Overweening Generalist

Sunday, November 10, 2013

Robots and Technological Unemployment: Further Considerations

In my last blogspew I wrote a bit about all the ideas and rhetoric I encountered as a kid in the 1970s, reading books and magazines from earlier in the century, when said rhetoric was about the End of Toil. And this seems possible, but we are stuck in a dumb-game about being unfathomably rich, or living in a constant state of biosurvival anxiety due to lack of money and the fear of poverty, homelessness, hunger, penury.

Maybe the ballsiest rhetoric about "all that" came out in 1930 - just as the Great Depression was setting in - by Lord John Maynard Keynes. In a short yet profound essay, "The Economic Possibilities For Our Grandchildren," Keynes - a polymath - wrote that the end of the economic game was in sight, and that many wouldn't know what to do with themselves, that working three hours a day is quite enough for most people, and that a few will know how to live a life of leisure - the goal of a true liberal arts education - while others will have a rough go of it.

"I feel sure that with a little more experience we shall use the new-found bounty of nature quite differently from the way in which the rich use it today, and will map out for ourselves a plan of life quite otherwise than theirs." - Keynes

Read this essay if you haven't before. (If it seems "tl/dr" skip to the II section.) He says that within 100 years this end of toil would be possible. As we write: 16 years and change from now. Which reminds me of a couple of studies that came out in the last 18 months.

We must know something about where we've been in order to understand where we are, and where we might be going. 

Profs. Erik Brynjolfsson and Andrew McAfee
A 98 page book appeared around January 23rd of 2012 titled Race Against The Machine. In a stunning move by an increasingly lame TV institution, 60 Minutes actually did a segment about technological unemployment that Brynjolfsson and McAfee had warned about, and allowed them on as talking heads. The segment featured much footage of robots in factories doing the work that humans previously did. Famous AI/roboticist Rodney Brooks is shown with one of his robots that can learn, pick up an object from the floor, works cheaper than a Chinese factory worker, can be programmed to do a new task by a human in a matter of minutes, etc. 

Everyone should have been talking and writing about Tech Unemployment after this, but few did. I think most of the population is clueless and in reactionary mode, while the Owner Class would rather the population not know what's going to happen to them. It was right there on your beloved teevee, people!

Here's the 14 minute segment, in case you missed it. That's Brynjolfsson in the pic.

In a blog post after the 60 Minutes piece ran, McAfee complained that other experts had misunderstood what they were saying. Near the end of the post he writes:

Previous waves of automation, like the mechanization of agriculture and the advent of electric power to factories, have not resulted in large-scale unemployment or impoverishment of the average worker. But the historical pattern isn’t giving me a lot of comfort these days, simply because we’ve never before seen automation encroach so broadly and deeply, while also improving so quickly at the same time.

Now: These guys are not my heroes. I've read their stuff. I object to their avoidance of talking about the human questions of suffering under continuing austerity and the defunct neoliberal economic model. In the 60 Minutes piece McAfee is asked about the human fallout, and he acts befuddled, saying only that "science fiction" is his best guide. Maybe he'd get too much crap from colleagues if he brought up Universal Basic Income? If you look at McAfee's blog there's nothing there about what to do about human suffering (that I could see), and in his book with Bryndolfsson they stress more "education" and "entrepreneurship," which I find tin-eared, or just plain stupid. Look at the education system NOW, look at the debt...and where are these new jobs that people would be "educated" to do going to come from? Servicing robots? What a joke. You just spent a dense 90 pages writing about the inexorableness of machines in the workforce. Fer crissakes! Read the Keynes essay from 1930! (Maybe if you rise so far in academia that you teach at M.I.T. [Bryndolfsson], or Harvard Business [McAfee] you aren't required to address ideas of human suffering?)

Interestingly, if you read the comments to McAfee's blog post I linked to above, the UBI is mentioned. 

Matthew Yglesias of Slate is pro-UBI, but thinks the idea of permanent technological unemployment is a myth...because in the past when new tech revolutionized production, it created new jobs. Here's McAfee's rebuttal. I find McAfee persuasive here. Do you?

Worse than McAfee, to my eyes, is Bryndolfsson's TED talk . How wonderful! The solution to technological unemployment? Work alongside a robot with advanced Watson-ability solutions! Because...it worked in chess. 

This is just pathetic. I applaud these two geeks for pointing out the obvious rapid influx of technological unemployment. They are just silly asses when it comes to what to do about the human fallout, in my opinion.

                                    a still from Fritz Lang's 1927 film Metropolis

Profs Carl Benedikt Frey and Michael A. Osborne
Both of Oxford. On 17 September, 2013 they produced a paper, "The Future of Employment: How Susceptible Are Jobs To Computerisation?" They were motivated by a 1933 paper by Keynes, about the possibility that machines will put most people out of work. They also cite Bryndolfsson and McAfee. They say 47% of all jobs in Unistat are at high risk of evaporating under computers/automation/robots/better AI systems...within the next 20 years. Round it out to 16 years and change, just to make it interesting?

Which jobs are susceptible to loss?
-data crunchers
-production labor
-office support/administrative support
-machine operators

Let's not even talk about booksellers, journalists, musicians, travel agents and a bunch more - who still exist! - but...you know what I mean?

Some things that could slow or speed up the loss of these jobs: regulation of technologies as they come online, and access to cheaper labor. In a paper by Frank Levy of M.I.T. and Richard Murnane of Harvard they address the types of jobs that will be lost to robots: "Each of these occupations contained significant amounts of routine work that could be expressed in deductive or inductive rules and so were candidates for computer substitution and/or offshoring." 

A.I. has gotten better and better at pattern recognition/machine learning and crunching Big Data, so a lot of clerical and administrative jobs are on the way out. 

Computerization will hit a bottleneck or technological plateau, then A.I. will be so good that it will replace most of the jobs in management, science, engineering, and even (this one really gets me) the arts. 

Look at the jobs not susceptible to automation. They mostly suck; the post-war boom and middle class labor movement seem a thousand years ago. The jobs that are hard to replace with a robot are low-wage: buildings and grounds maintenance, housecleaning, food preparation (although I've seen robots on video...nevermind), personal services like doing manicures and haircuts, personal care of the elderly (although I've seen videos of robots doing this work...nevermind), or any job involving abstract, unstructured cognitive work that's hard to write code for. And even with these jobs, software like Network Manager is often used.

Frey and Osborne advise more education to do the sorts of jobs robots can't do: "Acquire creative and social skills," they say. Is it me or is this just fucking ridiculous? It's almost worse than Brynjolfsson's "work side by side with the robots!" Just acquire social skills! Just learn to be more creative! 

Do these academics ever leave the Ivory Tower and talk to strangers in the streets? Message to Frey and Osborne: you were spurred to write your paper by a 1933 paper by Keynes. Please re-read his 1930 essay on "The Economic Possibilities For Our Grandchildren," and get real: work was, for the most part, something we as a species should try to cure. It's a malady. We have in our sights the cure. It's all about sharing the wealth enough so we can not be burdened with biosurvival anxiety and drudgery. And you'd be surprised how many of us know how to handle leisure. We will still "work" although we may not consider it that. Work may "be" play, but it will be productive. And how many boons have come to humanity when people saw some little problem that needed to be overcome, had the time to tinker, to "screw around" and made a contribution to humankind? Answer: almost all boons...

According to Frey and Osborne, only those occupations that require a high degree of creativity or "social intelligence" or other advanced skills can resist the rise of AI. I saw one paper - my notes are scattered so I can't say where - but two jobs that will last for awhile were (I'm not kidding and if any of you challenge me on this I will find the source to prove it): CEO and poet. 

This is where we're headed. And sooner rather than later, friends. It's time to think about what Life is for. Is it to compete in the rat race so you don't have to live under a freeway overpass? Or is money different than wealth?

(this is more than 14 months old, so it's "worse" [or better?] than this):


tony smyth said...

Yeah, interesting stuff. Ages ago I used to teach English in Fanuc, a japanese company that makes yellow industrial robots. This was early 1980s.In one class I asked if they could learn anything from America and they said 'no, we've learned everything we can'. This was the 80s!!!Thatcher came over to Japan and visited the Mt Fuji factory, where robots built robots. She wanted them to build ships in the UK, but had to be told that 'nope they couldnt do that'.

Theres a great book on the same lines as this blog called 'The lights in the tunnel' by Martin Ford.His particular focus is on, "What if technology progresses to the point where a substantial fraction of the jobs now performed by people are instead performed autonomously by machines or computers." This will inevitably lead, Ford and others in his camp believe, to an "extreme future scenario with 75 percent unemployment."

I read somewhere RAW writing about everyone should be a basic living wage, regrdless of if working or not. This of course would be anathema to Republicans, Tories and others, but eventually its going to come to that - either that or revolution and/or a fascist state (and we seem to heading to the latter fast anyhow!!).

Anyhow, Fords book ties well with this, though many may not like his prescriptions to solve the problems robots will cause.

Interestingly,robots are not considered a threat in Japan, but a means of relieving drudgery. Mind you Japan is a rapidly aging society with a declining number of young workers.

Now i have to prove I'm NOT A ROBOT in order to post this!!!!

Anonymous said...

What I find hilarious is that the
economic models and political ways
of the past have to struggle with
the rising tide of the future.
Where's Marxism when there are no
more workers left. Maoism without
any peasants. Capitalism with no
money coming in or going out. Will
we wind up in Cauty future where
everyone is a cop and everything is
against the law ?

Every one who starts to think puts
part of the world at risk, and we
seem to have risked it all at once.

I'm glad to know my poetry will
save me, if I can just offend a
bit to secure a spot in jail for
the necessities, I'll still have
a job to do.

Excuse me if I don't buy that Art
is in that great a danger from the
machines. SRI or Burning Man will
show you a compatible future.
Commodified art on the other hand
should be taken over by machines
to rid us of the pretensions of
those who practice it.

I'm assuming the SF referred to by
the academic is Mack reynolds who
had a workable system (except for
how we get there from here).

There's a big distinction between
a 'futurist' who is unconscious of
being full of shit and an SF type
who knows he isn't predicting any
'real' but is playing with ideas
to see how they work out.

We're never going to run out of
work, every new generation needs
to be taught about the world and
no one has ever learned enough
to say they know it all. The real
question is how to get the tokens
used to participate in the real
economy ?

Expecting the morally bankrupt
Republicrats to solve this problem
is like teaching a pig to fly.
They can't even understand why
Net neutrality existed at all.

It's also highly doubtful that a
system like fascism can exist if
there is no middle class base to
support it.

The current system of paying for
spying could be extended to pay
anyone who kept an eye on their
neighbors, it would also keep them
busy. Whether this is useful is
highly suspect.

It is quite clear that this does
need to be debated and publicized
until it has become an issue for

Compressorhead solves the RIAAs
problems, no more prima donna rock
stars and the promoboys keep all
the money from a $30 album on a
3cent CD. Given their tin ears a
tin artist should sound just fine.

Cleveland Okie (Tom Jackson) said...

I've been waiting to get my hands on a copy of Tyler Cowen's latest book, "Average Is Over," which predicts that income inequality will rise and that a small minority of people will prosper who are good at working with machines and computers.

Psuke said...

I suspect that many Unistatians are unconcerned about the rise of the robot because most of those shops have already be sent overseas, and all they can get now are the shitty jobs no one's figured out how to outsource (or automate) *yet.*

I confess I am somewhat ambivalent about automation because I'm curious how it is going to play out alongside of our diminishing resources, re: oil, copper, iron etc. Or perhaps they'll teach the robots to forage their own building materials from the trash heaps. Now *that* I'd like to see.

I feel we are in for a wild ride, and so many people (not just this continent) are going "lalalalala I can't hear you!" Or are futurists who say "We'll just upload ourselves and live virtually!" (Yes, I knew someone who advocated that as a solution. And he meant it.)

michael said...

@Tony Smyth: expert on Japanese culture and Fukushima and many other things:

Thanks for bringing up Martin Ford; I'd only recently become aware of him. Someone on some blog responded to something Bryndolfsson or McAfee wrote and cited Martin Ford and the Universal Basic Income idea, which obviously must happen.

It seems to me the basic work is in getting people out of those neurological ruts in thinking about little things like value, wealth, money, and what it means to live. And you're right: RAW was a proponent of the UBI, and other isomorphic ideas.

I find the Japanese cultural relationship with robots totally fascinating. And I have as yet not said outright that I'm very Japanese in this way: I think they're great! It's not their fault (or their makers' or programmers' faults that our stupid economic system puts my beloved humans in jeopardy. The only other problem I have with them is their - very real, as far as I can see - potential to get too good as warriors and kill us all. Aside from that, they're perfectly lovely.

That riff about proving you're not a robot made me laff out loud, literally: hey man: I'm sorry: if I were better at manipulating this computer, I'd tell it to let Tony Smyth just post with impunity.

michael said...


Your tone about all this harmonizes with mine really well: Yea, it's owners of the means of production vs. Would-Be Workers now, so where is the Marx? The 1% win it all! Yea, we get it: you're all better than us because of your billions. You're...just great. How cool it must be to be you. Gosh I wish I had worked my whole life in cutthroat competition in order to show I'm an Alpha! But I didn't. Good for you! You won!

Now: can we get real? Or do you want to make this a Blade Runner world? A worldwide prison planet? Really? This was your vision? If so, not all that impressive.

And good points about Art. Try as I might, I don't see it. How do we know most of that commodified art we see now isn't already done by algorithms?

I haven't delved far enough into Andrew McAfee to see what he means by science fiction, but even Heinlein once worked for Upton Sinclair's campaign to end poverty, and was for awhile a proponent of guaranteed annual wage ideas. From what I've seen, SF is teeming with ideas like the UBI. Edward Bellamy's 1888 Looking Backward was influenced by Henry George's alternative economic ideas. Today I read a couple of articles about Mack Reynolds, so thanks for mentioning him. (After reading the entry on "economics" in the Clute and Nicholls Encyclopedia of Science Fiction, the bulk of economic ideas in SF have run more towards the Social Darwinian ends of libertarianism (fuck the poor!).

In Clute/Nicholls Brian Stableford has an article on "money" in the history of SF. George O. Smith's 1945 novel Pandora's Millions sounds interesting: It is about "a desperate race to find a new symbolic medium of exchange following the invention of a matter-duplicator." This reminded me a little of 3-D printing and moreso the promise of nanotechnology...From that same article: "Jack Vance has been particularly ingenious in the invention of various monetary systems appropriately or ironically adapted ti different cultures."

I'm trying to gather some solutions to "paying for spying."

When you say your poetry will save you...always feel free to link to whatever you want here!

michael said...

@Tom Jackson-

I read a few reviews of Cowens's book and get the feeling he's very well-fed and secure in his own biosurvival. But I haven't read his new book.

michael said...


Here's the amazing thing: manufacturing jobs are coming back to Unistat! But: there's hardly any jobs in those factories, because of robots/automation. Shareholders LOVE this!

About the human suffering? Talk to the Invisible Hand.

Robots will not only forage for recyclable metals, but they will MINE them. There are already prototypes.

My impression is that most people don't understand at all what's happening, and think robots are pretty cool when they see them on some 75 second bit on Eyewitless News. There seems a small percentage who assume robots will take someone else's job, but not theirs.

I remember reading Hans Moravec in 1988: we will download ourselves into silicon and be immortal and blast off for the stars! Those who choose to remain meatware and not upgrade will face harsh consequences, much like being the retarded kid on the kindergarten school playground.

I covered him a bit here:

michael said...

Anon had written:

>There's a big distinction between
a 'futurist' who is unconscious of
being full of shit and an SF type
who knows he isn't predicting any
'real' but is playing with ideas
to see how they work out.<

This seems like a great point to me. And in my tone and bias (I admitted a few hrs ago in one of the comments around here that I think robots are pretty cool) is a projection, based on what I perceive in the acceleration of the digitization, the numbers from the "jobless recovery" (stock market through the roof/real unemployment at around 14% and stagnating), my impressions from the work of roboticists and what they say they can do (almost universally: you ain't seen nothin' yet), and a bunch of other factors.

I know Popper's work on the poverty of historicism and the peril of pretending you've got a special silver ball to predict the future. (RAW called Popper's stances on those Platonists over the past 2300 yrs "fundamentalist historicists").

Playing with ideas is my thing, like a SF writer. I find something All-Too-Human in futurist guys like Kurzweil and Moravec. They're both unfathomably brilliant, obviously. But they act like True Believers, almost zealots, when they talk about What Will Certainly Happen Because of Certain Laws of Technological Growth.

I don't want to come off as some sort of fundamentalist futurist, dystopian bend. Let us not forget the hidden good of what robots have done already, esp in the elimination of dire, nasty jobs....like bomb squad detonator.

Do I think technological unemployment is going to get a lot worse? Yes, that's how I figure it, knowing I could be wrong. But the suffering already: human caused and needless. That's why I may seem railing.

But yea: Kurzweil/Moravec rhetoric is potent, heady stuff. I'll drink some and check it out. For thinking about possible scenarios, give me science fiction...written by humans.

Psuke said...

I read that post (shall I confess I've been on a reading jag of your blog? Playing catch up since I discovered it only a few weeks ago, thanks to the Summer of Lovecraft post.) I agree that Kurzweil and the Singularitarians come off as True Believers, and I try not to argue Religion.

Are the robots mining waste dumps? Or just delving into the dangerous (and dwindling) world of mining virgin materials? Because that is heading towards a very dead dead end.

I am still curious how the ultra-wealthy think that having robots replace workers to cheaply manufacture items practically no one will be able to afford, (because, well, no money) is a growth strategy. Or maybe they'll just transfer the money in between themselves and the rest of us can come up with some alternate solution while their backs are turned.

Anonymous said...

Deathless Poesy

On Reincarnation

I haven't made a dime
on poetry this time


We can think our way out of any
problem once we get our back
against the wall, the solution
isn't obvious. History shows too
many examples of what usually is
the way out. Volcanos take down
the dominant civ, Minoan, Roman.
No thinker predicted Greeces
rise to philosophical prominence.
No one imagined the Feudal patron
model rising from the darkness of
wandering tribes scrounging for

The rhetoric about intelligence
seems to overlook that hampered
by the nanny State any change is
considered anathema because it
will effect some vested interest.

My money is on those who feel,
because if you can feel what you
do you will stop doing what brings

Convince the rich that selling
people money is a good business
plan and all our problems will
disappear. just keep an eye out
for the unintended consequences
of any solution.

Eric Wagner said...

Great blog as usual. I remember Travis McGee's friend Meyer had a boat called the John Maynard Keynes.

I would like to pay off my bills. I don't know about meta-solutions for the whole culture. It strikes me that many of my heroes had trouble paying their bills, from Robert Anton Wilson to Anthony Braxton.

I wonder if income inequality will resolve itself non-violently. What will The Invisible Hand do?

I also think some of us have trouble with leisure. When we do have time on our hands, we have trouble unwinding and making use of it. We still have a lot to learn from "Bob."

On other matters, I don't know of any good non-classical reference books. I've used The Rolling Stone Encyclopedia of Rock and Roll as a textbook, but it came out in 2004 and it doesn't blow me away.

I did just start a new blog for Christmas: http://zukofsky.blogspot.com/ .

tony smyth said...

Just by coincidence (or not) this was in a column in Todays Irish Times:

The rise of the 1 per cent in the Robotic Age
Rising global inequality may not just be a cyclical phenomenon
Rising global inequality may not just be a cyclical phenomenon; something structural could be going on. The rise of the 1 per cent might be a function of the emerging Robotic Age: there is little that can be done, now or in the near future, that can’t be done better by machines. Economic recovery, such as it is, will not generate nearly as many jobs as in the past. A percentage point of GDP growth is not the job creating machine that it used to be. Dystopian visions of the future of work see the activities of 99 per cent of us split between being either unemployed or writing apps that enable the 1 per cent to lead rich and fulfilling lives.
These fears are almost certainly exaggerated.

michael said...


In the Land of Oz, it seems robots are mining metals in waste, but it's hard to get a line on exactly what's going on, because I guess I don't know enough about the varieties of mining coupled with waste. O! My ignorance knows no bounds, truly!


For me, reading Singularitarians is important because their ideas represent how extreme things might get; they might turn out to be "right"; they give all sorts of insight into various aspects of sophisticated ways humans project in a visionary way; even if their forecasts are terrible and represent some gnostic variant in the human mind about "rapture" or "transcendence" their thinking contains tons of hard science; and finally, I agree with Stewart Brand who once said he follows the fringe movements to see "where the center might go," something like that.

There's a documentary on Kurzweil that I found quite moving: he loved his dad and his dad died far too young, and Kurz wants to do something to bring him back. It's almost painfully poignant. He's clearly a genius. His elaborations on Moore's Law represents one way to compare RAW's riffs on the Jumping Jesus Phenomenon...only RAW was less of a True Believer in what "will" happen by, say, 2030.

Your query about robots and the 1%: I have not seen ONE good answer for this yet. Lenin said the history of capitalism will result in imperialism, because capital must always be in search of new markets. Maybe they can raise the standard of living enough in Africa in order to market First World goods to Africans?

I don't see this an an adequate answer, and I had to stretch a lot just to put it forth.

michael said...

@ Prof. Wagner-

I've added your Zukofsky blog to that stuff over there ------>

Your riff on people not knowing how to handle leisure is right in my wheelhouse. John Dewey had a term: "occupational psychosis," which meant that people spend so much time and psychic energy at their jobs that, even when they have time off, that's all they can talk or think about: their co-workers, office politics, who's being treated fairly or not, what someone said, what's wrong at work, etc. They don't know what to do with themselves when they aren't expending their life-energies in making other people wealthy.

Chomsky said one of the reasons democracy in Unistat is such a sham is that people are so overworked (but not really complaining about it out of fear) that they come home exhausted, try to understand what's going on by watching TV "news", feel the need to accept what they see as "probably right", then drink and go to bed, year after year until retirement...in which they often die, because they never learned how to manage their actual non-wage-slave time in something that stokes their internal fires of creativity.

Up until the 20th century, "liberal arts" were THE royal road to learning how to be truly human..."leisure time" was the time to realize your human capabilities. The word was changed by PR and advertising types into something like the time to loaf in a La-Z-Boy, drinking beer and watching football. Consuming stuff: that's "leisure"!

It's been a loooong time since I've had a job with regular hours (it looks like I'll never have one again), but I'm never bored. Never. I may be emotionally "down" a lot because of perceived lack of self-actualization, but I regularly hear cracks from friends that they wished they had me as a teacher in high school or university, or other kind things.

I'm NEVER bored: books, writing, guitar playing, listening to music, talking with friends, watching old movies, hiking, doing yoga, meditating, doing research, bicycle riding, following sports ironically, following my own gnostic maneuvers.

In typing out this little rantish thing I realized this "problem" seems another Missing Public Discussion.

michael said...

@Tony Smyth-

"These fears are almost certainly exaggerated..."

Let me guess: what followed was a variation on "when we've had technical revolutions the jobs didn't go away....therefore it will happen again: these robots will create new jobs."

Maybe not. But I'll have something to say on this soon. I think it's wrong. (Obviously)

tony smyth said...

Yep, I think he's wrong too.Juts struck by the fact thta it came up in the Irish Times right after reading your piece.
According to Martin Ford, contrary to what mant people think, robots will replace not just blue collar, spanner and wrench type jobs. ers a link to a blog robot's potential effect on employment and society.

tony smyth said...

Oops, wrong link. Should read roboticnation.blogspot.com. "My brain is going Dave, I can feel it. Can I sing you a song?" "Go ahead HAL". "Daisy, daisy etc......"

michael said...


Not only did (seemingly?) no one notice those epochal changes, but another way to look at our democracy is that some of us see what's up ahead, but we're marginalized, demonized, persecuted. Then, when that which was warned about comes to pass, only a few notice that "back then" there were some who'd warned about it. We stumble into a crisis, then muddle our way out of it. David Runciman says it better here:

This gets to something you seem to be hinting at, and which is a major theme in RAW's work: the persecution of heretics. If you're a mad genius and have some heretical idea you're fucked...unless you happen to be operating at just the right historical moment.

I liked the lines "My feel is on those who feel..."

Re: unintended consequences: let's take those as given. They WILL occur, just as sure as every medication has side effects.

michael said...


Oh yes: Marshall Brain's robot site. Dude started the site How Things Work then sold it for $250,000,000, which forces my reserves of envy to the surface.

Gotsa read me some dat Martin Ford now...