it identifies plumbers as having a fairly low automation risk, which I think is reasonable.
Still, I'm skeptical how they identify automation risk, and I suspect they get it wrong in a way that makes small cities seem more vulnerable.
okay, but what if there is a home robot with significant dexterity for generic tasks exists for other reasons. would you stop its owner from using that for tree trimming?
my rule of thumb is simple, if a job only exists because of some amount of cognition needed with fairly simple decision making it will be gone regardless of the low pay. also, some higher end jobs like investment management are already getting automated given that they follow the same meme. its just that their cognition/pattern recognition tasks involved sifting through balance sheets (e.g. NYSE:AIEQ).
Investment management? Sure, that seems pretty easy to automate.
A home robot that can trim trees is FAR more complex. It needs to know about trees and what to trim, and talk to the customer. Maybe doable. It needs to analyze the environment to look for any dangers, fragile things, property lines. It needs to navigate heavy equipment. It needs to make a series of cuts, and make sure nothing too large falls after cutting. It needs to collect and dispose of the waste. This is way beyond any robotics we're seeing.
In a sci-fi world where robots can do EVERYTHING, they can also trim trees.
But we are not just talking about replacing humans entirely. How much can more advanced robot-tools speed up the process? Like the human operator talks to the customer, makes some gestures in AR to signal where to cut while improved robots / tools take care of lengthy / tedious work.
I totally see this happening and requiring much less human labor. And as with most automation, it will be a gradual process and not a one step replacement.
robot dexterity and cognition are both very very hard problems. Once we solve that, we just don't get human hand equivalents. We can make very tiny hands that can manipulate cells and perform intricate surgeries, or huge hands, that can move boulders and build a house in a day, or hands with 100s of snake like fingers manipulating 100s of things simultaneously and running a click farm.
Once we can build artificial brains more powerful than humans, and artificial muscles more dexterous than humans, almost every profession is at risk. In addition if machines are not made in a factory, but replicated organically like seeds because we figured out the magic DNA, then we've essentially created von-neumann probes and can conquer the rest of the galaxy.
Its possible that some mundane examples will be automated, but the law is not a deterministic code that can simply be interpreted and executed. Automation may make the job less tedious and might reduce the numbers of lawyers needed (or more likely, paralegals).
Many entry-level lawyers have historically been involved in document discovery. AI-based software and having a significant impact on the job market for freshly minted lawyers. Software now performs ediscovery.
From first hand experience, I would say that the willingness of clients to outsource discovery to cheap contract lawyers has had a far greater impact on the job market for fresh graduates than automation of discovery. The software for electronic discovery just isn't very good or smart. It also requires a ton of up-front work to train the model which makes it not cost effective except for the largest document reviews.
I've heard this example bandied about every time this topic comes up, but does this actually mean that those entry-level lawyers would be out of a job? Or does it simply mean that they would be freed up to perform other tasks? Wouldn't that be a win for a law firm that's presumably trying to grow and take on more clients?
The current job market for lawyers is kind of messed up. When in law school you can be applying and accepting job offers clerking or working at law firms 2 years before the start date.
I don't know what even more increased pressure on that job market would do to it.
You accept an offer conditional on passing the bar. But even once you're out of school, if you want to do a clerkship you have to apply for it 1-2 years in advance.
Quite a few paralegal jobs have been removed simply by adding document search capabilities. Automation that cuts the work down by 50% can have huge impacts even without dealing with all the edge cases.
I was under the impression that DoNotPay worked well because text is fairly boilerplate (if so can you really call that automation?). Plus, I don't think many people pay lawyers for parking tickets.
As far as law in general, I have heard convincing arguments that it would be difficult to make ML for defense and judging, given the tainted historical data (and likely that historical data will always be tainted as our morality and laws change).
Also, from my (very much outsider) observation of the legal profession there seems to be a lot of boring work that probably could be automated. So maybe not lawyers themselves, but legal helper workers, will increasingly be replaced.
It seems to me that knowledge worker support staff (paralegals, PACs, etc) are the current low hanging fruit in terms of automation. Firms without them will be able to turn higher profit on less revenue.
Right, knowledge professions are actually innately geared towards being automated. However its worth noting that the higher levels of these jobs cannot be automated, as well as the human aspect of the courtroom (judges). Whatever algorithm you apply is going to need datasets (i.e) parking ticket cases so it cant operate on the higher end of the legal system. The judicial system also provides a place for the meaning of laws to be decided by humans, you cant automate that.
The legal services industry in the US is around a $250 Billion industry, example of a automation of 0.004% of that doesn’t mean anything. Again the Toyota story doesn’t seem to contain any long term, industry wide statistical analysis. Without that it’s just an anecdote.
Millions will wake up with no goals and no structure to their lives because they have all of their most basic needs satisfied by UBI if the democrats have their way. Republicans will require mandatory public service to obtain UBI. Optimists tend to think this means more time for artistic passions and inventive endeavors. Pessimists think people will spend more time watching Big Bang Theory. Usually and answer is in the middle.
There are plenty of countries that have very liberal welfare payments. They're extremely easy to get and anyone that wants to can live on them for years and years without ever having a job.
You could look at those examples to see how society will be impacted...
Australia is a good place to start your research.
(It's not quite UBI because you have to apply for it and do something to get it, but it's pretty close)
Anecdotally, from family members in the medical industry who work with low income patients, people on welfare (who could otherwise work) don't seem to have very fulfilling lives. I think it's pretty clear that people need something to work for or towards in order to find some sort of meaning in their life.
It's definitely a relevant point. If a large group of people who are not working seem to be predominantly unhappy, that's legitimate information that may be useful for discussion about universal basic income. You can make a compelling argument that the reason for that unhappiness is not innate to a lack of work, but rather due to the particular idiosyncrasies of the implementation of welfare, but that's not obviously true a priori.
The commenter seems to have made this observation in good faith. It seems rude to me that you would insinuate the point is irrelevant and "almost seems trollish" just because you disagree with it, particularly when you did the bare minimum to mount a substantive rebuttal. It's true that we shouldn't conflate the unhappiness associated with welfare with some sort of ennui induced by a lack of productivity; but that doesn't preclude the point you're disagreeing with, which could still be true.
Anecdotally, from poor people, people would rather be physically healthy and have a place to live, even if they are unfulfilled, than be malnourished or suffer from untreated chronic illness and likely be unfulfilled anyway.
That's why instead of UBI I think we should make the government an employer of last resort for anyone who has exhausted their unemployment benefits and can't find a regular job. We essentially did that during the Great Depression through the WPA and other agencies. It functioned pretty well. There's no reason we couldn't do it again.
UBI seeks to avoid a permanent underclass by enhancing labor force mobility.
A long term policy of government as a mandatory employer of last resort with make-work jobs doing work for which there is no real demand is a route to creating a permanent underclass. (A short-term policy of government as mandatory employer of last resort in response to an unusual economic disruption where this serves to maintain skills in jobs for which market demand is expected to rebound, to provide bridging income support, and to leverage temporary labor surplus for public infrastructure is a different story. As is a permanent policy of government offering non-universal last-resort employment in regions or industries experiencing short-term downturn without an economy-wide collapse, though that obviously can't replace general universal safety-net programs.)
A UBI would certainly create a permanent underclass of unemployable people. Make-work jobs are better than no jobs at all because they effectively act as training. Some people need training in basic job skills like showing up on time and properly dressed for work, and following a supervisor's directions without mouthing off. And there are actually many simple jobs that would add value to our communities and will never be automated in our lifetime. Things like painting over graffiti or maintaining trails in public parks. It's a lot better to do that then pay people to sit around and stagnate.
I agree. Taxpayers should see a return for their welfare investment. The problem is, like in prisons, we now view manual labor as a cruel and unusual punishment yet this is the type of work we need the most in the US (rebuilding infrastructure)
I'm not sure if the stigma will go away, but a lot of the reason not to embrace the welfare lifestyle is the stigma associated with it. With UBI everyone is getting it, so if you use it to live off of without working... good for you. I'm still getting mine, too. With the cumbersome traditional means tested welfare programs you get handouts and I am over here working my butt off paying for them. In its nature it makes welfare systems hostile to the working class if they cannot benefit from them.
>Millions will wake up with no goals and no structure to their lives because they have all of their most basic needs satisfied by UBI
People keep predicting this utopia of people sitting around twiddling their thumbs with nothing to do as all basic needs are provided. These predictions go back to the pre-industrial era.
Inequality is the fuel of ambition and simply having the basic needs covered won't be enough for people in the same way it currently isn't for upper middle and upper classes -- they still compete.
The "work" we will be doing will just be different. 3D printing and autonomous vehicles are going to have a massive impact on manufacture and distribution. However people still need to create things -- and AI is currently just good at optimizing things not "new thinking".
Recently unemployed people have historically guillotined the rich during periods of massive automation and spawned communist rebellions instead of sitting on their thumbs.
I'm not sure how you're defining optimists and pessimists, since more time for artistic passions and inventive endeavors almost certainly also means spending more time watching TV or otherwise finding and making entertainment.
That is a pessimistic and simplistic view of a incredibly complex issue, and for millions of poor people antidepressants are already popular, when they can afford them.
The alternative is what, exactly? Continue having jobs automated out of existence, letting the middle class cease to exist while placing the blame on the segment of the population that has little influence over what trades will be valuable tomorrow? How about we subsidize coal mining jobs long so people can stay busy?
This kind of black and white thinking is pointless. Using this as a justification for ignoring growing problems is negligent. There are a lot of things that people could do, and probably would do voluntarily if they didn't have to do their dead-end retail jobs.
One option is taking care of the planet we are hell-bent on destroying. Imagine if we had potentially millions of people to take care of our forests and rivers? What about taking care of the elderly or teaching or any number of jobs that are practically poverty wages now that would be possible while living a middle-class lifestyle?
Any requirement to obtain UBI by definition means it's not UBI. And I've not heard a single proposal for "mandatory public service" in order to receive any benefit. Not to mention that UBI is squarely a libertarian (small l) proposal and neither the Democrats nor Republicans want it. Republicans only want it if it means slashing public services (one of only two ways it is even fiscally possible, which is a nonstarter for the vast majority of Democrats), and Democrats either don't want it or want it as part of a huge tax increase (the other way, which is - surprise! - a nonstarter for the vast majority of Republicans).
I don't know why you're being downvoted. I'm not a good fit for most labels so I've got quite a varied social network, and I've been pushing the idea of UBI out to them to gauge what folks think. It's approximately what you've mentioned here. Average people seem skeptical but open to the idea, interested but certain there's a catch. Republicans look at it like a way to cut welfare funding and Democrats look at it like a way to increase public assistance. I've got friends on both sides looking across at one another, wondering how something could be both things at once. I personally think most Americans would support UBI if it were spelled out clearly what it is and also what it is not.
A bit of anecdotal evidence of the pitfalls of welfare and what it could mean for UBI.
I grew up in an area of questionable safety. There was a woman across the street who spoke to my mother about how she would scam the welfare system to receive multiple payments every month. This was on top of food stamps and child support checks. She was making more than most people with full time jobs in the area while doing absolutely nothing for society with no desire to ever change.
People will take advantage of the system if they can. Spelling out what it is and what it is not would have to mean making people understand that it is only the minimum and somehow encourage them to expand their horizons. That seems like a cultural shift that needs to happen before UBI becomes prevalent.
> People will take advantage of the system if they can. Spelling out what it is and what it is not would have to mean making people understand that it is only the minimum and somehow encourage them to expand their horizons.
People on welfare often already treat it that way, doing under the table cash work while receiving benefits (many don't, either for moral reason (it is fraud, usually, both tax and welfare) or fear reasons (the punishments for such fraud can be catastrophic) or availability reasons (they just haven't found opportunities, which aren't exactly openly advertised). But it's still fairly widespread; UBI would normalize additional work and stop punishing it, and allow it to take place in the formal economy, which all by itself would produce exactly the culture shift you refer to.
Okay, but let's look at it this way: What do you suggest we do when fifty percent of the US workforce is no longer employed because of clever kinds of automation that smart folks like you and I personally weren't able to predict way back in 2018? What do we do? How do we keep babies from starving and all wealth from aggregating in the hands of a dozen people worldwide? Let's not talk about how likely (or unlikely) this eventually is and hedge our bets. Let's play like we're "Societal SREs" and do a little emergency management. Let's talk about what happens if the article's conclusions come to fruition. What do you suggest we do?
Because if we leave people to their own devices, they'll burn the whole fucking thing to the ground. Open literally any history book to see examples.
I was pulling a number out of a hat to make a point. But you should know that nobody has a crystal ball. Before 2008, institutional investors were saying they were prepared for "apocalyptic" losses of 10 to 15% of the housing market. Sometimes reality outpaces expectations. Regardless, you can substitute "a sufficiently large number of people" if it offends your delicate sensibilities.
> Millions will wake up with no goals and no structure to their lives because they have all of their most basic needs satisfied by UBI if the democrats have their way.
And? People are remarkably good at finding something to do once they are bored.
A lot of people veg out in front of the boob tube because they are so exhausted from working to pay for basic necessities that they simply don't have the energy to do anything else.
> And? People are remarkably good at finding something to do once they are bored.
I completely disagree. Bored directionless people, especially bored directionless men, are the worst thing a civilization can have. These are the guys that pick AK-47's and join terrorist organizations because they have no direction in their lives and they think they found it.
Sure, if a terrorist organization is the only outfit nearby which offers an outlet promising deeper meaning or empowerment.
But if people are given more than the empty prospect of future penury and distress, wouldn't they find values of discovery and creativity to be much more appealing?
Who picks up an AK-47 and joins a terrorist organization just because they have nothing better to do? I think that sort of thing probably comes from a deeper malaise of the soul, and an environment devoid of hope is probably a big ingredient there.
They're dangerous if you don't have an efficient way to pacify them, which we thankfully do. In the modern developed world, they can pick up virtual AK-47s in Fortnite or join a League of Legends team instead, which is much less destabilizing.
Forget about terrorist organizations, we have a long history of examples where disaffected young men with few prospects drop out of productive society and chase the danger, excitement and potential riches of the drug trade.
I've spent a lot of time building automated assembly lines and I think the writers of the article (and the underlying research) gets a lot wrong. I doubt the smallest cities will be prime automation areas since areas with high degrees of automation require a few highly-skilled robot babysitters that are unlikely to relocate into the provinces. I think large cities are much more likely to see high automation rates due to economies of scale and small citiies will either stay the same or businesses will simply leave.
I think you'll find enough highly skilled baby sitters in the provinces living there for personal reasons. McDonald's shouldn't have any problem finding someone in a city of 50,000 to watch their burger flipping robots machines.
There is a presupposition in this article: The automated jobs will leave a vacuum. That is, routine jobs will be taken away with no other jobs to replace them.
What if reducing routine jobs frees up minds for higher pursuits? In the same way that high level programming languages such as Python have increased the amount of programming jobs, as contrasted with writing in assembly. A quick search on Dice shows Python jobs outnumber assembly 8 to 1.
You only have to go back a little over 100 years to a time when agriculture accounted for over 50% of human labor. Now it's under 2%. Yet unemployment is still under 5%.
While agriculture likely represents the largest shift in labor during the 1900s, there were a huge number of jobs automated away during that time. Assembly lines, factories, and automation have been displacing workers since the dawn of the industrial age.
75 years ago there were basically no information technology jobs, and now it represents nearly 10% of the workforce, or four times the level of agricultural employment.
In summary, I agree with you; freeing people from mundane labor will lead to an explosion of jobs for which people are better suited; jobs which are more engaging and fulfilling as well.
There are always people who fail to adapt and get left behind, but the rest of the world moves on. People like to work.
Up until now, the people displaced from the jobs automated away have always been intelligent enough to offer utility beyond what automation can provide. But there's a tipping point when automation/machine intelligence exceeds humans at which point they offer no utility to employers of any kind. We haven't hit that point for the vast majority of workers, but the early indications are that we very well might.
But looking at the past as a predictor of the future is a very dangerous proposition. We saw that in 2008 when it came to the housing market. Sometimes now really is different from the past. And if we hit that tipping point, it will be much like the posited AI singularity. Vast swaths of humans will become surplus to the most efficient means of completing work. It's important to realize that the people who are talking about this issue as one we need to confront are envisioning a future that is fundamentally different from today or anything that has come before now and they're predicting that for a very specific reason. You can disagree with that reason, but you can't use previous examples of automation-displacing-jobs as a way to dismiss that thinking. Because what's being predicted is fundamentally different from what has already happened.
There's not much to learn from agriculture. People left farms for better jobs in factories. That's not a story of unemployment and displacement.
If a new source of jobs pops up that can employ 50% of the population at higher wages than they are currently earning, no one will complain about it (except maybe for the employers at their current jobs, but they will probably just buy the proverbial tractor).
Factory jobs started out being awful, and workers had to physically fight and die to get basic protections. Eventually we got a 40-hour work week, less child labor, etc.
Well it's still better than watching your family starve, sure. Reshaping our society to accommodate a lot of people who don't have to work (because machines are doing the work) is going to be a long and painful process.
Interesting :). Though, this sort of presupposes that the people whose jobs are automated away have the financial security in order to take the time necessary to create a new job/career.
I think a lot of people, in the U.S. at least, don’t have that level of financial security (plus the healthcare snafu).
What is financial security but the ability to afford to pay for one’s needs? Automation in farming has cut the price of calories, and perhaps too automation in other sectors will reduce the cost of living. Automation in medicine could be a big help for costs.
Unfortunately though, housing cannot be automated and that is very expensive. Though much of housing costs, at least for new builds, is from regulations and labor. Reduce regulations with better governance, and reduce labor with automation (3D printed houses?), and you’re on your way to part of a solution.
The true cost of housing is relatively flat across the developed world and especially the US. The actual cost is almost entirely a function of local politics.
We could have a flood of nice, brand new homes for $300k in Silicon Valley and nice older homes for $200k, in the same way a city like Tokyo or Houston does, if the people of California had not made the deliberate choice to pull up the ladder behind them.
You won't have brand new homes in $300k in SV because the land itself is worth more than that due to the heavy demand for housing in what is a comparatively small area, though you might get $300k condos in multi-story buildings if the zoning issues in the Bay Area are addressed.
Houston isn't a good example either. Houston has almost no zoning or building regulations, which was on ample display when entire neighborhoods flooded because developers weren't prevented from building houses in flood basins. (CA allows the construction of residential properties in floodplains, which are flat areas that might get flooded if flood control mechanisms fail. In Houston, houses were built in the flood control mechanism.)
There are also plenty of places in California where houses and residential units are being built. In LA, for example, several ten thousands of new residential units were built in the downtown/downtown-adjacent units in the past few years. In San Bernardino/Riverside counties, they still have housing stock available from the last building boom. It's not that the people of California pulled that ladder up; it's simply basic market forces: too many people want to live in the nice parts of California.
This idea that automation will free up others to perform "higher pursuits" (especially programming) seems completely wrong. The reason is that we have a tremendous shortage of programmers, at least in the largest cities but in a lot of smaller towns too. These jobs pay much more than truck driving or burger flipping. If people could do these jobs, they'd be paid tremendously more with better job prospects right now.
There's nothing stopping people from training to be programmers today and working as a spoiled or fortunate software engineer, just as I am today - nothing other than the challenges of training, the time required, their own interests and capabilities, and possibly unawareness of the opportunity. Maybe a few people would try programming if they suddenly had money and no need for a daily job, but they could also do it today.
YouTube and free SDKs, IDEs, and cheap servers are quite a democratizer.
But is it true to say that automation would fail to free up higher pursuits because there is a shortage of programmers? What would you be doing right now if there were no electricity?
The point I was trying to make is they could already be doing programming bc there are so many jobs and improving their lot in life but they aren't. Freeing up more people to do programming won't change the job equation. There are already openings to dev.
But the question is whether automation across all sectors would free up all kinds of pursuits, not just programming. Maybe that wasn't clear in my question, but that's what I was proposing.
It doesn't matter how cheap calories are if you do not have the income to purchase them.
If you like, change food to housing in my original comment. The point still stands: Higher minded pursuits require a level of safety and security to pursue. That level of safety and security does not exist if you do not have food security and housing security, and those things don't exist if you do not have an income.
This is why when economists blame lost aggregate jobs on automation they are being disingenuous. If the government thinks that there aren't enough jobs they can just raise the minimum wage and increase fiscal spending, creating jobs until there are.
There were similar claims made during the depression. It wasn't about automation then either - it was about wealth inequality and a lack of aggregate demand.
Roosevelt then spent and invested in infrastructure - building stuff like LAX - until the country returned to full employment.
> Ok so tell me more about how building LAX was "make work".
No, I'm asking if that's what you're proposing for the 21st century. What projects would you have be initiated that would fill the gap of some roughly 60% of unemployment, in the case of most cities on that map?
* Green energy infrastructure - wind/solar farms and the smart grids to support them
* Fixing roads, bridges and dams (all currently neglected).
* Building community centers and schools
* Caring for the elderly and infirm
There's a curiously high number of people who declare the jobs program that built LAX an obvious example of make work yet defend silicon vallley's silly startups and bitcoin mining (the purest example of make work) as "socially" useful merely because they were funded with private dollars.
I used to be pretty big on the UBI bandwagon but I've come back to Earth recently and the reason is pretty simple - while we certainly do not take advantage of the total sum of intellectual talent available today - we squander a huge amount of it to poverty and cultural isolation - even with global UBI and universal access to education there will never be a population where even the majority of people can do those jobs.
Tech is a bubble, and we dig ourselves in deep. But you don't need to go to the third world to find illiteracy and ignorance. You often don't need to go more than a few miles in any given direction to find entire populations of doomed people - they have no ambition, education, and most importantly no desire to ever have them. The smart ones leave, go to the big city, succeed in school and go to university and end up successful, but for every capable intellect there is often someone who just can't. Its no different than introversion vs extroversion - some people crave the new and different and some people crave consistency and pattern.
And we might like to think otherwise, but automation replaces the routine, not necessarily the physical. But a lot of people don't have the capacity for boundless creativity, in my experience it actually hurts some people to have to think outside the box that much. Like headache causing stress and frustration. They don't become scientists not because the opportunity is not available to them fiscally, but because they cannot think like that. At least not in a professional capacity with forty hours a week consistency.
I don't think we can eliminate that factor. And I think it is much larger than anyone who promotes UBI as leading to an era of scholars and scientists aplenty gives it credence for. Or for not recognizing that as a species we are barely evolved beyond animals that could only, at most, do simple tool manipulation and use. Who couldn't understand the concept of self. Asking for human brains to on average be innovative, imaginative, and capable of running in the highest states of consciousness we can reach on a near constant basis is asking too much of a lot of people.
It asks too much of me many days where I try to find something rote to do rather than face new problems that can get overwhelming, so from someone not on top of the brain potential spectrum, there is a reason why predatory repetitious skinner boxes work, why addiction is such a big problem, why people get stuck in their ways and why change is hard, and these aren't things we will ever eliminate from the species entirely without a serious investment in transhumanist innovation. These people will not become scholars, creatives, or thinkers.
That being said, I'm not convinced its even a problem. A lot of why people crave "meaning" in their lives is the fear of loss. Having a purpose gives you an anchor to always ensure you never go completely adrift from your life as it is. You can lose your job, home, etc but if you have a purpose and feel like you can do something it gives you the strength to keep fighting to survive.
With a UBI, the fight to survival pushed on pretty much everyone by capitalist economics as a means to spur growth changes fundamentally. I would be interested how societies pressure to have purpose and how we build our self worth would change with it.
No matter how you shake this, we're looking at 60+% additional unemployment for every city. Some fare much worse. And I remember from my history studies that the Great Depression was officially around 33% unemployment.
I've listened to the national talking heads and political figures, and nobody publicly is talking about this. Not the Republicans, and not the Democrats. However, the Dems seem to have a better idea by bolstering all sorts of subsidy and welfare programs. Those work well and good for keep individuals out of poverty, but falls short when working with communities of impoverished. These end up counting on, and using tax revenues from well-to-do areas. And these unemployment numbers show that there won't be 'well-to-do' areas.
The WPA was one such way out of mass unemployment. But given how effective automation is, I question its use this time around, if the current political atmosphere would even have the stomach to discuss such things.
This is one result from capitalism. Accelerate until we hit the wall. And this wall, is that we will no longer need "labor" provided by humans. So, what do we do with the 95% of people that must work? That's the $64,000 question.
My personal plan: I'm learning electronics from the ground up. I'm teaching myself control systems theory. I'm learning how things are put together - electronics are very similar to code libraries. You do a function block to do something, and you chain them together. This means that I'm untraining specialities, and learning how to analyse and fix quickly. I guess being a sysad also helps in that. I'm not putting much trust in the national or state level government. I see more done at the local levels, but without tax revenue/money, they're up shit's creek as well.
60% is unrealistic. It's still very expensive to make anything as reliable as a human being if we're talking about anything other than pick-and-place (and that was really expensive the first time around).
For a business to automate, it must invest substantial resources into the automation process and product. It must work in an industry where consistency is the most important quality customers are looking for.
Automation already affected most of what it's going to do in manufacturing.
Given the capital and talent being deployed, automated driving is on the way. Most other industries are not seeing similar talent and capital being invested into automation. Long-haul trucking is the industry I have the greatest concern for. Most of the others (e.g. short order cooks) will be fine for quite some time.
I have to admit I disagree strongly. How could you have this opinion and be in the workforce over the past 20 years? Automation isn't one day a human-looking robot shows up and does 100% of your job better than you did it and now you're unemployed.
Automation is we got a new machine in the shop that increased throughput, and we never replaced Betsy and George when they quit. It's getting that new cashier system that is idiot-proof and has pictures to tap of dollar bills vs. a keyboard or requiring any sort of math skills. It's redesigning the bathrooms to make cleaning take 30% of the previous time using basic automation and techniques.
It's a long slow reduction and dumbing down of the workforce that won't happen overnight, but has already been happening and on an accelerating pace for 20-30 years or so now.
Compare the skills needed to run a short-order cooking station 30 years ago to today - it's not even comparable. We would call such a position skilled or semi-skilled this day in age, as the average kid off the street couldn't be up and running in 6 hours of training. It would take months to replace that worker with someone in-kind, where now your labor pool for such jobs is basically "anyone who can show up on-time and sober" which of course pays considerably less.
I see automation as the long slow destruction of semi-skilled labor. You either are effectively a cheap very flexible human robot, or you get paid well in a highly skilled position. The middle is being hallowed out rapidly.
Tell that to the explosion of automation occurring right now. My local McD's just put in the touchscreen kiosk that'll ultimately replace the entire counter crew. The 'cooks' are already just a few folks running from machine to machine pressing buttons. That can also be entirely replaced (see other timely HN posting about MIT students' robo-cooker).
I find it odd that McD's went with something as archaic as self-ordering kiosks when you can just order from your phone.
Those kiosks are very slow and a bit cumbersome to use, and they put them smack in the middle of the waiting area in the restaurant. Huge waste of space.
But otherwise you'd have to download the McDonalds app and store your details in there. With the kiosks, you don't have to leave your personal details and don't have to sign up before you first use it. I guess this makes a big difference. I prefer the kiosks over ordering at counter staff but probably wouldn't want yet another app on my phone (I'm only at McDonald's every other month or so).
I don't have any of those apps for the same reason (barely use them, don't want to have even more apps). But I guess you still need to scan your card? The kiosks have proper card terminals, they work faster than adding card details to an app (with the exception of Apple Pay).
I also look at places like Chili's, Applebees, and the low-midgrade sit downs.
Whenever you see a tablet on the table, that's their way of automating staff on various levels at a restaurant's level. They sell it as a way to play games tableside and to request server assistance...
In reality, its so they can hire even less servers, have push-demand by customers when they wanted waited on, and sell more stuff quicker. It's also a way for management to ascertain actual engagement levels along with sales data.
It's another case of short-sighted automation segmenting out under management in low end sit down restaurants. You certainly don't see high end restaurants engaging in these shenanigans... yet.
Our (wife and I) decision is we place them face-down off the table. We prefer to engage with humans, not a idiot-box upper management dictates.
>It's another case of short-sighted automation segmenting out under management in low end sit down restaurants. You certainly don't see high end restaurants engaging in these shenanigans... yet.
It's already happening in some cases. A former colleague who is working in China (and I'm sure doesn't regularly eat at low-end restaurants) was telling me how in a lot of places you order from your table by scanning the QR code at your individual seat and pulling up the menu.
But even driving is going to have a massive impact, beyond just trucking. That will be the first to get hit, but then you hit the taxi services. And then the big hit - the entire auto industry.
How many people are going to decide they don't need to buy a new car, they can just sell their old one and have a fleet pick them up and drive them to work. Driving as a subscription. Then you lose car dealerships, and car manufacturing jobs, because we'll need fewer cars. Parking lots will become less necessary as the cars can just continuously drive around anticipating the next customer.
Of course, that is all assuming full autonomy, which I think we are still pretty far away from.
Car dealerships should have been disinter-mediated long ago. In my opinion, they add no value to the process of purchasing a car/truck; except for unnecessarily driving up the price. If i want expertise on which to buy, i wouldn't want the person whose incentive it is to get me into their most expensive car...i want advice from a neutral third-party who knows cars/trucks. I'm not an expert on capitalism, but it seems to me, that car dealerships - at least in this modern era - have been held up to exist in such an artificial way, which distorts pricing for me as a consumer. If car dealers failed to exist, yes, i would feel bad for the low-level car salespeople who now have to find another, different livelihood...but honestly, isn't this like trying to keep the jobs of the telegraph operators in our information age? Car dealerships should be left to go extinct.
I used to agree with you but there are several points where you're flat-out wrong.
> they add no value to the process of purchasing a car/truck
Unless you get a new salesman they typically are experts on the cars themselves. Yes there are exceptions but especially if you're shopping at higher end dealerships (think Audi and higher, basically) they're going to know everything about the car and be able to tell you exactly what model fits your needs.
> the person whose incentive it is to get me into their most expensive car
Car salesmen are paid on gross, and the dealership's profit on a $120,000 7-series is not going to be twice the profit as a $60,000 M2. In fact, highest gross is on used vehicles, and those are the hardest to price shop since every one is different.
> car dealerships have been held up to exist in such an artificial way
That depends on your definition of artificial. Audi isn't allowed to sell me a new Q3, by law. If you change the law and get rid of dealerships, guess what happens? Audis are now available at MSRP only. No haggling, no getting a good deal because something's been on the lot and is a prior year model.
The part you don't mention is the used car market. What happens to that when dealerships disappear? You can buy one from a private owner but then you get no warranty at all and you have to trust that they're not trying to screw you. I hesitate to draw a finance comparison, but don't dealerships offer some form of liquidity to the used car market? Audi corporate does not want to manage the trade in inventory. Won't the value of trade ins (and therefore used cars) drop like a rock at that point?
Tesla has showrooms. Tesla has service centers. They offer financing. I'm not a fan of car dealerships and spend as little time as possible in one when I buy a car. But there isn't a whole lot of evidence that if I started buying cars directly from the manufacturer, that car costs would be drastically reduced.
It's a real mix. Higher end malls in growing urban areas, in particular, are doing pretty well. Low end malls with bankrupt anchor stores in more stagnant areas, not so much. Within probably a 30-45 minute radius of where I live, there are malls and shopping centers that are always packed and ones that look about ready to close down.
My anecdotal experience is that people can't wrap their heads around such an enormous, complete, shift in the way labour and capital will work.
"It is difficult to get a man to understand something, when his salary depends on his not understanding it" - Upton Sinclair.
The few truck drivers I've spoken with all seem to think that autonomous trucks are going to arrive just after they retire. Similar stories come from nurses, fast food workers, and construction folks like drywallers. They all say "oh, that won't arrive for at least another 30 to 40 years!" I don't belabour the point because that just makes people fearful and angry.
In the USA there's a huge amount of people in depressed rural areas already. Their fear at being left behind by society has been manipulated into keeping the Republican party in power by various hatemongers. Now imagine the fear and anger they and many more will feel when 95% of the jobs disappear. What sort of person will rise up and lead this vast angry mob? How many of them will claim disability? Imagine 200 million people potentially starving and angry, many of them with firearms? (NB: I make no statement on the ethics of firearms!)
I am deeply worried about the era my teenage stepdaughter will live through.
The only thing I cannot wrap my head around is why we won't create new jobs. Are we truly months away from reaching the pinnacle of human achievement?
We've quite literally lost 95% of all jobs before. That part is nothing new. However, what we found was that when people were jobless and bored, they found new things to do. Things that nobody ever imagined people doing when they were busying doing other jobs.
It is not like human-level robots are anywhere on the horizon. At best we have simple pattern matching machines that can be trained to do certain tasks. But they cannot cross tasks. A self-driving car robot cannot start becoming a self-practicing doctor robot on a whim. A self-practicing doctor requires building an entirely new robot specialized to do medicine.
Up to this point, each time we lose jobs, the new jobs people invent have become more and more diverse. At one time, 95% of the workforce all did the same job. Now the largest industry employs something like 5% of the workforce. The numbers are getting smaller. And are trending to be even smaller still. It is not difficult to imagine a time when people will all have distinct jobs, that no other human does. The economies of building a robot to replace a single person starts to diminish quickly.
The industrial revolution replaced physical laborers, the information revolution will replace mental labor. If both physical and mental labor is replaced by machines, and you're an average or below average individual, what decent career prospects will you have?
Every job has a physical and mental component, so it is difficult to determine exactly where you are drawing the line. But if we say that a mental job is one that is done primarily sitting at a desk, I would suggest that the majority of jobs still fall under the physical labor classification. In fact, the agriculture sector alone, which the industrial revolution changed more than any other, is still about the same size as the entire information sector.
To dig deeper: If we look at the fastest growing jobs, according to the BLS, there are very few on the list that are not primarily physical jobs. Of the top 10, there are only three that I would consider as being more mental than physical.
If this is what a post-industrial revolution that replaced physical laborers looks like, I'm not sure why we're worried about some future information revolution. And that's without even accounting for the fact that the information industry itself is tiny.
i don't think job growth really matters so much as wages. Automation has had a pretty clear impact on middle class wages, while high-level and low-level wage jobs have continued to grow
The problem is how easily the workforce will be able to adapt to these new jobs. It's one thing to have new repetitive manufacturing jobs invented (as with the industrial revolution), and quite another for a growing market of engineers (electrical, mechanical, software, etc).
I second that. A year ago I had a discussion with a guy who works as a driver in mid-sized trucks that deliver supplies to ships. So at some point the discussion went to autonomous cars. He had just the reaction you mention so I opened YouTube and showed him a video from a Mercedes S-Class that drives itself around a town. He was literally shocked to the bone. He looked at it and was speechless for about ten minutes. It was like reality smacked him on the head. Right there he realized that the future he saw coming in 20 to 30 years was just less than a decade ahead. For someone in his early forties this is terrifying.
If we combine automation with climate change effects it seems that in the next 20 to 30 years humanity will have to face some very serious problems.
Automation isn't about saying one day "Whoop, youre through. This robotic maid will do everything you can do and more."
That doesn't happen. Instead, its computerized tools that "allow" you to do 1.2x more work. And 5 nurses means the owners get rid of the 6th nurse.
Then a new comprehensive patient device comes out, and can act like a .5 nurse. The owners buy 2 of them. And then they get rid of another human, or fail to hire when one leaves.
By the end, you have only a small handful of people, with technology as multipliers. And if that single person doesn't show up, then "bad happens".
That bad, with many instances, is the justification to finally depend on full automation.
I have always wondered how automation will adapt to things like ice on roads/parking lots/equipment, or salt everywhere like in Minnesota/Wisconsin. Is this being considered in sunny California/Texas?
I would love to see a breakdown by country. Given the endless rhetoric around manufacturing reshoring/offshoring/etc., it would be great to see if thats even something we should care about given how much automation already exists and how much more is about to come online. Owning the robots and supply chain will obviously have some value, but politically speaking, the value is often measured in jobs which seems like its minimally shortsighted if not downright disingenuous.
There is a lot of confusion between "automation" and "artificial intelligence", I believe that automation facilitates, and modifies the professions in the short term. AI is something that should be discussed in a broader sense.
I used to believe that automation will lead to greater unemployment. Now I think that two things happen: goods get cheaper which lowers the minimum amount someone needs to earn for a job to be acceptable. People who still have a great job have more spending our. This unlocks new jobs that previously nobody would have paid for and nobody would have accepted. But both maximum price somebody would pay went up and minimum price someone wants to perform went down. Recent examples are all these delivery services and obviously Lyft and Uber, although the car services are a little more complicated since there always were taxis.
I think there are three forces discussed here. Automation (removing work). Decreased cost good (reduced labor costs). Increased wealth (those with it spend more on goods and services).
Automation's effects on the economy larger depends on the pace it is achieved. If someone released a box that replaces truck driver's tomorrow and sells it for $1000 then there will be huge disruption. If drivers are replaced with automation slowly over 30 years the impact will be less because the rest of the economy will have time to absorb the impact.
Increased wealth is expanding the service industry, with petcare being a prime example. This will create more jobs, but those jobs are both at risk of being automated and might not be enough to replace automation.
Decreased good costs works for some things like toothbrushes and iphones, but they don't work so well for limited resources like gold jewelry or real-estate. It might improve quality of life in some ways, but it cannot help with everything.
Except it doesn't matter if general goods and services get cheaper if life necessities like healthcare and residence have their costs ballooning out of control. That $8 minimum wage job in ten years might be able to buy you twice as many trinkets at the same price but that doesn't mean anything if you have no where to live for less than $3,000 a month in rent.
Thats not to say that you cannot make housing and healthcare dramatically cheaper and more efficient, but we simply are not doing it, and there doesn't seem to be nearly enough momentum to see change happen in that regard. Lives rarely get better without a good protest, but there is no uniform movement with only big wig opposition to fixing zoning laws or socializing medicine. Its all working class vs working class infighting between haves and have nots.
from the article: "In a small town, there are likely a few small restaurants run by a few people who do many things—cook, clean, manage the books, etc. “Some of these tasks are easily enough defined to soon be automatable,” Youn says.
By contrast, in a larger city there will likely be some much larger restaurants that require more specialized knowledge and skills—perhaps a marketing team, or a lawyer who specializes in the restaurant industry—that cannot be easily automated."
I have a small factory, we use a fair bit of automation. I disagree with the above quote - I think that the larger restaurant will be more likely to automate first, and arguably the franchise restaurant model is automated marketing and business management. The larger places will go full robot in the kitchen sooner, the smaller places will continue to have one person wearing many hats for a long time. Larger places have more budget, can afford more automation, and the smaller places would have a harder time justifying a burger cooking robot if it would only be in use a small fraction of the time.
My welding robots are super fast. I have two so that I can reduce setup and changeover time. The robots are so fast, the main headache is feeding them enough jobs - selling the work. The setup and changeover is very slow. I spend 5-10 hours getting a new jig built, programmed and tuned up.
This setup time is never discussed in the media. Sure the robot will be amazing and fast and super productive - once you spent 10 hours programming it to do so. If the task is a 'once-only' and it only takes an hour, who in their right mind would automate that? This setup time will be the biggest impediment - humans are pretty quickly adapted, autonomous fleshy robots, that outperform the best automation when it comes to setup time.
I love automation. I have 6 automated machines, and a couple 3D printers, that you can push a button and go get a coffee, and there is the part when its done. Sometimes one button push can make a couple thousand dollars in products - that is a great feeling! But the setup time is massive, and this is the greater expense over the cost of the machine.
Its like an algebra equation. y=ax+b where 'a' is the setup time, and 'b' is the time for one unit.
With a human welding a simple part, 'a' would be 5 minutes, and 'b' would be 5 minutes.
With a robot welding cell like a Panasonic PA-750, the 'a' would be 500 minutes, and the 'b' would be 2 minutes.
If you need 100 of them, no sense in turning on the robot. If you need 10,000 of them, the robot welding cell will be done in 20,500 minutes instead of 50,005 minutes for the human welder. My Panasonic costs $2284 a month for 60 months on a lease, $1 to purchase at the end. The human costs $1000 per week in perpetuity, with increasing costs due to inflation if nothing else. The robot is half the hourly, and its paid for in 5 years. The human is never paid for, it always needs compensation. In this example, the human takes 500 hours more to do the task, and that is $12,500 more in wages over the robot. My 7 year old son can run (not program) the welding robot and make parts, which he does for Pokemon money on the weekend.
So where does that leave the human? Doing small runs! Which is better for the human anyway! Welding the same small part for weeks at a time makes you a little crazy. Its like watching the same GIF every 5 minutes for weeks at a time. Welding the same part every 5 minutes for 2-4 hours is quite doable, can be fun to get in the groove. I think in the future that production runs will become smaller as CNC machines proliferate. Everyone will want something custom, which is what humans are best at. High volume will be robot welded, because its boring and machines do it faster and better.
I manufacture in America. Using automation, I am directly cost competitive with China and India in my core products that we have been making (and optimizing manufacturing) for 14 years. Its 1/4 steel plate mostly, and our cost to build is directly in line with the best quotes from Asia. Robot welders are cheaper than any welder in Asia. We loose on thinner stuff, like 16 gauge and lower, and we lose on items that need a lot of hand work. But we are right there for heavier items. We did a large welding job for Chinese factory last year that was for the US market. Our price was the same, but shipping was less since the product was already here. We got the deal over their own factory. 28,000 units.
TL;DR - things are better for humans than these articles would have you believe.
There was also the hand made rolls Royce cars and the other robot assisted manufactured luxury cars. Which one suffered from poor quality perception (for good reason)? Automation can reduce errors and shoddiness.
Sure, for most things. I am finding this not to be the case with things like furniture, clothing, small finer metalworks that require something like a blacksmith, food, soap, etc.
I am not underestimating them, I am planning on it. All my capital investments are in automated equipment, that I know how to operate myself, and that are flexible in their use - they can make things for many industries, out of many materials.
The future is high mix short run manufacturing. Just in Time. Anything that has to be shipped and not flown, is never going to be 'just-in-time'. Locally made, custom just for you. This does not mean that automation won't be involved. The automation is the core of it - without the miracle of CNC (praise be!) the local guy wouldn't stand a chance.
My customers don't seem to care if it is robot welded or not, or made in the US or not. I doubt 99% of them could tell me which parts are robot welded and which ones are manually welded. Everyone claims that they care, but the top sellers in my segment are all made in Asia. So the evidence does not match the claims. The customer just cares if its what they want. Thats it. We use the "long tail" strategy in our market segment to differentiate with offshore competitors. We offer 3x the part numbers, and 2x the vehicle coverage, plus we do custom sizes and the others do not.
Automation, and flexible use machines, let me change styles to follow trends, and get low labor costs without resorting to offshore. We can customize parts one a one-by-one basis for clients if needed, and we do so usually once or twice a day.
I view my factory like a breakfast restaurant. I know I am going to sell a bunch of eggs, bacon, potatoes, toast - but I don't know exactly how much, or in what configuration. I have a menu, and I know what is roughly the most popular on any given day, but I never know exactly what the customer will want, and it doesn't matter since we make it fresh for them right when they order. I keep the raw materials in stock, we build about half our units on a fixed schedule based on statistical high movers, and the other half is built to order from the orders that day. There is minimal WIP on the floor, I try to keep it under 2 weeks.
Given the choice, people want to get things personalized just for them, or to have something a little unique. How many t-shirts do you see people wear on any given day, and how many are the exact same? Almost none, yet the basic form is the same for all.
TL;DR - things will be fine. Humans will have jobs. Have robots, hire humans.
Looks like software application development is listed as having a relatively high automation "risk", around 21%. Anyone know how this was calculated. Doesn't have to be a link for this particular site, just some good reading material.
My initial reaction was that this number is hard to nail down because on reflection, so much of what I do gets automated. That's the nature of software. When I started programming for money in the late 90s, I still managed memory. That's automated now through garbage collection (for me). I also used to do a lot of direct network and socket programming, also automated away. I managed my own database connection and pooling. JDBC came along and abstracted that away, but a decade later, I heard someone who used an ORM say he wouldn't want to write all that "low level" jdbc code ;) I guess you're getting old when the high level leaky abstraction of your youth becomes the low level code for a new generation of programmers.
My whole career has been one long progression of getting automated away, by this perspective. I remember when I proposed getting a web server up and running for a project in college, and a professor remarked on what I was doing. There was a time, he recalled, when doing a "spreadsheet" was fairly advanced and took some serious technical chops. The same would happen to web servers - but for the few years then they required specialized skill and had a big impact on business, this knowledge would be very valuable. The moment would pass, but something new would come up, and that would create all kinds of new apps that require an app developer.
To me, the moment when nothing new comes up - ie., when there is no new moment for human app developers, all software development is now automated - that is the moment when we replaced by automation. I wonder if that's what this article is talking about? If that's the case, 20% might seem high, though not out of the question. But by a more general definition, the one I discussed above, I'd ay the odds that my current skillset will be automated away is roughly 100%. I want to be clear I'm not (necessarily) talking about churn, I'm talking about new breakthroughs that fundamentally change the job. That kind of change I don't mind at all, I enjoy it, it's probably why I'm in this field. Churn, on the other hand, is what makes me want to quit.
I'm rambling on here, but it occurs to me that software development probably isn't that unusual. It's the jobs that don't change as much that are unusual. Lawyers, for instance, have certainly changed over the years, but we expect people to still be lawyers in 100 years. How many horse and buggy drivers expect that job to still be around? How many lathe operators or truck drivers expect this?
I'm thinking, maybe the best metric isn't the odds that a particular job will be automated, since for many (most?) jobs, that's pretty much 100%. The question is, what are the odds that a job will be automated with no path to the next thing. That's the fundamental transformation going on here. And yeah, it actually is happening. A horse and buggy driver can become a bus driver, but what happens when the concept of a drive goes away completely? Will anything emerge?
Another aspect of automation I'm interested in is the complete bindside. Music, I think, was one of the interesting automation stories of the last 100+ years. There were various attempts to replicated the human mechanical process of playing an instrument, and they all appeared very unlikely to do much replacement. Yeah, piano rolls made an appearance, and you could try this I suppose by hooking up some contraption to make a flute play.
But what happened instead was that it turned out you can actually store music physically in various materials and play it back. Music went from 100% live to less than 1% live in about 50 years. But we didn't really automate music, we replaced it with a similar product.
My guess is that this will be the story with software development. It seems absurd to me that this could happen through automation, but that's because I'm locked into one way of thinking about it. Kind of like trying to imagine how a crankshaft could power a bunch of strings tied to a bow and some pegs to try to recreate the physical process of playing a violin. It won't happen like that. Something utterly mind blowing, like storing music in grooves of different depth in vinyl, of all things, utterly unanticipated. Yeah, that's generally how this massive shifts happen.
Yes, I'm right there with you, and the cool part about our jobs being automated away -- besides not having to write all that boilerplate -- is that when the automation takes a steaming shit, we get to be heroes. I'm fine with this. I guess until it doesn't matter anymore, but then I suppose by that time, it won't matter for a whole lot of people, and we'll have something worked out.
I find it bizarre that they single out fishing as a job that's likely to be automated away. I don't see this happening any time soon. Fishing crews are already pretty streamlined, and many things on fishing ships are mechanized. I can see automation replacing more of the post-catch processing on giant factory ships, but I don't see robots costing many other fishing jobs in the near future.
The biggest impact to fishing will probably be that we've fished out most of the worthwhile wild food species. I would put my money on any significant growth occurring in fish farming, or things like lobstering, where we are effectively farming.
Regulations limiting catches and seasons to try to preserve the fisheries in some state of production have already driven many people out of fishing.
You're pointing out how a lot of fishing jobs have already been automated while trying to argue that automation isn't happening. The remaining jobs look difficult but that's only because we already got the low-hanging fruit.
Could you elaborate more? From my very amateur point of view, fishing seems like something that could be completely automated pretty easily, but it seems like I'm not seeing the complexity.
A large part of it is liability. In theory you could just send a self-driving boat out with a giant net to scoop up a few tons of marine life to bring back to shore and stuff in a processing plant. Or even better maybe make a self-sufficient platform that uses current to turn turbines and just scoops up fish, sorts them, and puts them on ice for cargo carriers to come by and pick up.
The problem is the same one as self driving cars on a more massive scale - accountability. When you have human bodies at every step of the process you have someone to blame if you "accidentally" kill a protected species, overrun your quotas, put tainted meat in the market, etc - if everything is operating on its own you get the same situation where self driving car operators are going to try to blame the software engineers if something goes wrong, while the company gets to take the fallout.
And the fines on illegal fishing are actually legitimate compared to the token paltry fines in other industries (depending on the country, of course).
The economics are completely on automations side here, you could get 97% good fish for a tenth the price with some substantial up front investments today. But that 3% false positive / sorting error case with the inability to blame line workers kills the business.
The ocean is a dynamic environment. Nets get tangled, lines get tangled, wildlife interferes, weather and sea conditions are harsh and unpredictable. The ocean isn't a nice flat square corn field that a combine can lazily navigate through. We've just barely figured out how to automate picking strawberries in a nice safe greenhouse.
Also in reference to manufacturing processes, it appears that Toyota has learned some hard lessons in automation and is actually moving in the opposite direction: http://www.thetruthaboutcars.com/2011/02/toyota%E2%80%99s-se...