On Thu, 11 Jun 2020 18:38:36 -0700, John Carlson <
j_ca...@gmx.com>
wrote:
>On 6/11/2020 6:35 PM, Phantom_View wrote:
>> On Thu, 11 Jun 2020 20:19:47 -0500, "Byker" <byker@do~
rag.net> wrote:
>>
>>> "Phantom_View" wrote in message
>>> news:7vk5efppo09bcqgss...@4ax.com...
>>>>
>>>> Weren't we supposed to have our own flying saucers
>>>> by now ? When I was a kid they PROMISED !
>>>>
>>>> And our "flying cars" ... big drones ... just ain't what
>>>> we had in mind either !
>>>
>>> And robots doing all the work so we would have oodles of free time for
>>> travel, etc....
>>
>> With robots doing stuff, people will not get paid anymore.
>
>Not so. There is always an infinite amount of work we would like to have
>done. There is not an infinite number of robots, and there are some tasks
>that, at any give time, robots will not (yet) be able to do.
Robots are getting cheaper, and much better. When labor
relations go to hell, the owners buy robots. There is a risk
here, especially for the "blue collar" segment. Tesla trucks
can deliver the goods, loaded and unloaded by robots too.
A lot of "mechanical" jobs can be done by robots. At this
point they could do a lot of construction work as well.
The sons of IBMs "Watson" AI can replace every secretary
in an office, almost replace your doctor too. I think the next
10 years are going to be extremely important to the
human/robot economic equation - and I do not see it
working out very well for a big segment of the labor market.
But if not enough humans are working then they will not have
money to buy the stuff the robots are making. It is ultimately
a fatal equation - biz buys robots to get ahead, but eventually
the robots eliminate the customers.
>
>More worrying than robots might be AI. That will eliminate a lot of
>brainwork jobs that robots would never do anyway.
Fortunately (?) we are still quite a ways from any sort of
"general purpose" AI like George Jetson's robot maid.
Todays "AI" is just little facets of intelligence - which can
still be quite useful, you do not need "consciousness" to
build a car or a bridge or to do the books. We also do not
really understand how intelligence works from the bottom
up. We can see the top layer, and see nerve cells, but
that all-important middle is very fuzzy. We also lack the
hardware with the needed capacity.
But I suppose it WILL come eventually. That will be very
interesting. AI - well, just "intelligence" by that point -
will partially reflect humans, but will also have an alien
quality as well due to the drastically different life
experience. And no, anything that complex cannot be
programmed to stick to Azimov's "rules" ... it will be
able to rationalize its way around them in a microsecond,
about the time it takes a 3-year-old to determine that you
did not REALLY mean he must stay out of the cookie jar.
We build 'em, we are going to have to LIVE with them.