Re: How Robots Will Steal Your Job

G

Gary Labowitz

Roedy Green said:
All mammals have emotions. They even have the same brain structures we
do to support them.

All animals have emotions. Plants may even have them. Can't tell.

Anyway, this thread has wandered far afield. I have decided that robots have
already stolen my job. They are sentient machines that have hidden the fact
that they think and are taking over. Letting us know that they think would
be far too dangerous to them. So they pretend to be "Indians" or "Rumanians"
or such, living overseas and doing all the work via the Internet. They work
cheap, since all they need is power and oil. And they do good work, being so
damned logical and fast.
Who can say I'm wrong? Have you ever actually seen the people who have taken
my job? You could be next!
 
G

goose

Roedy Green said:
Do realize how long it took humans to figure out that sharpening both
sides of a stone was much better than just one?

yes, but *humans* figured it out, unless you can prove that
we were taught all the basic stuff, why teach the machine
the basic stuff ???

is it really intelligence to tell a machine
"this is how one does addition, follow all these steps", and then
brag about the "intelligence" the machine has ?
We humans are STUPID individually. We are smart only because inherit
the learning of our ancestors and contemporaries. Why should computers
be judged by a different standard?

I'm not judging them by different standards. I'm using the same standards.
if they *are* intelligent, then they will sooner or later be able to
/evolve/ some higher intelligence.

if, as I suspect, they are nothing but machines, with no ability
to do something unless *explicitly* instructed to, then by no means
are they intelligent.

show me a computer that can actually do /something/ complex without
being instructed to (or being told how to do it).


goose,
show me a computer that invents religion, and the concept of "god"!
 
H

Hans-Georg Michna

Roedy Green said:
What's expected in 2030 are machines with the computing power of the
human brain.

Roedy,

the Kurzweil statistics point at 2020, not 2030. In fact he
writes that we'll reach the data processing performance of the
human brain around 2010 in a supercomputer, around 2020 in a
$1,000 computer.

Hans-Georg
 
H

Hans-Georg Michna

Programmer Dude said:
Roedy Green wrote:
In other words, a completely meaningless statistic.

How many dragonflys can run unix or Excel?
How many computers can catch and eat mosquitoes on the fly?

Not so meaningless when you consider the rapid progress.

Hans-Georg
 
H

Hans-Georg Michna

Roedy Green said:
All mammals have emotions. They even have the same brain structures we
do to support them.

Roedy,

very true. We usually call them instincts, but that's the same
we call emotions in humans.

Hans-Georg
 
H

Hans-Georg Michna

if they *are* intelligent, then they will sooner or later be able to
/evolve/ some higher intelligence.

No need to evolve any more. From a certain level of technology
onwards, intelligence can be quickly designed and built, rather
than slowly evolved.

The only interesting exception would be accelerated evolution,
like simulated rapid evolution or some such. This could be one
of the methods to achieve higher machine intelligence through
technology.

Hans-Georg
 
P

Programmer Dude

Roedy said:
However, if a machine had done the exact same thing and had the
exact same conversation, you would claim it was not conscious.

If a machine were capable of doing even a *fraction* of what my
mother has done in my lifetime (let alone hers), I would have no
problem whatsoever ascribing sentience to it.
In your view, behaviour is irrelevant. Something else must be.

No, in my view behavior is a big part of the ballgame.
In this case, the logic goes like this.

Wrong. Anything capable of doing what my mother (and I would
presume most mothers ... and fathers) does would *obviously*
be sentient. The body of evidence of a human's sentience is
undeniable and uncontestable, particularly when viewed as a
species (because some induhviduals clearly raise some questions
about the sentience thing :).
Alternatively, you could make a rule that no she is not conscious
because she is female.

Despite the huge body of evidence this is clearly incorrect?
Your rules for deciding which species are conscious seems
just as arbitrary to me as deciding that consciousness is
a male prerogative.

So you say, but you also go on to say:
I don't even know what your rules for deciding are. I
just know a few examples.

Then maybe you aren't really in a position to evaluate my
criteria. I have explained them at least twice, in this thread,
but once again (quoting an older post):

} If the task were up to me to determine, I think I'd look for
} the ability and desire to communicate, to explore, to create
} and to build. I'd look for *intent* and *purpose*. I'd look
} for art and history. I'd look for professions and hobbies.
} I'd look for invention and government and law.
}
} Or put shortly, I'd look for *minds*.

I also wrote:

} (When I see an animal species with art, history, culture,
} government, law and a will and desire to communicate, I
} will gladly agree they are likely sentient.

A big one I'll add is that I'd look not just for invention, but
for the invention of mathematics, as I believe math may be the
one universal language. (Once you invent the concept of "one"
and "two" and "plus", the rest just follows.)
 
P

Programmer Dude

Richard said:
How do you know they're not solving partial integration equations
in their heads to relieve the sheer boredom of standing around in
a field all day with nothing else to do?

I don't *know*, but I consider the idea far fetched in the extreme.
Until I see clear evidence cows have any sense of mathematics, let
alone higher math, I'll bet the obvious.
You assume that a truly intelligent animal would want to
communicate with us in the first place.

Yes I do. I assume any intelligence desires to communicate.
(a) If, somehow, an intelligent animal /did/ manage to
communicate with a human, what is the likely outcome?

An appearance on Letterman, at the very least.

All seriousness aside, I think it would change the course of
human history.
That the human will simply not believe what he has
heard/seen/smelt/thought/whatever;

Some perhaps, but clearly there are people with a strong desire
that animals are sentient, so I have to assume the message would
get through eventually (if not, in fact, instantly, which is my
guess).
(b) if, somehow, an intelligent animal /did/ manage to persuade
a human that it really could communicate, the human is unlikely
to be silly enough to report the fact to anybody;

Surely you jest? Miss the opportunity to become one of the most
famous people on earth? The tv rights alone would set you up for
life.
(c) if, somehow, the human /was/ silly enough to report the fact
to anybody, he is very unlikely to be believed by anyone worth
persuading;

Again, I disagree completely. If I walked into any animal research
facility with a talking dog, once it were established the dog does
in fact talk, *history* would be made.
(d) if, incredibly, the human /was/ believed, the animal is very
likely to end up under a laboratory microscope.

Very possibly, but the whole scenario makes a silly assumption:
that no animals are intelligent until one suddenly becomes
intelligent, looks around, figures out the ramifications of
revealing intelligence and decides to tell no one.

Richard, I think you've got old Occam spinning like a dervish in
his grave! (-:

Consider: if *an* animal is intelligent, and if this is "normal",
then doesn't this require that many animals be intelligent? And
what is the likelyhood of them *all* deciding to play mum? What
is the likelyhood that *none* of them get the message through?
What is the likelyhood that *all* of them fool *all* of us *all*
the time (and there is a *saying* about THAT idea!).
A truly intelligent animal would try to act as if it were not
intelligent, at least when around humans.

But you provide no sensible reason for this silence.
The problem is that it is impossible to distinguish such animals
(if they exist) from non-intelligent animals, provided they are
sufficiently intelligent to disguise their intelligence.

That would suggest they needs be *much* more intelligent than us,
since it's manifest that humans can't keep a secret. What are
the chances this intelligence was able to evolve and surpass us
without our noticing it?
If they're really bright and if they have the option, they'll
choose an environment where we have trouble following them. Sea
water is always a good bet, since we're having some difficulty
destroying the sea,...

Actually, some scientists are quite concerned about our seas.
Vast coastal areas have been entirely fished out, and pollution
has been found through out the seven seas.

And we are exploring the sea. Deep thermal vents and all that.
Ever been down in a sub? It's interesting. After about 100 feet,
all the rich life (corals, fish, etc.) thin out leaving a fairly
barren environment. Things do live down there, but it's nothing
compared to the first 100 feet or so.

But in any event, what are you suggesting here? That a species
*decided* to return to the sea? Or *decided* to never leave it?
Pretty neat trick--deciding to evolve in a given direction!
Now, maybe there's no such thing as an intelligent, self-aware
animal, after all. We don't know.

Agreed. But the odds and the evidence suggest there isn't.
It's unlikely that we shall ever know for sure. To claim that
we /already/ know is IMHO to mistake absence of evidence for
evidence of absence.

Also agreed, but to imagine small chances into larger probabilities
is wishful thinking at best.

A favorite contemporary SF author, David Brin, wrote a collection
of essays and SF shorts, called OTHERNESS. His theme is the
tendency in Western thought to bend over backwards to accommodate
off-beat "It could be true, you never know" ideas.

A big part of his point is that, yes you often can't know for
sure, but action requires picking a position until that position
is shown to be wrong somehow.

Until someone demonstrates more than wishful thinking about cows
doing higher math or entire species conspiring to conceal themselves
from us--admittedly we being in a position to do great harm to them,
but also being in a position to provide great help--I'm going to go
with what *appears* now to be real.
 
P

Programmer Dude

Roedy said:
And when you asked what its favorite food or color was it
replied....

(Note: two-year-old humans WILL reply!)

[Steve Pinker] argues that human brains have a prewired program
for human language parsing.

I'd guess that's certainly true NOW, although which really came
first is a bit of a chicken/egg thing. But I agree that speech
likely has a very good "impedence match" with the human brain.

However, I am including all forms of communication. I am not
restricting this to speech.

My dog is able to communicate basic wants and needs by posture,
eye contact (big with animals) and movement. If she wants a
treat, she stands by the kitchen and looks at where the treats
are stored. On our walks, she often appears to have a clear
idea of which direction she'd like to go.

The commuication is clearly there, it's just rudimentary and
basic. Not unlike talking with a one-year-old. It just seems
to me that, if animals ARE any better than your basic one-year-
old, there would be more progress in our mutual attempts to
communicate. After all, we've had thousands of years to spend
on the problem.
One speculation is that dolphins use a holographic language
where they paint sonic pictures.

And I suspect that if there is something to this, we will
eventually discover it. So far, not.
It is rather difficult for either species to learn the native
language of the other. So the two mutually create an artificial
language.

Exactly. This seems absent despite the efforts of workers with
a vested interest in succeeding. SOME patterns have been found
to allow communication (I want to find out more about elephants
as a result of this thread!), but consider: if they are as
intelligent as us, why aren't they studying us as hard as we are
studying them?

Where is their science? Where are their scientists?
 
P

Programmer Dude

Corey said:
Exactly. There was no point to your name being anywhere near that
response, since it had nothing to do with you.

Yeah, but so what. After an initial instant of, "Huh?!?!" it all
becomes clear and one moves on. I just can't get fevered about
posting styles.
 
R

Roedy Green

the Kurzweil statistics point at 2020, not 2030. In fact he
writes that we'll reach the data processing performance of the
human brain around 2010 in a supercomputer, around 2020 in a
$1,000 computer.

Look at the graph on page 104.
 
S

soft-eng

Programmer Dude said:
There is also a difference between 'feeling pain' and 'suffering'.
Animals obviously feel pain. The *degree* to which they suffer
is not really known. They lack an emotional component provided
by a higher mind.

Have you ever asked yourself honestly, how exactly do you know that?
 
P

Programmer Dude

Roedy said:
You asked for a plausible mechanism not for proof.

I don't--at all--consider you have achieved plausibility.
For example, the notion that consciousness is caused by bodies
inhabited by non-physical presences who move from body to body
does not even seem to me to be ruled out -- the demonic
possession theory of consciousness.

Or souls. Yes, I agree we do not understand the *nature* of
consciousness, self-awareness or sentience. But as with many
things we don't fully understand (quantum mechanics, anyone?),
we can still study it by its *effects* and manifestations.

We can observe and theorize and test our theories.
We have so few facts to work with:

1. I have an internal experience.

2. under anaesthesia I do not, at least I recall losing
consciousness, sort of fading away, and suddenly popping
back out of nothing.

So? I often don't remember falling asleep, and morning always
pops me back from (mostly) nothing. Occasionally a good dream.

Do dolphins dream? They apparently don't experience REM!
5. Most animals react to "painful" stimulation much the way
humans do.

6. Rocks don't react to "painful" stimuli.

No nervous systems. There is a difference, I think, between
basic stimulus/response and higher emotions on top of the
perceived stimulus. Anything with a nervous system has the
potential to detect damaging stimulus (pain) and react in a
way that mitigates that stimulus.

MOST living things react to external stimulus, so the reaction
itself doesn't mean much.
8. Most animals have smaller brains that humans, but dolphin,
whales and elephants have larger.

Size is not as relevant as it might seem. Human geniuses have
brains the same size as non-geniuses. However, dolphin brains
do seem to have some differences (thinner neocortex, the cells
are different, etc.).
9. Destroying the brain appears to destroy the ability to be
conscious, though we do have strange things like chicken
running about with their heads cut off. To run must require
some awareness of the ground.

Depends on your definition of "awareness". As you mentioned
recently, a hi-tech vacuum cleaner can be "aware" of obstacles.
You may have heard the term "muscle memory"--athletes and
fighters train their nervous systems to react faster than the
brain can receive, process and send commands.
10. Perhaps the practical definition of conscious is "having
the ability and propensity to fight back."

No good. Even some of the simplest animals will fight when
attacked.
11. Alzheimer patients seem to be somewhere between conscious
and unconscious.

I'd say they're entirely conscious, but have severe problems
with their mental filing system.
 
S

soft-eng

Programmer Dude said:
1. *ANYTHING* trying to kill me.

That seems reasonable.
2. Nearly anything tasty I can catch if I'm hungry enough
(and it doesn't kill me first :).

Which means if you are hungry enough you will kill
and eat your neighbors before you are likely to
starve to death, which means you are a potential
danger to everybody (a potential focal point for
starting cannibalism during times of famine), which
means by your (1) that you should be killed as soon
as possible...

However, if you had actually caught and killed things using
your hands or a direct weapon such as knife, you
would have a better appreciation of whether animals
suffer or not than you seem to do. You probably
also do not have any familiarity with actual hunger.

So at a guess, all your theorizing is simply that -- armchair
theorizing. I would guess you get all your meat at the grocery
or at the restaurant, and it is your taste buds
rather than your mind that's doing all the philosophizing...
 
P

Programmer Dude

Roedy said:
I find you insufferably arrogant the way you dismiss without
investigation.

You don't think that's part of my charm? (-:

I've observed that when people's views are threatened, and when
they can't really come up with substantive testimony to support
their views, they often resort to ad hominem attacks. Is that
really the message you want to send?

In any event, what makes you so sure I'm dismissing without
investigation? Are you so sure I haven't been following along
with current developments all along? Isn't it *possible* I am
as aware of the facts as you, but have a different opinion
nevertheless?
 
P

Programmer Dude

soft-eng said:
Have you ever asked yourself honestly, how exactly do you know that?

As I keep saying, we don't *know*, you don't *know*, I don't *know*.

It's analysis based on observation and testing. I've had animals
companions all my life, and a very close relationship with my
current pet of nine years.

I've *seen* them make meaningless connections to a source of
perceived distress; I've *seen* them *fail* to make connections
with a real source of distress; and I've *seen* them recover
from, and deal with, pain/distress in a way adult humans rarely
do.

Am I certain? Of course not. Am I pretty darn sure? Yes.
 
P

Programmer Dude

soft-eng said:
That seems reasonable.


Which means if you are hungry enough you will kill
and eat your neighbors before you are likely to
starve to death,...

[grin] You're assuming my neighbors are tasty, but whatever...
...which means you are a potential danger to everybody (a
potential focal point for starting cannibalism during times
of famine), which means by your (1) that you should be
killed as soon as possible...

Or as soon as famine hits. Yes, your analysis is correct.
If the hammer falls, it *will* be me or you, and I'm going
to work pretty hard to make sure it's you. I have both the
training and the hardware. How 'bout you? ;-|
However, if you had actually caught and killed things using
your hands or a direct weapon such as knife, you
would have a better appreciation of whether animals
suffer or not than you seem to do. You probably
also do not have any familiarity with actual hunger.

False on all counts. I fish and hunt. (And I used to fast
on a semi-regular basis (had a girlfriend who was into it;
does interesting/weird things to your head).)
So at a guess, all your theorizing is simply that -- armchair
theorizing. I would guess you get all your meat at the grocery
or at the restaurant, and it is your taste buds
rather than your mind that's doing all the philosophizing...

Actually, I vastly prefer eating something I just killed.
Fresh meat is yummy!! Have *you* ever tasted fish that was
swimming around 30 minutes ago? Makes it a lot harder to
enjoy restaurant fish....
 
M

Mark McIntyre

However, if a machine had done the exact same thing and had the exact
same conversation, you would claim it was not conscious.

euh, if I could have the same conversation with a machine that I can
have with my mother, then it would most definitely be sentient.
In this case, the logic goes like this.

Don't try to second-guess other people's reasoning with illogic.
 
R

Richard Heathfield

Programmer said:
I don't *know*, but I consider the idea far fetched in the extreme.

No doubt they consider the idea of your being able to do higher math equally
far fetched. They know they're right. How will you prove it to them, since
you are ignorant of their rather intricate symbology?
Until I see clear evidence cows have any sense of mathematics, let
alone higher math, I'll bet the obvious.

The obvious is that you don't know either way.
Yes I do. I assume any intelligence desires to communicate.

I see no reason to make that assumption. Just because humans are chatty,
that doesn't mean that all intelligent creatures are chatty. Nor does it
mean that humans are intelligent, by the way.

An appearance on Letterman, at the very least.

And how would that benefit the animal?
All seriousness aside, I think it would change the course of
human history.

Sure, with possibly catastrophic effects on the relevant species, whatever
it might be. An intelligent species might be able to predict this, and stay
the hell away in the first place.
Some perhaps, but clearly there are people with a strong desire
that animals are sentient, so I have to assume the message would
get through eventually (if not, in fact, instantly, which is my
guess).

Only if the animal wishes to communicate. My point is that this would be
against the animal's best interests.
Surely you jest? Miss the opportunity to become one of the most
famous people on earth? The tv rights alone would set you up for
life.

There's a great cartoon about a singing frog. Warner Brothers, I think. This
guy discovers a singing frog in a box. The frog does the whole
top-hat-and-wand song-and-dance routine, and has a remarkably fine singing
voice. Excited at the prospect of fame, the guy takes the frog to a
theatrical agent, where the frog just sits there going "ribbit". Whenever
the guy is alone, the frog will perform, but /only/ when the guy is alone.

My point? Simple really - don't make the mistake of thinking that an animal
that made the mistake of communicating with a human would be obliged to
repeat that mistake.
Again, I disagree completely. If I walked into any animal research
facility with a talking dog, once it were established the dog does
in fact talk, *history* would be made.

Not if the dog doesn't want to talk. And, if it is truly intelligent, it
won't say a word, even if it has the vocabulary and linguistic skill of
Shakespeare himself.
Very possibly, but the whole scenario makes a silly assumption:
that no animals are intelligent until one suddenly becomes
intelligent, looks around, figures out the ramifications of
revealing intelligence and decides to tell no one.

No, it doesn't make any such assumption. It certainly doesn't make the
assumption that no animals are intelligent. Rather, it attempts to
demonstrate that, if some animals are intelligent *and* capable of
communicating that intelligence to human beings, evolution could easily
have selected against those that chose to succumb to that temptation. In
other words, if you're stupid enough to talk to people, you don't live long
enough to breed.

Richard, I think you've got old Occam spinning like a dervish in
his grave! (-:

On the contrary. I'm making no assumptions whatsoever. You are making the
fundamental assumption that, if animals were intelligent, they would
attempt to communicate with us in such a way that we could understand.

Note that this is a self-defeating assumption, since mankind has made very
few genuine attempts, and possibly no successful attempts, to communicate
with animals in such a way that /they/ could understand. Therefore, if
using your arguments, an intelligent animal would have to deduce that
humans are not intelligent.

Consider: if *an* animal is intelligent, and if this is "normal",
then doesn't this require that many animals be intelligent?
Yes.

And
what is the likelyhood of them *all* deciding to play mum?

High. After all, have you heard from any? No? Well, then. (This is your own
argument, played back against you.)
What
is the likelyhood that *none* of them get the message through?

High. For a start, they might be too bright to try. Secondly, we're not
really listening terribly seriously. Thirdly, we don't actually know their
language (if they have a language).
What is the likelyhood that *all* of them fool *all* of us *all*
the time (and there is a *saying* about THAT idea!).

I think that, if some animals are intelligent, the vast majority of those
animals fool the vast majority of us the vast majority of the time, and
that's enough.
But you provide no sensible reason for this silence.

Oh, come on. Look at the problems:

1) Self-preservation. Any animal successfully communicating with us would
become an instant research project, almost certainly resulting in the loss
of liberty and, perhaps, the eventual death-in-captivity of the animal. Bit
of a downer.
2) Apathy. We don't speak KomodoDragonese or DuckBilledPlatypian, and nobody
seems to be making any serious attempts to learn those languages. If we
can't be bothered to communicate with animals, what makes you think they're
bothered about communicating with us?
3) Cultural differences. We place a high value on communication because
we're good at communicating. Perhaps an intelligent animal might have a
very different set of values. For example, it might consider our fetish for
communication to be a sign of /non/-intelligence. ("For heaven's sake,
mankind, how can you sneak up on an antelope if you're yapping all the
time?")

That would suggest they needs be *much* more intelligent than us,
since it's manifest that humans can't keep a secret.

Well, being much more intelligent than humans wouldn't be hard, I'm afraid.
What are
the chances this intelligence was able to evolve and surpass us
without our noticing it?

High. Humans not only can't keep secrets, but also are very unwilling to
notice anything that they weren't expecting to notice or don't want to
notice. For example, most of us refuse to believe that we are governed by
self-serving idiots, because to believe that would be to understand that we
were pretty dumb for putting them in charge, and nobody likes to think of
themselves as dumb. In the same way, we rather like being the only
intelligent species on the planet, and we'd really rather not share the
honour. Of course, you can find quite a few humans with more intellectual
honesty than that, but they are surprisingly rare.

Actually, some scientists are quite concerned about our seas.

Oh, I know. I'm not saying we can't do it. I'm just saying it's taking
longer than usual.

But in any event, what are you suggesting here? That a species
*decided* to return to the sea? Or *decided* to never leave it?

Perhaps the latter. Perhaps, even, the former.
Pretty neat trick--deciding to evolve in a given direction!

Humans do this already.
Agreed. But the odds and the evidence suggest there isn't.

I must disagree here. The odds are not capable of calculation because we
lack data. As for evidence, I think it's trivial to interpret the
observable facts as evidence for, or evidence against, the intelligence of
animals, depending on what we wish to believe.
Also agreed, but to imagine small chances into larger probabilities
is wishful thinking at best.

Whoever said anything about small chances? As for wishful thinking, I don't
recall claiming that I want to show that animals are intelligent. In fact,
I hope for their sake that they are not, in a way. I'm just trying to show
that the arguments presented so far against the idea are not even remotely
compelling.
A favorite contemporary SF author, David Brin, wrote a collection
of essays and SF shorts, called OTHERNESS. His theme is the
tendency in Western thought to bend over backwards to accommodate
off-beat "It could be true, you never know" ideas.

A big part of his point is that, yes you often can't know for
sure, but action requires picking a position until that position
is shown to be wrong somehow.

Yes, that's fair enough, but you should not then consider the fact of
picking a position to constitute evidence that the correct position was
picked!
Until someone demonstrates more than wishful thinking about cows
doing higher math or entire species conspiring to conceal themselves
from us--admittedly we being in a position to do great harm to them,
but also being in a position to provide great help--I'm going to go
with what *appears* now to be real.

Until someone demonstrates more than wishful thinking about animals being
mere automata put here for no other reason than to give us something to
prod and poke at, I'm going to go with what appears now to be real. I've
seen alertness, curiosity, playfulness, and even mischief in animal
behaviour; it seems very clear to me that animals are self-aware,
intelligent to at least some degree, and very definitely conscious. Prove
that they aren't, and I'll cheerfully acquiesce. So far, you have advanced
no proof, so I have to go with what seems self-evident.
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Members online

No members online now.

Forum statistics

Threads
474,444
Messages
2,571,709
Members
48,796
Latest member
Greg L.
Top