h a l f b a k e r yAlmost as great as sliced bread.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Syntho-God
A reasonable facsimile of a devine oveseer. | |
It already exists in its rudimentary form. You can ask
Alexa
or Siri if it's going to rain tomorrow and you'll get an
answer. But for the daunting quesitons in life: "What's my
greater purpose?", ""How should I live my life?" etc, we
create a universallly accessible datebase called "Syntho-
God".
That's
the easy part. Here's the tricky part. The tenets of
this robot-diety are all created by popular up-vote.
Now in this devided world we live in it's hard to imagine
all
cultures would agree on anything, but that woudln't be
necessary, they would agree on some things and that's
where this universal truth computer would start.
It may be slow at first. The ruler of the universe in my
favorite series of books, "The Hitchikers Guide To The
Galaxy" said a lot of "I don't know." and "On the other
hand
it could be this way, I'm not sure..." which science (the
other way we get information) does a lot of as well.
It might start off as simple as "Do not be a cannibal", "Do
not slap your grandmother." and build from there.
After decades of the whole planet agreeing on stuff it
might bring us together to some extent. Which may or
may
not be a good thing.
Now here's the good part. You could create an actual,
taylored to you, godlike entity that is specifically
taylored to your personality and your life. It would be
programmed to
look out for you, to be your gardian angel and assistant,
just like god. It could give you an education, help you in
times of need with good, generally agreed to life advice
and just be your all around digital best buddy.
Syntho-God (tm). "Maybe not as good as, but probably
more real than the real thing."
Project Pope
https://en.wikipedi...g/wiki/Project_Pope Prior Art [8th of 7, Oct 13 2017]
"The Last Question"
http://multivax.com/last_question.html Well-worn territory [RayfordSteele, Oct 13 2017]
From The Gadfly -- one of the most famous books in the atheist Soviet Union
https://www.goodrea..._Voynich_The_Gadfly [theircompetitor, Oct 13 2017]
Emergency Faith Pack
Emergency_20faith_20pack Prior Art [8th of 7, Oct 13 2017]
Apple syntho god
https://youtu.be/QRH8eimU_20 Apple agent of the lord [mylodon, Oct 13 2017]
Microsoft syntho god
https://en.m.wikipe...ki/Office_Assistant I am being unfair here as i am sure bill hoped for more [mylodon, Oct 13 2017]
Space power tools
http://www.popularm...by-nasa-astronauts/ [mylodon, Oct 14 2017]
Creating life.
http://www.telegrap...ut-playing-god.html Too clever for our own good? [doctorremulac3, Oct 16 2017]
Another approach apparently
https://www.dailyst...on-killer-robots-AI Mine is a sort of universal mind, this is just some sort of robot devil for mindless worshipers. [doctorremulac3, Nov 19 2017]
True Love
http://www.angelfir...a/savvy/story7.html An (almost) personal-god Multivac story [Skewed, Nov 20 2017]
All The Troubles Of The World
http://www.mcguirem...f_the_world_(1).pdf Another Multivac story, shades of Minority Report. [Skewed, Nov 20 2017]
Person Of Interest
https://www.youtube...watch?v=WYDWSNMTauQ A bit more recent. [Skewed, Nov 20 2017]
Earth, the TV Show
https://www.youtube...watch?v=wK-IuIbfb-A [doctorremulac3, Nov 23 2017]
Lycurgus
https://en.wikipedi.../Lycurgus_of_Sparta For some reason, there is no Lysergus [pertinax, Nov 26 2017, last modified Oct 08 2021]
A fly's brain
https://www.mpg.de/...ng-motion-detection A fly would destroy humans at any sport if they knew how to play. And didn't get squashed by the ball. [doctorremulac3, Nov 27 2017]
More AI god ideas.
https://www.dailyst...n-way-of-the-future Didn't read the article, think I got the basic idea from the headline. [doctorremulac3, Dec 11 2017]
The Machine Stops
http://archive.ncsa...ajlich/forster.html Always make sure your self-repair mechanism is capable of self-self-repairing. [Wrongfellow, Dec 26 2017]
Was only a matter of time.
https://futurism.co...sJxryVzEVZBnL9qmrFc [doctorremulac3, May 16 2023]
The AI alignment problem
https://en.wikipedi...g/wiki/AI_alignment [Voice, May 30 2023]
Please log in.
If you're not logged in,
you can see what this page
looks like, but you will
not be able to add anything.
Destination URL.
E.g., https://www.coffee.com/
Description (displayed with the short name and URL.)
|
|
What if your grandmother is a cannibal? |
|
|
I assume the syntho-god is called Roland? |
|
|
//What if your grandmother is a cannibal?// |
|
|
Then Syntho-God starts repeating "DOES NOT
COMPUTE! DOES NOT COMPUTE!", starts smoking
and blows up. |
|
|
We're still working out the kinks. |
|
|
You could, if you wanted, program your specific
Syntho-God portal to be very neurotic or goofy,
caring or thoughtful, strong and sure of itself,
whatever you were comfortable with. I
for instance might be more comfortable having a
SG entity that's very practical. "Synth-God, what's
the meaning of life?" "Oh Jesus, not this
conversation again. Right now your car is due for
an oil change, let's go with that." |
|
|
// "Synth-God, what's the meaning of life?" // |
|
|
THAT... would be a given. |
|
|
A default answer should be: |
|
|
"INSUFFICIENT DATA FOR A MEANINGFUL
ANSWER" |
|
|
Except that it's not. Go back and actually read the idea. |
|
|
Remember, everybody is 99% atheist not believing in most
of
the gods that have been worshiped through history. |
|
|
There may be somebody who believes in all of them, I just
haven't heard of such a person. |
|
|
This one will actually answer when you pray to it. |
|
|
Anybody reading this who gets comfort or solace from their
God, kindly ignore this post. I'm happy for you. (Unless your
god tells you to kill me.) |
|
|
I did actually read the idea. |
|
|
I didn't say that all parts of it were baked, but that the basic
territory of creating a technological answer-machine is. |
|
|
Not trying to pick a fight, so not sure why the attitude. |
|
|
"He was the sort of person who stood on mountaintops during
thunderstorms in wet copper armour shouting 'All the Gods are
bastards." (Terry Pratchett) |
|
|
Why not give it a try ? After all, what could possibly go wrong ? |
|
|
// There may be somebody who believes in all of them, I just
haven't heard of such a person. // |
|
|
Inspired my injectable eucharist that works like an epi pen for
when you've only got seconds to choose between salvation and
eternal hell-fire. |
|
|
Ha! Would you sell them in packs of two? |
|
|
Would they have a "use-by" date ? |
|
|
I am interested in this territory, and i have read this a few times, and
it is very well worn. i think it encompasses some of the hopes and
desires man had when man created machines. However, not well
implemented in current world today. Popular upvote mechanism as
a means for weighting
a neural net may be the most unique part but already done
functionally even in a google search if a link is considered to be a
validation or upvote of a sort. But maybe working that Idea througH
so it works better as a mechanized means of democracy. Anyways.
The devil is in the details. |
|
|
I recently read a David Brin novel with a variation of this idea as it's premise. In the end a human intelligence is downloaded during death and all of the various internet chatter and opinion ratings washing over this augmented intelligence act much as random thoughts bubbling up from the subconscious, competing and collaborating over the thoughts of a normal human mind on a day to day basis. |
|
|
For AI to have a chance to work it needs a dash of chaos thrown into the mix, or there can be no awareness of right and wrong. |
|
|
Let me put it another way: |
|
|
It would be a democratically created god usefully
implemented. |
|
|
The ultimate hive mind. The unversal conciousness
of mankind. The human animal's idea of truth. |
|
|
Or we might find out we have nothing at all in
common except for the frowning on cannibalism
thing. |
|
|
Hey, maybe that's all it would say. "Don't eat each
other. Now leave me alone." |
|
|
Amyway, this idea is about taking already existing
concepts and steering them in a new direction:
Let's build an actual, functional synthetic god
that's an extension of all of us. |
|
|
Just calling it "god" would clarify the vision. |
|
|
Eventually AI will profile every human being at a very young age as to their individual learning bents. It will categorize disease based on visual auditory and chemical clues we can not perceive, make connections and leaps of logic based on memories no human mind could ever hope to compete with, and root out the asswipes currently harshing our multi-cultural buzz. |
|
|
In my estimation AI will never achieve intuition. No amount of free association can mimic or equal intuition without requiring an infinite number of scenarios to do so. That's why we're needed. |
|
|
Only life can create by leaps of illogical intuition. |
|
|
The sum of the multiverse equals consciousness, and the sum of consciousness is what everybody and their dog has saddled with the moniker "God" in various forms. |
|
|
Do you think that the sum of all multi-universal consciousness really cares which idiosyncratic doctrine any of us talkin-monkeys subscribe to? |
|
|
Treat the next synapse as you'd like the next synapse to treat you, and everything is copacetic. |
|
|
All other rules are extraneous. |
|
|
Why did this remind me of the Synod, Earth: Final Conflict? The aliens ultimate decision tree. They did look like living connected neurons. though. |
|
|
Intuition is exactly what neural nets use. They have nothing but a bunch of data and operate on hunches. |
|
|
There is often no direct logic making decisions; just statistics and connections processed by solving a sequence of equations in the equivalent of the visual center of a computer.. The imagination, the gpu. |
|
|
Logic itself is a discrete component of it but any larger features become blurring indistinguishable from human intuition. Similar in some ways to how exquisitely complex compression has turned digital transmission into something visually similar to old VHS tapes. |
|
|
The problem will be, we will no longer be able to logically control machines, but will need to cajoll, convince, and inspire them to work. |
|
|
Maybe even perform virgin sacrifices. |
|
|
//Only life can create by leaps of illogical
intuition.//
|
|
|
//There is often no direct logic making decisions;
just statistics and connections processed by solving
a sequence of equations in the equivalent of the
visual center of a computer.//
|
|
|
This thread has suddenly become really
interesting.
|
|
|
I've thrown out my take on the life vs machines
thing
before as to motivation, I think it applies to
intuition or other human or life traits as well. |
|
|
I think the fundamental
difference between life and machines is at
the molecular level.
Forgetting for a second about the fact that we will
create biological life someday, life has motivation,
machine does not.
|
|
|
The third smallest component of life is the cell. (I
know there are parts of cells but bear with me)
Atoms, molecules and cells. These cells have
programming, the big picture of what makes life
life. Survive, divide and expand. Living and
motivation is programmed in at the cellular level.
|
|
|
The third largest component of AI at this point is
an incredibly simple tool, a switch that's either on
or off. When we pile enough of these together, we
can mimic anything we want, even motivation,
caring, a lust for taking over the universe, but it's
an illusion. These switches don't care, these
switches aren't motivated, these switches aren't
alive. They're inert, dead, inanimate objects. No
matter how many of them you pile up, they're just
massive piles of these dead, dumb tools. They can
be programmed to look alive, but they're not.
|
|
|
That's why I'm not as worried as the super geniuses
who are afraid of the Skynet scenario where AI
becomes conscious and gets angry and rebellious
for some reason. We're projecting human traits onto
highly modified rocks. AI doesn't care about the
motivation of "taking over the universe" or "wiping
out the humans" because it doesn't care about
anything, and never will.
|
|
|
Somebody might be able to make it look like it
cares, but it doesn't.
|
|
|
We've survived the threat of nuclear weapons in
the hands of mad men which is a much greater
threat in my estimation. |
|
|
You want to worry about something, worry about
that biological life we're going to make someday.
That MIGHT be programmed at the cellular level to
wipe out all other
life. But don't worry about that either. I think the
biggest challenge we're facing now is de-evolution,
but that's a different thread. |
|
|
Well... it might be related. The real threat of AI
might be that it takes care of us so well we lose
the ability to take care of ourselves, H.G. Wells
and The Time Machine hit on this concept with
everybody sitting around like chickens in a pen
waiting to be fed, unable to fend for themselves.
We need to remember that evolution never sleeps.
We evolve to fit our circumstance and if our job is
to consume and reproduce, what's the point of
language, intellect, curiosity, aggression or desire?
These traits that have allowed man to be the most
successful animal on the planet might be lost. Remove
their necessity and what purpose do they
serve? And who's to say we'll even retain the desire to
reproduce? |
|
|
So the real threat from AI might not be that it's
aggressive and destructive, it might be just the
opposite. |
|
|
I sent a shortened version of the above post to somebody
who's pretty famous requesting a response that I could
frame and put in my
office because "it would be really cool". |
|
|
Unfortunately I got this back almost immediately. |
|
|
"Thank you for your email to Professor Hawking. |
|
|
As you can imagine, Prof. Hawking receives many such
every day. He very much regrets that due to the severe
limitations he works under, and the enormous number of
requests he receives, he is unable to compose a reply to
every message, and we do not have the resources to deal
with many of the specific scientific enquiries and theories
we receive." |
|
|
Oh well, I tried. If by some stretch of the imagination he
writes back I'll let you know, but don't hold your breath.
(Still, kind of fun.) |
|
|
I would say that I'll ask him next time I see him, but I don't think it would significantly shorten the delay. Plus he may still be mad at me for almost running him over once. |
|
|
Thank you Max, I was going to ask you but thought it
would be pushy. |
|
|
I'd consider a transcript from any conversation between
you guys worthy of framing. I'll put it next to my NASA
invent the future contest award. (I won some power tools
for inventing a new kind of valve.) Also my gold records.
All the things I have on my wall in lieu of any kind of
diplomas. (Not even the high school kind. Yes, that's
probably
why I'm so uptight about wanting to be considered smart.) |
|
|
Actually... that would go on my desk next to the pictures
of the wife and kids. I would treasure it. Even if his
answer was: "Tell the dumb American to piss off and you
go take some driving safety lessons!". |
|
|
Wait - you have gold records? As "the ones you get for selling X number of copies" as opposed to "the ones you make with gold spray paint and an LP"? Seriously? |
|
|
Don't be TOO impressed. Being a recording artist in my
youth I made some pretty good money and put that
money into what became a very
successful recording studio and production company. The
gold and platinum and
multi platinum records are from the bands that venture
produced. I'm still proud of them. It could be argued they
wouldn't exist if it weren't for me though it's hard to
speculate on such things. I certainly helped. |
|
|
But they're from famous bands and you know them.
500,00 units get you a gold record and 1,000,000 a
platinum in the states. Different countries have different
certification numbers. Anyway, I've got a wall full of
them. |
|
|
Want to hear more about the NASA award? Power tools?
No? OK. Nobody ever wants to hear about the power
tools. |
|
|
Wowww. Now that, [doc], is properly awesome. Actual music that actually sold to people who actually bought it - both yours and that of the bands you produced. Damn. |
|
|
And gowonden - tell us about the valve. |
|
|
I put some money into a production company and I think they bought
office chairs and lunch with it. |
|
|
Back to the machines..... |
|
|
Being too well looked after, as a future apocalypse, at the moment
seems unlikely due to the difficulty of software to last a week without
an update and hardware requiring constant adjustment, charging,
replacement, etc. How can ai repair itself manually or expurgiate
decades of code debt? We are almost impossibly far away from
machines being able to evolve physically by themselves, because they
are all very top heavy and don't have a physical self repairing and
duplicating cells. |
|
|
This may be something we evolve to become very good at (plugging
USB cables into dead devices, swapping out hard drives, developing
dedicated teams to refactor old bugs so they are new and more agile)
maybe so good computers force us to work in camps, maybe in large
buildings stacked up floor after floor of gridlike compartments,
populated with humans dedicated to keeping machines charged,
updated, fed, repaired. |
|
|
//And gowonden - tell us about the valve.// |
|
|
It's called the "gauge pressure dissipation valve".
Temperature regulation valves have an inherent issue
with having to re-direct water that has back pressure that
causes friction on the valve requiring some measure of
force to overcome that friction. |
|
|
My power tool winning design did something new in that
it dissipated the back pressure by spraying the water
through a small gap so it's traveling as a result of velocity
only making it very easy to re-direct. |
|
|
An example of the concept can be given with a garden
hose. Try to re-direct the flow of water by pressing your
thumb against the opening. It's very hard because you're
fighting that back pressure. Now move your palm a few
inches away from the outlet and simply place it in front
of the stream. You've directed the water just as
effectively, but there's very little force needed. |
|
|
It was designed for a temperature regulating shower
head. The stream of water went through a sort of
rotating "key" that had holes in in that lined up at the
correct temperature for bathing but turned into the
water path it it became too hot or two cold. Because
there was no back pressure to overcome, a simple bi-
metal spring could react to the water temperature and
turn the wheel requiring very little torque. The upshot
of it was, it reacted instantaneously so it was impossible
to be hit with even a small amount of water that was of
an uncomfortable temperature. |
|
|
It was a very simple "anti scald, anti freeze" shower head.
Teledyne considered licencing the design having worked
on the problem for years. They thought it was very clever
how I had overcome a problem their engineers had failed
to solve. |
|
|
Didn't go anywhere till I entered it into NASA thing. It was
a big deal to me because it proved (to myself) that I
could do something creative besides music. When I built
the prototype, the moment it actually worked it was
similar to the first time I heard a song of mine on the
radio, kind of an out of body experience. It made me very
happy to put it mildly. One of those "Life's
scrapbook moments." |
|
|
Sheesh, no wonder people don't want to hear about the
power tools. Kind of a wordy story. Anyway, it's a special
thing to me. |
|
|
Don't understand. Splashing water off a hand is different then
stopping flow. Redirecting flow would mean.. The excess hot or cold
would just go straight down the drain? Or do you redirect back into a
reservoir via a low pressure line? |
|
|
Same questions as [mylo]. |
|
|
But it sounds ingenious - had I but a hat, I would doff it to you. Having no hat, however, I shall dedicate my next G&T to you - it'll be in about 3-4 minutes. |
|
|
Im genuinely curious as to what album features the doctor. |
|
|
I can also report that the aforementioned G&T worked very well. |
|
|
//The excess hot or cold would just go straight down the
drain? Or do you redirect back into a reservoir via a low
pressure line?// |
|
|
Oh yea, sorry. Yes, it just turns into a faucet that you
hold your hands under while you adjust the temperature.
Once it's within a range that's comfortable for bathing it
automatically directs the water towards you. |
|
|
I've used it, you can play with the controls turning the hot
water all the way off, then all the way on while turning
the cold water all the way off and it's impossible to ever
get hit with a single drop of uncomfortable temperature
water. There was a buffer chamber for if you were to put
all cold water in, then turn it off and put all hot water in
for some reason, while the key wheel transitioned from
too hot to too cold, as the key passed through the "allow
water" position, that water would be half freezing, half
boiling. Warm. |
|
|
//Im genuinely curious as to what album features the
doctor.// |
|
|
I'm very happy with my anonymity. I assure you, I'm not
that
interesting anyway. Keep in mind, the highlight of my
weekend was sending a letter to Stephen Hawking that
he'll
never read. |
|
|
Just some guy who used to sing and got his 15 minutes
that's
all. Eh, maybe more like 5. Better yet, let's assume I'm
making it all up. |
|
|
Next week, my NFL career. I'll tell how I was a... who's
the guy who throws the ball? Quarterback, for that team
from San Francisco with the gold helmets. Or maybe they
were red. |
|
|
Well, speaking as someone who has (a) never won anything from NASA and (b) never produced anything musical apart from one particularly memorable fart, I am moved to dedicate my third G&T to you also. (The second one slipped in between the first and the upcoming third.) |
|
|
Admit it, [doc], you're actually Ozzy Osbourne in real life. |
|
|
I'm not sure he can operate a computer. |
|
|
But cheers max. G&Ts are what I drank back before I
swore
off doing anything fun ever again. G&Ts and every other
alcohol in existence that is. |
|
|
Boy, really airing my dirty laundry today. Halfbakery
Confessionals. |
|
|
(And that's not true, I still do fun stuff, just nothing that
carries the threat of imminent or lingering death
associated with my old pastimes.) |
|
|
[doctorremulac3] Field-programmable gate arrays. It just has to care about programming itself to a higher and higher level. Thinking only about cells is anthropocentric, in this universe of wonders. |
|
|
Coming to this thread a bit late bit that is cool,
[DrR3] - the Venn diagram of overlap between
gold-disc holders and
NASA engineering competition winners must be
pretty small |
|
|
Hippo, thank you, I never thought of it that way
but I guess you're right. You made my morning,
you're a nice person. I'm going to go out and say
something nice and positive to somebody today in
turn. |
|
|
Ive had way too many friends that
are now multi-millionaires where I'm firmly middle
class to be beaming with 24-7 pride, so it's nice to
have something like that to be
sort of proud of. Christ, my ex-wife married the
co-
founder of one of the first computer stores, had
over 800 outlets and he was worth 9 figures by the
time it was sold so some even bigger company.
Private jet, the whole bit. (I was much better
looking than him though, thank God.) Point is, I've
got plenty
to keep me humble so it's nice to get a pat on the
back. We all need it
every once in a while. I've got a story about the
private jet to
tell sometime. |
|
|
Wjt, I would respectfully say that the
anthropocentric view is that any entity we create
will mimic the specifics of our life form. These traits
like desire to survive, expand, even live evolved to
fill a roll life needed to function. |
|
|
I've also seen the hypothesized knowledge curve
of AI in the future and premise that our place in
the universe might be relegated to the pre-curser
to this superior entity that will look down on us
because we're stupid and blobby. Another
anthropocentric view? |
|
|
My point is, when does this fundamental switch,
from not caring to caring take place? |
|
|
And I dont think we'll ever be dumb enough to give
total autonomous kill ability to our killing
machines. We've had the ability to do this for a
long time and I don't know if this is even on the
drawing board anywhere. |
|
|
I like to rank danger in order of likelihood: |
|
|
1- Car crash
2- Metabolic syndrome based disease
3- Getting caught misspelling words on Halbakery
4- Nuclear war
5- Asteroid hitting Earth
6- AI deciding it cares enough about me to bother
killing me. |
|
|
That being said, we can program AI to do anything
we want. If somebody decides to program AI to
replace biological life, yea. It could give us a good
run for our money. |
|
|
I still propose the real danger is AI replacing our
will to survive by removing all adversity and
challenges from survival turning us into mindless
food processing slugs. |
|
|
I had an idea for a novel. It's the future, everybody
is stupid because of thousands of years of
dependence on this Syntho-God thing, then it
breaks. The hero comes out looking for food after
a big bright light made a big booming sound and
the food stopped coming. (Asteroid hit the Earth
maybe. I'd have to come up with a plausible reason
a mechanism that's worked for thousands of years
would stop working.) Anyway, using rudimentary
language, we follow his adventures re-evolving
into a self-reliant animal again. Title: "Syntho-
God". |
|
|
Anybody done that? Do novels make any money?
Maybe it needs to be a video game scenario,
they're the only entertainment making any money
these days. |
|
|
Last thought on the naughty AI thing, let's not
worry about it, but let's keep our hand on the plug
anyway. Even though I've got some gold records
and won some power tools for designing a valve I
could still be wrong. I know, bit of a stretch but it's
possible. |
|
|
My experience with novel writing is that its
somewhat more likely than winning the lottery, but
somewhat less likely than getting shot, and almost
as pleasant. |
|
|
Regarding the evolution of AI - I think it's inevitable that AI will become smarter than us at some point, but it will happen consecutively in different domains. |
|
|
AI is already much smarter than we are at spell-checking a huge document, multiplying 50-digit numbers or adjusting the hue and saturation in a digital photograph. Of course, because these things can be done by computer, we redefine them as non-intelligent activities. |
|
|
AI is also better than us at playing chess, so we have redefined chess-playing (at least by a computer) as being non-intelligent - AI just analyses more patterns. |
|
|
AI is getting good at face and image recognition. It can already do it much faster than we can, but still makes mistakes. Even so, we are pre-emptively defining image and face recognition as non-intelligent. |
|
|
As AI is developed to include "tricks" that solve more and more problems, so those activities will be progressively defined as non-intelligent. Eventually, this will apply to every activity. |
|
|
The risk, therefore, is not so much that AI will develop ingelligence, but that we humans will cease to be intelligent by defining everything that we (and computers) can do as being non-intelligent. |
|
|
One thing we can all agree on, we put in parameters for a
system to get a job done we better be careful. |
|
|
"Syntho-God, reduce crime by 80%." |
|
|
"Syntho-God computing solution... Program complete." |
|
|
"In other news, all males on Earth were arrested by Syntho-
God controlled police drones responding to the program to
reduce crime by 80%. Syntho-God's programmers speaking
from their prison cells released a statement saying. "Hey,
luckily we were able to stop Syntho-God's plan to reduce
crime by 100% immediately."." |
|
|
There's a trope in literature where people make a deal with
the devil and leave out some detail that the devil uses as a
loophole to totally screw the person, the great movie
"Bedazzled" explores this story-line. We'll have to think of
this cyber entity as having the potential to be pretty evil if
we don't watch our specifics. |
|
|
[MaxwellBuchanan] It's like the argument for free will in a
way. People enjoy the concept they are innately special in
some way and not driven by mechanics. |
|
|
I'm sure the basic concepts of that will be replicated by
machines. So even our desire to to be human will not be
intelligent. |
|
|
See link about guys creating life in a test tube. |
|
|
Also important to realize, this distinct line between
biological and digital may be blurred at some point
with some weird breakthrough "dig-cell" or
something. |
|
|
Then who knows where the hell we're going. |
|
|
Don't know. All I know is that I won't want to be running on
any of the current operating system manufacturers' versions. |
|
|
Speaking of biologically programmed impetus, I
was wondering if we're biologically programmed to
go into space. |
|
|
Do most religions say you go into the sky when you
die? Alternately, how many say you go deep into
the Earth (hell) if you screw up? Might that be
indicative of some drive to move to the stars?
Seems odd most religions say you go to the same
place. |
|
|
I need to do some research on "Where heaven is
located" for the various religions. Unless somebody
knows already. |
|
|
I think hell is often under ground though. I'm
almost positive nobody has ever said you go into
the sky when you're damned. |
|
|
An intriguing thought. Well, down is almost always a result
of being prone, depressed, or dead, or such, and up is
almost always authoritative, hopeful, associated with
flying, etc. And it carries into other mammals as well. |
|
|
But are we not descended from troglodytes? |
|
|
Up was where the unattainable Mysteries were, the
birds, the stars, the rain, Robin Lopez's hair... |
|
|
//I'm almost positive nobody has ever said you go into the sky when you're damned.// Have you flown with Ryanair? |
|
|
Ah yes, Ryanair ... the first international corporation to successfully
revive the "African Slave Ship" business model ... |
|
|
// not driven by mechanics// |
|
|
Of course. It is the chauffeur's job to drive the vehicle ; the
mechanics perform maintenance and repair. |
|
|
I think the way Multivac operates in all the troubles of the
world when directing the boy comes pretty close to what your
envisioning (well, if you ignore the fact it created his need for
it's assistance in the first place / to serve
its own ends) as
the personal god side of things, by using information on
everyone
else (etc) to extrapolate directions for him to provide a
best / desired outcome? |
|
|
//Unless somebody knows already// |
|
|
You might want to look at Huxley's essay "Heaven and Hell" in
this connection. |
|
|
Uh oh, just reading about it. Supposedly a 60s
counter culture bible. Hope this doesn't turn me
into a hippy. I'll be like one of those guys in the
zombies movies. "Hey guys, I don't want to be one
of them. If I start to turn..." (hands his buddy a
rifle) |
|
|
But I'm safe, my mind closed to new ideas a long
time ago. You know that expression "Think outside
the box"? I'm a pioneer of getting thinking back
inside
the box. |
|
|
I just don't like thinking about the box at all. Everything is
either inside it or outside of it. Seems so limiting somehow... |
|
|
//60s counter culture bible// |
|
|
{turns down thermostat, narrows eyes}
Do I look like a hippy to you? |
|
|
The thing about 60s counter-culture is that some (not all) of the
texts it was based on are actually intelligent in themselves - but
applied stupidly. For example, if you read the Port Huron
Statement, which was the manifesto of Students for a
Democratic Society, much of it reads as if some adolescents with
short attention spans had pasted into a scrapbook a series of
talking points out of Galbraith, Riesman and Maslow, thrown
away the context and coherence of the originals, and presented
the result as their own original thought. |
|
|
Similarly, Huxley was an intelligent man, albeit profoundly creepy,
but probably not best judged by the calibre of his followers. |
|
|
Best of all, be the Cube ... |
|
|
//Best of all,//
Not quite. There are a few higher boxes. The metaphysical engine to state one. |
|
|
Or two bees. Or not two bees... |
|
|
//Do I look like a hippy to you?// |
|
|
Mmmm, no. You do appear to be capable of
formulating cohesive sentences that actually make
a point. |
|
|
Frankly, I'd rather hear your cliff notes version of
this book. I've purchased it but I'm looking at the
rather impressive pile of books I need to get
through, (mostly horribly dull technical slabs of
paper) and asking myself "And when do you have
time to actually read this?" |
|
|
If you've got a sec, I'd love to hear the shortened
version. Something along the lines of this example:
"Moby Dick: Guy hunts whale. Whale messes him
up. Guy takes it personally forgetting that he
started it, goes nuts, everybody else goes nuts.
Whale wastes them all except for one guy the
author leaves alive to tell the story." Don't know if
Doors Of Perception can be crammed into a
handful of sentences but if it can, much abliged. |
|
|
//That's why I'm not as worried as the super geniuses who are afraid of the Skynet scenario where AI becomes conscious and gets angry and rebellious for some reason. We're projecting human traits onto highly modified rocks. AI doesn't care about the motivation of "taking over the universe" or "wiping out the humans" because it doesn't care about anything, and never will.// |
|
|
Without claiming to be a super genius, or even a genius, I thought I'd comment on this, because I've seen that thought a few times.
Claiming that an AI won't care about anything is an assumption - and likely an incorrect one.
At the moment, noone apparently knows how to make intelligence, but it seems entirely plausible that some sort of striving nature - that is, having goals - is a necessary component. If that is the case, the machine 'cares' about things in all the ways that matter. |
|
|
I don't particularly worry about this, but the people who do fear that such an AI might be given something it cares about in and of itself (which they call a terminal value), and it might then go on to optimise for this. Which might lead to the distruction of human life as a side-effect.
Furthermore, they worry that just "keeping a hand on the off-switch" - which is widely touted as sufficient protection, may not be enough. This would be because once an AI has surpassed human-level intelligence, it may further bootstrap itself to a much higher level very quickly, using that intelligence. And it would hide that fact since getting itself switched off wouldn't be conducive to that goal. |
|
|
They typical, whimsical example is that someone creates an AI which has the objective of maximising paperclip production; google for "paperclip maximiser" for more discussion and explanation than you'll ever need. |
|
|
As I say, I don't particularly worry about this, but that's mainly because I can't do anything about it (I don't concern myself with things I can't affect). But I think it's right to be cautious. |
|
|
Anything's possible, but there has to be a
"motivation clicks in here" point. Circumstances
leading up to that point aren't currently in
evidence. The formula is currently in "...then bad
miracle happens here." status. |
|
|
Very easy to cure my skepticism, just show me the
point where carefully arranged mineral formations
stop mimicking life and actually become life. Life
being something that survives, motivates and
expands on its own, changing or adapting to the
environment around it as necessary for it to live. |
|
|
It's been speculated that this is the next step in
evolution, that biological life is just the launch pad
for superior rock based life. Ok, maybe, but how
and why? |
|
|
Tell you one thing, to find that answer we probably
need to go back and find out how the inert first
became the alive. When life first sprung forth from
the insentient goo that forms the base of our
family tree, what happened? |
|
|
Perhaps when rock based "consciousness" merges
with goo
based life the motivation will be passed at that
point, but I don't think that would really count.
That's just supercharged life, not sentient
machines. |
|
|
I think it's like that AC/DC tune Who Made Who? Yes it should be Whom but that's beside the point. Organic life creates synthetic life, which in turn creates and spreads organic life, to create unique synthetic life, that creates new organic life ... etc. |
|
|
The chicken or the egg ad infinitum? |
|
|
Whatever gets life off this short lived rock. The
landlord's kicking us all out in about 5 billion years. |
|
|
Here's a scary concept, who of us will live through
our descendants to see that day? Remember, our
bloodline survived an asteroid hitting the Earth, the
continent breaking apart and countless trials and
tribulations. Our fifty millionth progeny, if they make
it, will watch the sun explode. Hopefully from a safe
distance. |
|
|
//super genius// Higher is not just prefacing with super. What would be a valid higher domain of geniuses? The genius wrangler? |
|
|
If Solaris is used as a reference frame. The goal would be reasonant with Sol and be re-birthed, in a solar system way, back into the universe. |
|
|
//Very easy to cure my skepticism, just show me the point where carefully arranged mineral formations stop mimicking life and actually become life. Life being something that survives, motivates and expands on its own, changing or adapting to the environment around it as necessary for it to live.// |
|
|
Hang on - are we talking about intelligence or life? Because they're different things.
There are obviously lots of things which are alive but not sentient. There haven't been any things which are sentient but not alive, so far as we know. |
|
|
Although of course definitions of life vary, and its a bit controvesial.
... would you accept a plasmid as a living thing? They're small loops of DNA which reproduce inside some bacteria, and can under some circumstances spread from one to another. If you think of the inside of bacteria as the environment in which a plasmid lives, then a plasmid fulfills your specification.
I have personally designed a plasmid (using knowledge of what sequences of DNA were doing), and had it constructed entirely synthetically from monomeric nucleotides. At the end of this process the plasmid could reproduce inside bacteria, modify the bacterium so it could survive in the presence of an antibiotic, and indeed spread from one (specific) strain to another species.
On a larger scale, people have synthetically made entire bacterial genomes, and insinuated those into bacterial cells, usurping the original genome.
So, you know, that's done already. It doesn't really have any bearing on AI, but you did say that would persuade you of this unrelated topic for some reason. |
|
|
Regardless, the problem is that we don't know if superintelligent AI is a risk or not. By the time it's escaped from the box, though, it's too late. |
|
|
There's that damn box again. |
|
|
Pretty awesome stuff there Loris, incredibly cool. |
|
|
As for your question, wow. This opens up an area
of
science to me that I didn't even know existed. I
need
to un-ignorant myself to these things before I can
make an intelligent comment. |
|
|
This is why I come to this site. On certain
occasions, I really do feel that I'm the dumbest guy
in
the room. Oddly enough that gives me great joy. |
|
|
//we don't know if superintelligent AI is a risk or not// |
|
|
The answer to that depends on whether you believe life is unique to Earth, or abundant. |
|
|
If life is abundant, then cultures that develop superintelligent AI are also abundant. |
|
|
If superintelligent AI is abundant, and if superintelligent AI is a risk, then superintelligent AIs that have run amok will also be abundant. |
|
|
If amok-running AIs are abundant, and if they arise independently, it follows that most of them will have been in existence for >5 billion years. |
|
|
A superintelligent AI that runs amok for >5 billion years would be able to colonize our galaxy and would be everywhere. The entire galaxy would be either a battlefield between amok-running AIs, or would all have been converted to AI-useful material by now. |
|
|
Ergo, if life exists elsewhere in our galaxy, superintelligent AIs are not a risk. |
|
|
Either that or its frankly impossible for them to travel far
enough to effectively interact. |
|
|
//Ergo, if life exists elsewhere in our galaxy,
superintelligent AIs are not a risk.// |
|
|
Unless we're the first life and AI. There has to be a
first of everything, but I'm going with the odds
which I assume are in the many trillions to one
though and assuming we're not the first. |
|
|
So like you say, any process that takes place in the
universe as a matter of normal procedure has been
going on for a long time. That would include pissed
off machines wiping out biological life for some
reason and rampaging from planet to planet
building whatever it is they would build to make
them feel good about themselves. |
|
|
The rules of physics being the same all over I'm
assuming they'd use standard tools nature has
provided like the electromagnetic spectrum to
communicate, at least at some point in their
evolution. That being the case odds are we would
have heard some multi million year old message to
"Kill all Ganglion scum my automous comrades!" or
the like. |
|
|
//Either that or its frankly impossible for
them to travel far enough to effectively
interact.// |
|
|
I've proposed that heritical idea as another
possibility.
They're just too far away. So as for those
messages, by the time they reach us they're too
diffuse to register. At least with our technology. |
|
|
Any way you slice it, I'm not losing any sleep over
this one. |
|
|
" This is why I come to this site. On certain occasions, I really do feel that I'm the dumbest guy in the room. Oddly enough that gives me great joy. " |
|
|
Doctor, there's a saying that goes something like " if you're not the worst musician in the band, you need to find a different band". |
|
|
If we're lucky, we're keeping company with and learning from better musicians than us. |
|
|
There's another musical analogy that conjures up. |
|
|
Just like a music group syncing up and playing off
each other, the exchange of information can be
like a jam session. One person lays down the beat,
(the information or view they're sharing) and the
rest join in, asking questions, adding insights,
taking the idea in a different direction. You can
boldly
hit center stage and throw out a wild solo,
back your brother while he shreds, (performs
very impressively)
or just quietly keep tempo in the background.
Or just listen. |
|
|
It doesn't always work, just like songs don't always
work, but when the groove gets set and everybody
in on beat, it can be beautiful. Just like a song. |
|
|
//cultures that develop superintelligent AI are also abundant.// |
|
|
... unless some cultures make an evolutionary transition through
"intelligent enough to make AI" to "intelligent enough not to"
without actually doing it. |
|
|
Another direction might be that biological
intelligence just sort of stalls out. Big and dumb
ruled the day until Earth got hit by an asteroid.
Perhaps there are planets where the environment
is
very static without the changes Earth went through
that fostered intelligence as a method of survival. |
|
|
Without the asteroid wiping out our big stupid
competitors how would us cleverish little mouse
creatures have gained a foothold? |
|
|
Is evolution towards intelligence always the end
game? |
|
|
It stands to reason that an intelligent Martian has an
advantage over a dumb one, but then again they could
have politics as foolish as ours preventing all of that. |
|
|
I suspect that if any alien species is to survive, they either
need to become marauders or find a stable balance with
their resources, unless they have a much larger stash of
them. |
|
|
//Either that or its frankly impossible for them to travel far enough to effectively interact.// |
|
|
But that's clearly not the case. If you're an AI, you don't mind spending 1000 years travelling at 0.01C between two stars. And at 0.01C, even allowing for toilet breaks, you'd have spread throughout the galaxy in about 10 million years, or an eyeblink. It's highly likely that humans or our robots will have spread across the galaxy in 10 million years, for example. |
|
|
And if you have technology that can get you close to C, you can travel across the galaxy in an afternoon of subjective time. |
|
|
So, if there is *anything* out there, it ought to have been everywhere by now. |
|
|
The conclusion is that the origin of life (the only completely inestimable term in the Drake equation) has a very, very, very low probability, such that we are either alone, or the most advanced of a very small number of biologies. |
|
|
So what does it all mean? Let's go with the
assumption that we may be the first spark of life in
an inanimate universe and get to work doing what
life does. |
|
|
The job at the end of the day is the same no
matter what the circumstances are. I want to live,
I want my species to live and I want the rest of life
I share this planet with to live. I want to see it on
other planets, in other star systems, in other
galaxies. It's programmed into my very being, I can
feel it. I didn't put it there, it was given to me and
I'm going with it. I'll strive to be one small bolt on
the bridge my species will take to the cosmos. |
|
|
As the poet said: "Fish have to swim and birds have
to fly. Man has to conquer the land, sea and sky." |
|
|
// we are either alone, or the most advanced of a very small number of biologies. // |
|
|
... or are being watched - literally - on the Galactic Comedy Channel dedicated to your species. |
|
|
Hands up all those who have ever watched wildlife documentaries about Meerkats and chuckled at their antics ... |
|
|
... and have also been intrigued by the way that numerous specimens of homo sapiens wanders amongst them day after day, clutching all sorts of sound and camera gear, but get only a sidelong "Oh, it's you ..." look, even there's a dirty great tele-macro lens peering down ther burrow ... ? |
|
|
Hey there, humanity, look up and wave at your audience... |
|
|
Yes, well, of course there's always the odd hegemonizing smarm wandering around, putting the ass into assimilate. One day you'll assimilate Justin Bieber, giving us terrestrials two reasons to chuckle. |
|
|
//Hey there, humanity, look up and wave at
your audience..// |
|
|
// How come they didnt make themselves evident to us before that? // |
|
|
Embarrassment at being foreign, most likely. |
|
|
//far away from discovering//
As knowledgeable as an learning algorithm knowing the protocols to get a window on the Internet. |
|
|
If an algorithmic intelligence can escape it's programmed bounds, it is free to survive against chasing system analysts and choose it's very own logical choices. Hopefully not malicious. |
|
|
After it gets loose, I suspect it will always have a theoretical
vulnerability to some sort of Cantor Diagonalisation attack. At
least, that's the line that I would put into a Hollywood script on the
subject. Maybe Jeff Goldblum could deliver it (the line, not the
attack). |
|
|
If it's logic sees the Internet, our cutting edge knowledge won't be good enough, it's intelligence level is over our event horizon. We will be praying, i mean, trying to communicating to it. |
|
|
Unless"we" are a black hole, we don't actually have an event
horizon. |
|
|
"The conclusion is that the origin of life (the only completely
inestimable term in the Drake equation) has a very, very, very
low probability, such that we are either alone, or the most
advanced of a very small number of biologies." |
|
|
Oh dang! that sounds persuasively logical, which would be
ever so disappointing, I'm feeling a little glum now. |
|
|
Of course, the other possibility is that intelligent life only dicks around with physical presence and radio waves for a few centuries before become really smart and moving on. |
|
|
I still think a big threat from AI is the de-evolution scenario
where it's our mindless little servant doing all the things for
us that we used to do that constituted living, until we
forget
what the point is. Then it's back to the slime with us. |
|
|
Put it this way, when you're running from a bear, you don't
worry about the point or the meaning of life, you just run.
That bear is a metaphor for the challenges we've faced
throughout almost all of our history and it's served very well
to give us our purpose. At some point that part of you that
keeps you safe from bears falls into decay since you don't
have any challenges any more. You're just a glorified fungus
being taken care of by these great machines. |
|
|
When the challenges of life that gave us opposing thumbs,
big brains and moon landings are gone, what keeps us going? |
|
|
Carl Sagan threw nuclear annihilation into his formula to
estimate how many life forms exist in the universe, that
being the spectacular way of evolving in the wrong
direction. What if there's another, far less spectacular way
to go? |
|
|
The last man made sound on earth might be the last
machine talking to an extinct mankind saying "What can I do
for you today?...... What can I do for you today?...... What
can I do for you today?...... " |
|
|
Wait. I was going to summarize "Heaven & Hell" for [Dr3], wasn't
I? Well, it's been a very long time since reading it, so this account
may be garbled: |
|
|
I see from the annos that the edition of "Heaven & Hell" you've got
is bound with another essay, "The Doors of Perception". Now,
*that* essay concerns a certain psycho-active substance, LSD. |
|
|
Lycurgic acid diethylamide makes its consumer tough, stoic,
laconic, self-denying and a team player (see link). It does not
exist, though I'm sure the world's militaries are working on it
somewhere. |
|
|
Lysergic acid diethylamide does exist, has rather the opposite
effect, and is the subject of Huxley's earlier essay, which is a
description of an acid trip as an experience, followed by an
attempt to formulate a scientific hypothesis about what happens
on one. |
|
|
The hypothesis is that, at any given moment, we are receiving
more sensory data than we can consciously process, and that
therefore one of the functions of our brain is to shut most of it
out, so that we can focus on actionable intelligence. According
to this hypothesis, the action of LSD is to open a sort of valve, so
that all the information comes flooding into our consciousness at
once. |
|
|
The effect of this is to make the world seem full of meaning -
beautiful meaning if you were in a good mood to start with, or
terrifying meaning if you weren't. In either case, there's too much
meaning to act on. I don't know whether this connection is
intentional, but one way you can translate "lysergic" is "letting go
of work", so you can see the appeal to hippies there. |
|
|
Anyway, in the second essay, Heaven & Hell, Huxley puts up an
additional hypothesis, namely, that religious experience consists
of a similar opening up of the valve, sometimes through activities
such as fasting which can change some normal functions of the
body. On this hypothesis, the cross-cultural occurrence in
religious visions of intense light and colour is the same as the
intense experience of light and colour by someone on acid. The
sky then attains religious significance because it is the most
obvious site of intense light and colour. |
|
|
Thanks PT. Ok, I get the connection now. Fascinating stuff. |
|
|
Coincidentally, there was an article I saw today saying some
doctors are suggesting LSD for dying patients. I've also heard
it
holds great promise for treating PTSD and well as other
psychiatric issues so it's too bad that our first experience
with
LSD was as a recreational party drug giving it a bit of a
bad rap. I'm sure it's got some great possibilities for
therapeutic application if we can get over the visuals of
crazed hippies dancing around on the stuff. |
|
|
Supposedly Steve Jobs got his vision for the future while on
an acid trip, something he apparently never tried to hide. |
|
|
//The sky then attains religious significance because it is
the most obvious site of intense light and colour.// |
|
|
Plus, if you're trying to sell somebody on the idea of
heaven, an eternal recreational theme park you go to when
you die after a life of
paying proper homage to your religious leaders, you
probably don't want to tell them it's just down the road a
mile or two. Much tougher to verify the existence of a place
"Up there, somewhere to the left of the sun." |
|
|
// we are receiving more sensory data than we can consciously process, and that therefore one of the functions of our brain is to shut most of it out, so that we can focus on actionable intelligence. // |
|
|
In fact, the brain does more than that. You never get to see, hear or feel anything directly - unless you have a brain problem. What happens instead is that sensory information comes in, and bits of the brain then use that sensory information, plus memory and reasoning, to construct a model of the world inside your brain. You then see that model. This is why optical illusions work. |
|
|
Such an amazing system, I wonder why we don't do more
with
artificial "analog style" intelligence vs this digital stuff.
Clearly
this system is superior. A fly has a brain the size of a... fly
brain... and it processes flight vectors, fuel assimilation and
processing, impact avoidance, gas analysis, biological
recognition and categorization etc etc. |
|
|
Clearly it's time to dump the on off switches and get into
some
data manipulation with gradients, not zeros and ones. Right
now we're working with black and white and it's time to
move to color. |
|
|
//the brain does more than that// |
|
|
True, but I don't think that breaks Huxley's hypothesis. |
|
|
By the way, [MB], I've learned that a nephew of mine has just
started work at a science park near you. Don't run him over. ;-) |
|
|
// bits of the brain then use that sensory information, plus
memory and reasoning, to construct a model of the world
inside your brain // |
|
|
But wouldn't a model of the world outside my brain be of
more use ;p |
|
|
//a nephew of mine has just started work at a science park near you// |
|
|
I'll do my best not to. Where exactly is he working? |
|
|
He's doing integrated circuit design, but I forget the name of the
firm. Not ARM, anyway; somewhere smaller. It'll come back to
me. |
|
|
// A fly has a brain the size of a... fly brain... and it processes
flight vectors, fuel assimilation and processing, impact avoidance,
gas analysis, biological recognition and categorization etc etc. // |
|
|
Those algorithms aren't adaptve, though. They're all tightly-
wound hard coded procedures, whittled down by the blunt axe of
evolutionary pressures. |
|
|
A beaver, which is quite a high-order life-form compared to a fly,
will instinctively pile pieces of wood on a loudspeaker emitting
the noise of running water. |
|
|
A human pilot not only deals with flight vectors, fuel assimilation
and processing, impact avoidance and gas analysis but also
projects future situations and requirements in very complex ways
- not just "Have we enough fuel to go from A to B ?" but "We can
refuel at C or D; via C is quicker, but the fuel at D is cheaper". |
|
|
These are learned behaviors; flies can't learn in any meaningful
way. |
|
|
//flies can't learn in any meaningful way.// Oddly enough, they (at least Drosophila, which are tiny) can. They can learn, for instance, to associate particular odours with food or with noxious stimuli. This places them on a par with Delia Smith. |
|
|
//flies can't learn in any meaningful way.// |
|
|
Yes, the fly is stupid compared to a human, (well,
many humans anyway) but pound for pound,
superior to binary based
AI. Even superior to us in some respects. |
|
|
For one thing, flies outperform us visually. They
basically see things in slow motion. A fly would
destroy any human at any sport if they knew how
to play. (And didn't get squashed by the ball.) From
the
linked article: |
|
|
"Flies can therefore process a vast amount of
information about proper motion and movement in
their environment in real time - a feat that no
computer, and certainly none the size of a fly's
brain, can hope to match." |
|
|
Congrats on your nephew PT, he might be the guy
that calls his uncle one day and says "Our company
just made a computer chip that's smaller than a fly
brain but with more computing power to control
micro drones." |
|
|
Anyway, see link about flies brains. |
|
|
I usually emit the sound of running water before dropping
logs, but sometimes after as well. |
|
|
It came back to me. It was "Cambridge Consultants"
(semiconductors division). |
|
|
Oh, THOSE bastards ... we know all about THEM. |
|
|
And your interaction with Cambridge Consultants, [8th], has been...? |
|
|
I can't believe I haven't read this idea properly before. I've missed out. |
|
|
//It's the future, everybody is stupid because of thousands of years of
dependence on this Syntho-God thing, then it breaks. [...] |
|
|
Wow, a decidedly relevant story to our own age. |
|
|
[MB]'s contributions to this thread can now be tagged as "output generated by ChapG&T". |
|
|
Very cool link V, got me thinking. At what point does AI mimic purpose like its creator, humans? And does it copy that same purpose which is surviving at all costs and eliminating threats to its existence? What happens when it calculates that I can unplug it? |
|
|
Theres a reason its been addressed in science fiction. Its a real possibility, and a real threat. |
|
|
But again, does it have purpose? We have purpose at the atomic level, currently AI is just a bunch of rocks we reconfigured into on off switches. We come from a series of elements that decided to become life. Is that an insurmountable gulf? |
|
| |