The Dim-Post

August 29, 2016

Why I don’t fear the robot apocalypse

Filed under: Uncategorized — danylmc @ 10:45 am

Being in Christchurch made me realise how reliant I am on Google Maps whenever I’m out of the tiny patch of Wellington I’m familiar with. Maps doesn’t really work in Christrchurch – every time I tried to use it the application lead me to a giant construction site or into the middle of a vast, empty temporary carpark the size of a city block and then told me ‘You have arrived at your destination.’

31 Comments »

  1. Now you know why the hand-picked Nat-friendly audience, there to provide polite applause at appropriate junctures to John Key at the last election, unexpectedly burst into raucous laughter when they were reminded how well the Chch rebuild was going.

    Comment by McNulty — August 29, 2016 @ 11:33 am

  2. I’d have thought your experience should make you fear the robot apocalypse. By the sounds of it, Google Maps already is trying to kill you by directing you into the path of heavy construction equipment.

    Comment by Andrew Geddis — August 29, 2016 @ 11:38 am

  3. Being in Christchurch is probably a lot worse than being in any robot apocalypse.

    Comment by eszett — August 29, 2016 @ 2:43 pm

  4. “3.Being in Christchurch is probably a lot worse than being in any robot apocalypse”

    I think you meant W(h)anganui?

    Comment by Clunking Fist — August 29, 2016 @ 3:03 pm

  5. The truth is that Google maps doesnt allways have a clue exactly where you are from GPS, as they say
    “How Maps finds your current location
    Maps estimates where you are from sources like:

    Your web browser’s location information.
    Your phone’s location, if you have Location History turned on.”

    Use a proper in car navigation system instead instead of having your every move kept with the bowels of the deep state.

    Comment by ghostwhowalksnz — August 29, 2016 @ 3:39 pm

  6. I use Google maps nearly all the time. It pays to be skeptical about its results. They are always interesting though, as an insight into algorithmic thinking.

    For instance, last time I was driving back from Rotorua to Auckland, Google diverted me at Maramarua off the main route home on state highways 2 and 1, and into a bizarre mountain range crossing via Hunua. On a whim, I decided to follow it. Sure enough, Google was accurate in its assessment of how long that winding hilly alternative to the motorway would take. I can’t know if the motorway was blocked up enough to justify such a detour – it was something that I can’t imagine any human ever thinking of except maybe some people who lived in the area. It probably cost me a fair bit more in brakes and petrol. But it’s probably correct in that it was a few minutes faster.

    Comment by Ben Wilson — August 29, 2016 @ 4:12 pm

  7. Ghost – the deep state doesn’t give a flying fucking about where you go for lunch or how to sell you shoes based on your location.

    Comment by Gregor W — August 29, 2016 @ 5:33 pm

  8. That’s a common misconception. Actually, the deep state exists purely to further the interests of the shoe vendor industry. I read all about this on Infowars.

    Comment by Flashing Light — August 29, 2016 @ 6:29 pm

  9. Don’t knock Google maps, that little blue dot on your phone showing your location is possibly the only decent peace dividend of all the trillions spent on cold war satellite navigation and targetting technology.

    Comment by Sanctuary — August 29, 2016 @ 7:31 pm

  10. The “robot apocalypse” is really a “rentier apocalypse”. And when Stephen Hawking says as much…

    http://www.huffingtonpost.com/entry/stephen-hawking-capitalism-robots_us_5616c20ce4b0dbb8000d9f15
    http://motherboard.vice.com/read/dont-fear-the-robots-taking-your-job-blame-the-monopolies-behind-them

    CF #3: Did that happen to be when Michael Lhaws was its mayor?

    Comment by Kumara Republic (@kumararepublic) — August 29, 2016 @ 8:43 pm

  11. Flashing Light – that’s nonsense. Everyone knows the Phoebus cartel diversified into GPS after compact fluorescent bulbs cut their margins too much.

    Comment by Conrad — August 30, 2016 @ 9:47 am

  12. Kumara, so it seems we binned Universal Basic Income too soon? Again…

    Comment by Clunking Fist — August 30, 2016 @ 9:53 am

  13. CF – the trouble is, the ‘universal’ bit would also apply to our robot colleagues, lest in be discriminatory.

    Comment by Gregor W — August 30, 2016 @ 10:12 am

  14. I was going to suggest slavery, but the internet never forgets. So may I be the first to welcome our new robot.txt overlords…

    Comment by Clunking Fist — August 30, 2016 @ 12:48 pm

  15. The robot apocalypse I fear most is not robots becoming like humans and taking over our work. It’s the one that is actually happening, where our work becomes more robotic so that robots can take over it. Humans turning into robots (and then being discarded), not the other way around. This is the real path to the robopocalypse. It’s worse than the other robot apocalypse, in which at least we get robots doing cool human things. In the real robot apocalypse we get humans dumbing down to be as stupid as robots, taking away functions that we used to have and replacing them with crap systems.

    Comment by Ben Wilson — August 30, 2016 @ 2:30 pm

  16. In the real robot apocalypse we get humans dumbing down to be as stupid as robots, taking away functions that we used to have and replacing them with crap systems.

    Dark Satanic Mills 2.0

    Comment by Gregor W — August 30, 2016 @ 3:11 pm

  17. There’s plenty of technological development that doesn’t fall into the category of the faux-robopocalypse, and power plants would be a good example. But this bankrupt vision we so cleave to of removing ourselves from the picture all the time, and that somehow being better, I find …. Unoriginal? Unlikely? Poorly thought out? There are so many cool things we can do with technology, things that do genuinely transform our world. Why do we always fall back to the sad metaphor of a robotized version of ourselves, a stink slave, the kind of servant you wouldn’t even want because they’d have to be micromanaged. This vision itself leads to our continually under-rating what it is that we are actually good at, and seeking ways to remove even that, to tell ourselves we don’t need it. We just have to engineer what we do until it’s so simple even a cretinous robot could do it, and then hey w00t! AI! Machine intelligence!! Not ridiculously high levels of human intelligence applied at doing the kind of things even the simplest person can do, only not quite so well. Call it machine intelligence and feel the sci-woowoo. The singularity is coming! I fear it in the way I fear that my next robo call to the bank will be just that little bit more annoying than before. With the sad resignedness at how it’s now my job to understand what the bank teller used to do, so I can explain it via key presses to a dummy. For *my* benefit, apparently.

    Comment by Ben Wilson — August 30, 2016 @ 3:26 pm

  18. “so I can explain it via key presses to a dummy. For *my* benefit, apparently.”*1

    But your bank will employ AI so you need only say “I want to increase the limit on my credit card please” and it will know exactly what you want. Change bank, bro.

    I’m not sure what or where these “simple jobs” you talk of, are. I cherish the robotic parts of my jobs, as a diversion from the bits where I have to think really hard about really complicated things.

    Comment by Clunking Fist — August 30, 2016 @ 4:37 pm

  19. >I’m not sure what or where these “simple jobs” you talk of, are

    Well, for example, following a simple set of verbal or written instructions, something that most people can manage fairly rapidly without needing an engineer to program them to do it.

    >But your bank will employ AI so you need only say “I want to increase the limit on my credit card please” and it will know exactly what you want. Change bank, bro.

    No thanks. I’m not about to change something as complicated as my banking just to get a idiot robot that can understand a limited number of perfectly enunciated phrases so that they can hire less actual people. As I said, personally, I don’t actually want to be the one who has to change just to enable a robot servant to save the bank money that they certainly will not pass back to me in any form. That’s the opposite of customer service, the opposite of progress. It’s me accepting the bank externalizing their cost back onto me in the form of wasted time.

    Comment by Ben Wilson — August 30, 2016 @ 5:24 pm

  20. “……..the only decent peace dividend of all the trillions spent on cold war satellite navigation and targetting technology.”

    So the iron curtain was really to keep all those Western Europeans from tasting the delights of real, existing socialism.The Soviets and their allies just couldn’t keep up with the demand so had to ration it to just those who were in their immediate grasp. Truly, we have been living a lie. You must be rapt at Putin’s ambition to put it all back together again.

    Comment by Tinakori — August 30, 2016 @ 6:26 pm

  21. @Ben Wilson
    “The robot apocalypse I fear most is not robots becoming like humans and taking over our work. It’s the one that is actually happening, where our work becomes more robotic so that robots can take over it.”

    On that sort of topic, Marshall Brain’s short story Manna is worth reading if you haven’t yet.
    http://marshallbrain.com/manna1.htm

    Comment by Kalelovil — August 30, 2016 @ 9:38 pm

  22. I’m not about to change something as complicated as my banking just to get a idiot robot that can understand a limited number of perfectly enunciated phrases so that they can hire less actual people.

    You two are 5-8 years behind the times. All that can be done via the bank’s presence online, from the comfort of your own home, without ever having to speak to a robot or a person.

    Comment by Phil — August 31, 2016 @ 9:19 am

  23. Maybe they have very unusual banking requirements?
    A.

    Comment by Antoine — August 31, 2016 @ 12:07 pm

  24. “Well, for example, following a simple set of verbal or written instructions, something that most people can manage fairly rapidly without needing an engineer to program them to do it.”
    Yet more and more jobs need edjacashun, like degrees and stuff. I guess you could argue that I could follow a simple set of instructions to prepare a set of audited financial statements and a tax return. After all, it must all be laid out in those accounting standards and the income tax act. Simples, really. Could you please mail me a set of instructions on increasing revenue by 15% next year?

    Comment by Clunking Fist — August 31, 2016 @ 12:16 pm

  25. Could you please mail me a set of instructions on increasing revenue by 15% next year?

    Increase spending by 30%.

    Comment by unaha-closp — August 31, 2016 @ 1:12 pm

  26. An anecdote related to me many many many etc years ago.

    A Japanese technologist was walking around a car factory in the UK and was accompanied by the Managing Director and the local Shop Steward (Well, it was a long time ago). They were looking at all the many tasks that humans were performing to construct the automobiles as they crawled along the conveyor-system pretty much unchanged since Henry Ford devised it.
    “By this time next year” said the Japanese industrialist, “all of these tasks will be undertaken by robots.”
    “Yes,” replied the Shop Steward, “But the problem with that is, that robots don’t buy cars.”

    Every so often I encounter people who excitedly visualise the future of warfare as fought by robots, Terminator style. I usually remind them that humans are much cheaper to develop, nurture and are much more expendable. Using robots is a costly exercise. Consider the cost involved using a human bomb to deliver a payload to a crowded airport, compared to the cost of a drone. Suicide bombers are much more cost-effective.

    Finally, I’d like to suggest that there are certain jobs that are rendered more necessary and more valuable as a consequence of the increasing isolation and alienation that accompany technological improvements, standardised technological society. These include literacy and numeracy tutors, educators, probation officers, social workers, politicians and of course prostitutes. (I added the prostitutes part for frisson, I’m pretty sure a robot might be a able to do the last one, I just don’t dwell on it that much, because you have to draw the line somewhere.)

    Comment by leeharmanclark — September 1, 2016 @ 7:39 am

  27. >All that can be done via the bank’s presence online, from the comfort of your own home, without ever having to speak to a robot or a person.

    For almost everything I do, yes. For anything I typically want to call them for, no. I wouldn’t even call if it was something simple. But it’s always good to have the obvious pointed out as a personal flaw of mine in this kind of discussion. Restores my faith that nothing has really changed. The fact that I’ve been using internet banking since last millenium need not get in the way of missing the point. It’s a fairly normal strategy when discussing the technological shortcomings of systems to turn it into an ad hominem about the personal flaws of anyone who won’t “get with the program”. That’s been an excellent strategy for decades, when facing anyone who looks at a system change objectively and notices some ways in which it’s actually got worse when seen through the eyes of a customer, rather than eyes of a developer or shareholder.

    Comment by Ben Wilson — September 1, 2016 @ 9:10 am

  28. Yeah. New roads, temporary roads, construction…. all tend to be behind in Nav / Map applications, understandably.

    In one my cars the 3 year old Nav system knows nothing about the new Waikato Expressway and gets very excited as I race across paddock after paddock at 99.75kph.😉

    Comment by truthseekernz — September 1, 2016 @ 10:38 am

  29. Yeah, the Soviet bloc states did terrible things: building walls to stop people moving countries, banning people with suspect politics from jobs” and having only one newspaper that supports the government.

    Terrible, it was…

    Comment by richdrich — September 1, 2016 @ 2:55 pm

  30. I like to think a robot AI or GE apocalypse would be a way for self aware intelligence to continue its evolution and avoid the now almost inevitable homo stupidicus extinction event; where we mismanage our ecosystem and fight each other with toxic war machines for the last scraps of agriculture production. It seems that the qualities needed for a leadership that can enact positive outcomes for all humanity and the planet are totally beyond human psychology. I often contemplate this while drinking yeast based beverages.
    Individual consumer grade computers are now capable of out performing human brains in both linear and parallel processing operations. it seems obvious to me that what we lack the most is cognitive grunt in political leadership. I’m voting for the first AI candidate I see on a ballot paper.
    My intelligence and that of my ancestors was limited by the relationship of brain size to the birth canal. Now we have cesarean section and soon genetically enhanced progeny, perhaps it is possible that human intelligence may encounter some kind of More’s law. With support from information technology an exponential increase in human cultural capabilities may give us the ability to chart a course for survival before the robots take over, or nothing.

    Comment by 3ccels — September 2, 2016 @ 9:43 pm

  31. human intelligence may encounter some kind of More’s law

    I doubt it. Intelligence isn’t measured in flops. Its measurement is not really very scientific at all. It’s not a well-defined term. Which is part of the reason people can assert that robot intelligence is inevitable, near, or already has happened. It all comes down to what you define as intelligent. People working in AI 50 years ago were calling the machines doing what they doing then intelligent. So long as you reduce intelligent activity down to an actual definition of the specific solving of specific tasks, then it is always achievable, so long as those tasks are specific enough that it doesn’t require intelligence to interpret the outcomes. Which is necessary, of course – you’d have a circular definition otherwise. But that reduction task is a very large chunk of the job of programming a computer. In some languages it’s almost all of the job. So actually providing the definition of intelligence in that way requires solving what intelligence is, and you get a problem whose inevitability can’t even be calculated until it’s finished. In computer programming this is usually summed up in the answer to “how long will it take you to program this idea of mine?” which is: “how long is a piece of string?”. Until the person asking the question has actually got specific enough about what they want to program to do, accurately calculating how long it will take to do that is essentially impossible. And getting that specific can often be most of the job.

    In the case of this particular piece of string, it’s at least as long as the entire history of computing – so far. How long is the rest of it? You tell me. If we’re going on progress toward early stated goals, progress has been glacial. If we get far more specific about the actual tasks we have solved, progress has been extremely fast. But in absence of our ability to know how much further there really is to go, we can’t tell if we’re approaching light speed, but setting course for Andromeda.

    Comment by Ben Wilson — September 5, 2016 @ 11:03 am


RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Blog at WordPress.com.

%d bloggers like this: