[ExI] The Robot Big Bang

Anders Sandberg anders at aleph.se
Sat Feb 21 23:54:30 UTC 2015


Carsten Zander <Carsten.Zander at t-online.de> , 21/2/2015 4:12 PM:We should establish the term "ROBOT BIG BANG" 

...
The Robot Big Bang will have the following characteristics: 
- Robots will become very cheap and will spread rapidly around the world. 
- All simple activities can performed by robots. 
- Most people will lose their jobs to robots. 
- All people will need a basic income. 
- Robots themselves will be produced by robots. 
- Robots will transmit their skills, knowledge and abilities to other 
robots. 
- All people will be able to produce most things on their own with the 
help of robots and 3-D printers (3-D printers are like robots). 


Hmm... why do you think all these things have to come to be together? Some are likely causal: cheap robots that can do most activities would likely lead to a lot of lost jobs - but note that "simple activities" does not cover most jobs actually done. Robot production does not imply micromanufacturing, and so on. 


Still, the population of robots does increase rather rapidly.
http://earlywarn.blogspot.co.uk/2012/04/global-robot-population.html
However, the data it builds on is pretty chunky:
http://www.worldrobotics.org/uploads/tx_zeifr/Charts_IFR_PR__04_June_2014.pdf
At least the industrial robotics market happens in fits and starts, making exponential extrapolations risky. Figure 3 of 
http://www.unece.org/press/pr2001/01stat10e.html
also shows that quality adjusted robot prices have been declining since the 90s while labour costs have been increasing - but since the task abilities are so different, it is pretty hard to pinpoint a time when the curves will actually cross for real in generic tasks. 


So it seems reasonable to expect a near future point with more robots than humans (but we already have more microprocessors than humans). A more important one would be when the cost (weighted by the job skill pool) of robots becomes smaller than labour costs - but it will be a rather messy thing to measure since the job skill pool also changes as a response. So I expect the robot big bang to make sense from a sufficient distance just like the industrial revolution, but close up there is little to see. 


Understanding these complications and that there likely is a big automation shift matters. As does explaining it properly to decisionmakers. I am a bit worried that right now it turns into a simplistic "The robots are coming, so we need basic income", which means some politicians will immediately accept or dismiss it depending on their views of basic income, and hence deduce that robots are either a problem or not a problem...


Anders Sandberg, Future of Humanity Institute Philosophy Faculty of Oxford University
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20150222/c9abdb54/attachment.html>


More information about the extropy-chat mailing list