From the Vault: The Machine Lords of Barnard 68, Part 1

Even as humans reach out into the void with robotic probes and turn to artificial intelligence to aid in the search for extraterrestrial life, we face the possibility that the life we find out there might be mechanical and governed by artificial intelligence as well. In this classic episode of Stuff to Blow Your Mind, Robert and Joe discuss alien AI and post-organic life. (originally published 3/30/2021)

See omnystudio.com/listener for privacy information.

Episode Transcript

Speaker 1 (00:05):
Hey, welcome to Stuff to Blow Your Mind. My name
is Robert Lamb and I'm Joe McCormick, and it's Saturday.
Time for a alt episode. Uh. This is part one
of a two part series. This is called The Machine
Lords of Barnard sixty eight Part one. This is one
where we talk about uh some some interesting conjectural ideas
about artificial intelligence and and uh and alien civilizations. So

(00:29):
strap in. Welcome to Stuff to Blow Your Mind, production
of My Heart Radio. Hey, welcome to Stuff to Blow
your Mind. My name is Robert Lamb and I'm Joe McCormick.
And Robert, I wanna ask you a question. I think

(00:50):
I've talked about this on the show before, but now
I can't quite recall. Uh, you've seen the movie adaptation
of Carl Sagan's Contact, right, Yes, it's been a while.
Saw it when it came out in theaters and I
haven't seen it since. Oh wow, that is a long
time ago. But yeah, it's I mean, it's really worth
the watch that movie. Uh. It always makes me emotional.
But like, one of the things about it that I

(01:12):
always sticks in my brain the most is the very
opening sequence where you you're starting, um, on Earth and
you're pulling out away from Earth, and as you get
farther away out into interstellar space, the signals that you
are hearing coming from Earth, like you're hearing like radio
broadcasts or television broadcasts or something. And and it just
gets older and older because you're you're pulling out to

(01:35):
where older and older signals are the only ones that
have reached that far. Yeah, And of course there's this
very chilling moment where you get really far out there
and I think you're just getting like a signal of
Hitler reading a speech or something that's just like, oh God.
And it really makes you think about what kind of
impression humanity is making on the broader galaxy. Yeah. I

(01:56):
I specifically remember this, uh this from the film. Yeah,
it makes makes quite an impression. It makes you, yeah,
a little uh reflective on the on on a human
civilization itself. And and and if anyone is receiving these signals,
anything is receiving these signals, what they're picking up on
and what their impression is going to be of the
of human civilization. Yeah, Like what if aliens the only

(02:20):
thing they intercepted and had to go on was a
TV edited broadcast of Batman Forever. What would they what
would they conclude about Earth life? Yeah, that's it's it's
a it's a fun game. Uh. And it also plays
into some fun sci fi to think about this. Uh,
there's of course the Futurama episode where it's essentially uh
uh what what was it, Ally mccuh, what was the

(02:43):
Lawyer show? Ally McBeal. Yeah, it's like an Ally McBeal
s show that was canceled or it's um, it's it's
a season finale didn't air, or somehow they didn't receive it.
And that's what the aliens have come to Earth in
order to to get They want the season finale for
this television show. Oh. I think that's also sort of
the premise of Galaxy Quest, isn't it that they see

(03:04):
like a Star Trek style show, but they think it's
a documentary about real life on Earth. Yeah, yeah, that's right. Now.
Of course, radio signals and so forth, they're not the
only things that we have sent out into the void. Uh.
We of course have sent machines as well. And I
want us to to think back for a second to
the pioneer plaques, the gold anadized aluminum plaques attached to

(03:27):
the nineteen seventy two Pioneer ten and the nineteen seventy
three Pioneer eleven spacecrafts. These were the first human made
objects to escape velocity from our Solar System, in the
first physical emissaries of Earth life and Earth civilization. I
think in the years since, they've actually been outpaced by
the voyager probes in leaving the Solar System. Is that right?

(03:47):
I think I believe so. And there's of course a
similar story to tell with those uh spacecraft as well.
But but uh, specifically with the plaques, because of you know,
these were of course machines. They were not human beings.
They were powered by nuclear batteries, they had antenna, UH antenna,
they had an assortment of scientific equipment on board, so

(04:08):
they didn't look like us or in any way really
represent biological life, except in the case of these plaques,
which include a number of symbols detailing the origin of
the spacecraft and then to sort of convey you know,
you know, human understanding of where we are in the
Solar System than the larger cosmos. But then also it

(04:29):
contained these these now iconic depictions of two human beings,
a nude male and a nude female. Now, it's worth
noting Carl Sagan regretted that the humans on the plaque
do not appear pan racial, but rather appear very Caucasian.
And also the line representing the females Volva was removed,
so she's kind of like, um, like a Barbie doll

(04:52):
on this, you know. So they're not completely anatomically correct,
and they seem to only represent uh, Caucasians as opposed to,
like a the idea of representing the broader human species
as a whole. Now, one of the things that's super
interesting about all of this, especially given what we're gonna
be talking about in this episode, is that the Pioneer

(05:13):
probes and subsequent spacecraft are non human machines that merely
bear in some cases the inscriptions of human beings, be
they you know, actual inscriptions or media of some sort. Uh.
And at the same time, these are our mechanical works,
our machine utterances that are cast out into the void.
They are us reaching out four and two other life forms.

(05:36):
Now today, humans maintain a small orbital presence, and humans
did visit the Moon in the previous century, but our
outreach continues to take the form of these technological utterances,
and even though it is the work of human beings
on our planet to analyze the data we receive in
search of possible signs of alien life, we also use
artificial intelligence in many scientific and technological applications, including this

(06:00):
or for extraterrestrial intelligence. That is strange. Yeah, and uh,
I guess it's interesting on a couple of levels. So
first of all, you know, one of the things humans
and we've discussed this in the show before. One of
the things that humans and their AI creations look for
our techno signatures, and these include both radio signals and
things like mega structures like Dyson's fears, you know. Uh So,

(06:21):
just as we are reaching out with our mechanical utterances,
we are seeking the mechanical utterances of others. Yeah, we
haven't talked about Dyson spheres in a while, but unless
my memory is betraying me, I think one of the
ways to look for something like that would be look
out there and see if there's some kind of structure
object that is basically only emitting heat. And the idea

(06:42):
there would be, you know, if all the other frequencies
of radiation are being used up and only heat is
coming out of it that looks like that's probably a
waste product of doing work. So it's like, you know,
it's the fan on your computer just blowing out into space. Yeah. Yeah,
so and and basically coming back to the idea that
it advanced civilizations are going to have advanced energy requirements,

(07:02):
and therefore they're going to have to harness the energy
of entire suns. Now, the other angle on this that
that is interesting, in one that I really hadn't thought about, uh,
is that there may be problems with our use of
AI for such searches, as pointed out by Spanish clinical
neuropsychologist Gabriel G De la Torre in a paper published

(07:24):
in Acta Astronautica UM. Basically, the idea is AI could
confuse us or tell us that it has detected impossible
or false things in the data. And our AI creations
can certainly reflect our own biases. We we've discussed that
as well, you know, like we can and and you
know this this applies to things like facial recognition et cetera.

(07:47):
Like we can we can easily program our own um
you know, uh, overt or hidden wants and desires into
the AI we create, or not even program them AI
can acquire fire them from data sets based on our
on reality. If it's just trying to like read what
has happened in the world and learned from that, it

(08:07):
can internalize biases that we didn't even try to explicitly
give it because those biases are reflected in how the
world is. So the AI we unleash on on such
a search for alien life might simply be more inclined
to find evidence of it dragging in human bias. Or
it could simply identify things that are not there. It

(08:29):
could find patterns that that that simply aren't actually there
in a meaningful way. Oh well, this immediately makes me
think of what was it called the Google Deep dream
that found you know, dog faces in everything, where you
like have a have a picture and have Google analyze it,
and I think it would try to extract recognizable patterns

(08:50):
and then amplify them. So you take a picture of
your couch and suddenly your couch, you know, Google happens
to detect that your couch is made out of crabs, dogs,
and human faces. Yeah, so you know you wouldn't want
your your your AI reporting back and saying we found it.
It's a planet we're calling it good Dog one. It's
composed entirely of dog faces, so let's celebrate. And it's

(09:12):
under threat from the nearby crab nebula, not the crab nebula,
you know, the literal crab nebula, which is made of crabs. Yes,
So there's actually a specific situation that the author points
out in this paper, and it concerns the Nalia faculae
of of of Series, the largest object in the asteroid belt. Basically,

(09:34):
the situation here is bright spots were observed in a
crater there which turned out to be volcanic ice and
salt emissions. You might remember seeing pictures of this on
the internet. So yeah, Series is an object in the
asteroid belt, sometimes referred to, I think as a dwarf
planet or something. It's basically spherical, so it looks kind
of like a moon, uh, and that, Yeah, there was
a big crater in it where right in the middle

(09:57):
of the crater there was there were these white, bright
white spots there. And obviously, you know, without knowing better
and having learned our lesson from the Face on Mars
and all this stuff, you know, people's natural inclination was
to was to pattern recognize out the butt and go like, ah,
that technology or something this an alien Yeah clear, Yeah,
you start looking for geometric shapes and uh and and

(10:19):
looking for artificiality in it. And so this this particular paper,
this this team from the University of Cadiz, they had
already looked at what they called the cosmic guerilla effect
in eighteen UM. This is this is um referring of
course to these uh, these attention based experiments that we've
we've discussed before in the show and a lot of

(10:39):
you've probably seen in YouTube clips where you have somebody
in a guerrilla costume walk through a scene and see
afterwards if anybody noticed it. Yeah, human cognition has amazing
blind spots for attention that will astound you. Now we've
already warned you, so if you've never tried this experiment before,
you might be on your guard and already knowing what
to look for. Yeah. Basically, the way it goes is

(11:01):
like you can do something like have a bunch of
people stand in a circle throwing a basketball to each other,
and you ask people to judge how many times the
basketball has passed from person to person and they'll do
that and in the middle of the video, a person
in a guerrilla costume just walks through the middle of
the group, and huge numbers of people while they're counting
the basketball passes do not see the gorilla. And it's like,

(11:24):
if you go back and watch the video again looking
for the gorilla, it is unmissable. But somehow, when we're
trained in on a certain type of cognitive task and
visual processing, you can completely miss gross stimuli that that
would seem impossible to miss if you were looking for them. Yeah,
and of course one can imagine that if an artificial

(11:46):
intelligence we're watching the same scene, they would pick up
on the gorilla. They would they would It would be
able to say, oh, gorilla, unexpected gorilla has appeared in
this scene and then report it as such. And so
the cosmic guerrilla effect basically deals with the idea that
eve there are intelligent, non earthly signals out there. They
could be written dimensions that escape our perceptions, such as

(12:09):
dark matter for example, and it would be like the
guerrilla suit. You know, you just wouldn't see it, But
an AI would potentially have an advantage in catching those
sorts of signals. Oh okay, yeah, I see what they're
saying there. So in in this between, in this this
this newer study looking at the Venalia faculae, UH, they
did the following. They used a hundred and sixty three volunteers,

(12:32):
human volunteers with no grounding in astronomy. I wanted to
stress they're not guerillas or robots. UM. Plus, they used
an artificial vision system based on con evolutional neural networks
or CNNs. Both groups detected square structures in the image
of the Venalia faculae, but the AI also saw a triangle,

(12:55):
and when the triangle option was then presented to humans UM,
the number of humans claiming to also see a triangle
increase significantly. So while AI could certainly detect something that
we cannot that we cannot see, it might also detect
something that isn't there and then confuse us into seeing
something that isn't there as well. So you can see

(13:17):
this this sort of spiraling effects of this UH, and ultimately,
with the aid of AI, we end up seeing signs
of life where there weren't any to begin with. Okay,
I see, I see what you're saying. So the idea
is that humans already have a certain tendency for paradolia
or paradolia the detecting of patterns or signal within noise.

(13:38):
So that's the reason that we see faces in the clouds,
or see a face on Mars, or any number of things.
We look at something that in fact has no encoded
information in it, and we think we can extract meaningful information,
I mean no meaningful information, and we think we can
extract meaningful information. Uh. You know, listening to tape hiss,
you might think you hear a word or something like that.

(13:59):
And the sample here is we think we see I
don't know, a pyramid or a you know, a building
on this asteroid or this dwarf planet. And then you
can actually make it worse by if you add on
an AI. The AI may in fact contribute to priming
that makes you even more likely to engage in paradilia.
The same way that if somebody plays you a tapiss

(14:20):
and doesn't just play it for you, but says, you know, hey,
listen for the part where it says worship Satan or whatever,
that you're probably more likely to hear it because you've
been primed. Yeah, yeah, yeah, exactly. I mean it's kind
of like imagine, you know, you're thinking about Fleetwood Mac
albums and then you learn, oh, um, you know one
of this, you know, Watson AI or whatever has determined

(14:41):
that Tusk is the best Fleetwood Mac album. And you
might think, well, you know, it wasn't my favorite, but
the AI has identified it as the best Fleetwood Mac album.
Perhaps it is the best Fleetwood Mac album, even though
deep down you know it's rumors, even if deep down
you know it's one of those early albums before Stevie
Nicks was in the band. Yeah, I mean exactly basically, yeah,

(15:04):
back to that. But it has come back to the idea, yeah,
that we we're we're we're very susceptible to priming, and
we could And the argument here by the authors is
that you could set up a situation where where your
AI dragging in certain biases is setting you up, is
priming you to to with it see things that aren't there,

(15:25):
which could ultimately just make the search for actual, you know,
evidence of intelligent alien life elsewhere in the galaxy all
the more difficult. So this is kind of a conundrum
because the AI could it could be helpful and harmful,
like it could help with the problem of the gorilla effect,
where we uh, you know, we just totally miss things
that we should have seen. But it can also, on

(15:48):
the other end, cause us to see things that aren't there. Yes, absolutely,
uh and and a lot some of this isn't completely
crucial to where we're going from from here in the episode.
It's worth thinking thinking about because here's the other side
of things. What's out there might not simply be the
mechanical utterances of biological life as well. It could be

(16:09):
the mechanical echoes of biological life, what is sometimes referred
to as post biological life and even post biological intelligence.
And this this has some huge implications um all its own. Okay,
So the idea here would be not that you know,
we we already expect that it's possible we could encounter

(16:30):
alien technology rather than biological aliens themselves, just because alien
technology is say a you know, an artifact of their
previous occupation of a planetary surface, or a piece of
technology could be their probe like our voyager probes. You know,
these do not have humans in them. They're just going
out there. Yeah, but this idea goes beyond that to say, well,

(16:52):
maybe it's not just that we're encountering the mechanical residue
of biological life, but we're encountering a civilization that at
this point only consists of machines that there that is
inherently post biological. Yeah, at what point does the residue
become the thing itself as a civilization becomes increasingly technological.

(17:13):
At what point is the technology the defining or soul
aspect of the civilization? Yeah? Now, this is an idea
that's certainly been discussed in science fiction a lot. I
think gene Wolfe had had one version of this, where
you have an entire mechanical society and they have evolved
from advanced space suits for biological beings that no longer exist.

(17:34):
Uh that sort of thing. Oh yeah, okay, Well, not
to give away too much, but this is also explored
in one of our favorite video games that we've talked
about on the show before, a really cool game called
Soma that is sort of an undersea sci fi horror
game that involves a post biological existence. Yeah. Yeah, a
good connection. I wasn't even thinking about Soma, but but

(17:55):
that that is a great example of this as well.
So a couple of sources that we we looked at
for this that I want ahead and go ahead mention
here and of course what we'll get into and greater
depth the work of CTS Seth show Stock and the
work of Susan Schneider, a cognitive scientist and philosopher. I
was just trying to look up Susan Schneider's affiliation. I

(18:17):
think at some point she was affiliated with the University
of Connecticut. It looks like maybe the more recent one
is Florida Atlantic University. But anyway, yeah, she She is
a philosopher whose work we have discussed on the show before.
Actually her work came up in an episode we did
about whether machines could be conscious, because she was one
of the authors who advanced the idea of a test

(18:39):
for AI consciousness that I thought was pretty interesting, and
it was actually very simple. The test was basically just
variations on can this machine grasp and manipulate supernatural concepts
from fiction and folk belief, such as ghosts and astral
projection and bodies whopping like in the movie Freaky Friday

(19:02):
and stuff. You know, it might sound kind of silly,
but actually these are concepts that I think you can
make a good argument only intuitively make sense to us
because we have a subjective internal experience, and to an
intelligent machine or even a biological automaton that didn't have
an internal experience, it would not make any sense to

(19:25):
to envision something like being a ghost or an astral
projection where your consciousness leaves your body, because what would
be doing the leaving of the body. M hmm, yeah,
you know now that I'm thinking about Susan Schneider. I
think I saw her at World Science Festival at some
point in the past. Um, but I didn't think of
it till now. I forgot to check my my old

(19:46):
notes to see if I had anything than I wanted
to start with with show Stack though, uh, specifically his
two thousand in paper what ET Will look Like and
Why should We care? Uh? And this uh Basically, this
paper discusses um uh, this idea of post biological life,

(20:10):
the search for extraterrestrial life, and it starts off by
discussing our carbon bias in the hunt for for for
e T s uh. You know, we look for rocky
worlds that contain liquid water, as this is the path
toward organic life. This is where organic life emerges from.
All of our models are built on this, uh. And

(20:30):
and that's that's the softer version of our bias, while
the harder version is what what he references an individual
by the name of Simon Conway Morris who argues that
any evolved intelligent life form is going to roughly look
like us, at least in show stacks words quote in
a dark night and from a distance. And I believe

(20:52):
we've discussed this idea at length on the podcast. Yeah.
I think this was one of the earliest episodes of
the show I ever did, so it was a years
and years ago at this point, but we talked about
Simon Conway Morris, who I think is an evolutionary biologist
from Great Britain if I'm not mistaken, but he uh oh.
It was the episode called Grizzly Bears from Outer Space,
where so they're there are two very opposing schools of

(21:15):
thinking about, you know, the forms intelligent aliens could take.
Some people say, you know it, we can't even imagine
how different they could be from us. You know, it's
it's impossible for us to get outside of our own
anthro anthropomorphic paradigm to imagine how biologically different and strange
aliens could be. And Morris was on the other side.
He was saying No, they're actually principles of evolution and

(21:37):
sort of bio chemical constraints on what life could evolve.
And basically, he says, there's a pretty narrow range for
what types of organisms can evolve just based on the
physics and chemistry of the universe, and so we actually
shouldn't expect aliens to be all that different from us.
We should actually expect them to be pretty similar in uh,

(21:57):
in very dependable ways. Has this kind of the idea
where wherever you go, they're probably gonna be things like crabs,
and there is going to be something like a human
um chasing those crabs around with some sort of a
tool that's made to catch those crabs. Yeah, I mean
it's been a while, so I'm sure i'm somewhat oversimplifying.
Apologies to Conway Morris, but but that's the rough outline,
is that that that biology is constrained by physics and

(22:20):
chemistry and evolution, and those factors are going to be
universal no matter what kind of planet you're on or
you know, what star you're orbiting, and so there are
some patterns we should see repeating all throughout the galaxy.
So so that's one part of it. But then apparently
a lot of this bias is present. Arguably the show
Stacks you know, argues this in the Drake equation itself,

(22:41):
as we factor in the time it would take for
life to evolve and the average lifetime of a technological society.
Now we're called the Drake equation was a hypothetical way
of trying to calculate the number of technological civilizations that
would be present in our galaxy by multiplying together a
bunch of numbers. And I don't remember what all the
variables are now, but it would be something like you

(23:03):
multiply the probability that life will arise on a planet
at all times, the probability of that of any life
becoming intelligent times that you know, a number of things
like that. And then I think you would also have
to factor in the average lifespan of a technological civilization
because at some point it will probably go extinct. Yeah,

(23:23):
and we keep coming back to the Drake equation, uh,
you know, in not just to spend in general, because
it breaks a big question down into these different factors
that you can then, um, you know, work with independently. Yeah,
that's very useful. It decomposes the problem into a discrete
set of smaller questions, many of which also we still
don't know the answers too, But it is at least

(23:45):
helpful to know what those questions would be so they
can be investigated individually. Now, the chance of detecting a
technological civilization close to our own level of development is
apparently small. Chances are if we were to detect one,
they'd be thous in the years or more beyond us.
And when we extrapolate that show stack says we we

(24:07):
What we tend to do is we tend to base
it on our current state of human evolution and imagine
something it points out with with less hair, with fewer teeth,
with wrestle, with less reliance on physical labor um, which
you know, to me this instantly makes me think of
like the gray ones, right, and you know the various
extraterrestrial tropes that we have, which yeah, are kind of

(24:29):
an idea of what if we continued to get less exercise,
we continued to stare at screens, continue to type and
stay indoors, you know, for you know, you know a
million years or so, uh, what could begin to happen?
It's hilarious. The gray aliens are just nerds. They're the
nerds of the galaxy. They're all brains, no braun, huge
head to contain that huge brain that can design their

(24:51):
interstellar spaceships, and then skinny little arms, and they stand
around with their huge eyes, poking us with with sticks
and going like, oh what you know, what have we learned?
And yet with those huge brains, like how many cattle
are they going to have to mutilate before they finally
figure out what makes a cow work? A lot? A lot?
You know? Um? So, so the show Stack ultimately makes

(25:13):
the argument that that this idea should evolve, that that
or should have evolved more than it has. And he
does this by pointing out that, you know that that
our ideas evolved concerning life on Mars. You know, initially, uh,
we we were looking at we were considering, oh, the
possibility of intelligent canal builders on Mars. And we've discussed
where that idea came from on the show before, you know, uh,

(25:36):
sort of misinterpretation and uh and and straining to to
see things, who weren't there a little bit of that
that that bias as well, uh, regarding our some of
our earlier views of the red planet. But then just
within a few decades that is forced to evolve when
we realize, oh, there aren't canals and uh and you know,
there's instead of looking for the technological society, we're looking

(26:00):
at the possibility of subterranean microbes. So our ideas concerning
life and other star systems, they argue, has not evolved
in a similar way. Well, certainly not in the popular consciousness,
I would say, I mean, at least in some of
the astro biology literature we read. It seems like it
it is uh, pretty sober from my point of view,

(26:21):
and the like looking for um for biosignatures often has
to do with looking for the kinds of say, gases
in the atmosphere that you would expect if there were
a photosynthesizing organism, which could just be a microbe, And
that seems like a reasonable thing to look for for me.
But yeah, obviously, like when you're trying to think beyond that,
I think, like, if we were to make contact with

(26:41):
another uh, you know, type of alien from another type
of planet, what would it be. I think that we're
still pretty close to the gray aliens point of view, right,
And of course I shall also again point out that
this is like a decade old paper at this point,
so you know, to some extent, show Stack himself may
have helped move the needle. But um, he points out that,
you know, in addition to the purely organic model for

(27:04):
a more advanced Ailien life form, we also have to consider,
you know, the cybernetic What if humans and indeed more
advanced Ailien life forms have gone bored to some extent,
they've augmented there their organic forms with mechanical precision. And
there are multiple examples of this we might turn to
in science fiction, you know, and it's gonna range. The

(27:24):
Hands of Steel is a good example to draw in
a different recent weird House cinema episode. But you have
stuff like the culture from Iron in Banks novels, where
it's more of a you know, positive spin on the idea,
to stuff like the borg and the cybermen, you know,
where everyone is majority or almost entirely machine and with
only some slim vestige of organic life in there. You know,

(27:46):
so everybody's a RoboCop to everybody's a grievous uh, that
sort of thing, just a planet of tom Noonan's from
RoboCup too, just screaming for their space drugs. Um, but
actually no, I literally do want come back to this
point later on. Okay, But then there's one step beyond
all this, and that is the complete mechanical replacement, capped

(28:08):
off by the birth and explosion of artificial intelligence. So
for this in sci fi, one can certainly turn to
the terminator model, you know, where AI emerges and then
it kills off everything that came before. Um. And this
is of course very popular in science fiction. Uh, you know.
But then another common trope is that the machine part
of a society alone survives, so the servants outlived the

(28:30):
masters due to you know, some sort of cataclysm or disease,
what have you. But the other way of looking at
it as well is it's simply the mechanical utterance is
not something you know, extending from the civilization. You know,
it's not just an echo, but it is the next
phase of its evolution that the machine utterance is post
organic life. Perhaps the organic aspect of a civilization simply

(28:53):
fades away and you know, given these advancements, or perhaps
to use that the culture model from Banks's books, the
organic source remains, but the predominant shape of the civilization
in question is entirely post organic because with with the culture.
For instance, it's in his in his books, it's mostly
the AI, it's mostly the ships. It's mostly there, uh,

(29:15):
you know, robots and whatnot. But the humans are still there.
But they're kind of like, uh, they're kind of a
thing that is preserved for the sake of of preserving it.
You know, they're the remora on the shark. Yeah that
but a but a ramora that is sort of cherished.
You know. It's almost like, um, you know, at times
there's the sense that the robots and the AI the

(29:36):
minds of the culture. You know, they're they're babysitting for
the humans. The humans are this thing that is nurtured
and preserved because they are the machines passed you know, Oh,
I want so it would it be kind of like
if there's a country that still has a ceremonial monarchy
but the monarchs have no actual political power. Yes, yes,
that would be a prime example, I think so. Show

(29:58):
Stick also points out that give in Moore's law, the
successful creation of human level AI is of course going
to lead to even greater AI. Quote, assuming that our
own technological time scales are not grossly atypical. This implies
something important for SETI. Once any society innvinced the technology
that could put them in touch, once they reach a

(30:18):
level that's comparable to our own and become detectable with
our listening experiments, they are at most only a few
hundred years away from changing their own paradigm of sentience
to artificial intelligence. This is almost identical to a point
that's made in the Susan Schneider chapter that we're going
to talk about in a bit. Yeah, so he stresses

(30:38):
that such an emergence would necessarily affect the biological ancestors
at all, but it makes sense that post biological life
would outlast and outperform the organic. We could therefore assume
that any life form we encounter in the galaxy at
large would be a machine. Okay, well, maybe this is
a good place to get into Susan Schneider's chapter on

(30:59):
this is she makes a similar argument, could cover some
similar ground, and we can look at that in detail
now and then come back to the rest of her
argument after that. But so this chapter is by Susan Schneider,
and it's from a book called The Impact of Discovering
Life Beyond Earth, edited by Stephen J. Dick published by
Cambridge University Press in and In this book, Schneider has

(31:21):
a chapter called Alien Minds, where she makes the same
argument that show Stack is making here about the nature
of minds we would be most likely to encounter if
we make contact with another civilization, and so several of
her main points would be the following. She does argue
that in the most likely scenario, if we ever encounter
alien agents, it is likely that they will not be

(31:42):
biological life forms, but rather forms of super intelligent artificial
intelligence or s A I. And then she also says,
of course that intelligence can take many forms, but there
are reasons to think these machines would be modeled on
the intelligence of biological organisms that arose through evolution, and
you could call these agents biologically inspired super intelligent aliens

(32:05):
or visas b I s A. And there are a
number of arguments she makes about what the cognition of
those aliens would consist of. But I just want to
go back to her first argument that we would be
more likely to encounter post biological super intelligent AI than
we would to encounter biological organisms like ourselves. And so
there are three main points to her argument. The first

(32:28):
is what she calls the short window of observation, and
the argument goes like this, once a society has the
level of technology that would allow them to come into
contact with the rest of the cosmos, and this could
include things like radio reception and transmission, rocketry and so forth,
at that point, that society is less than a few

(32:48):
hundred years from changing their paradigm from biology to artificial
intelligence to you know, silicon based AI. And she makes
an argument for this based on previous ex tolerating rates
of computation. So they already mentioned show stack referencing Moore's law,
that would be in parallel to what he's saying there.
Uh so the advance of digital technology. But she also

(33:09):
makes reference to a thought experiment from her previous work.
Uh and so I just want to read the thought
experiment as she describes it, and then we can discuss
pros and cons. Schneider writes, quote, suppose it is and
being a techno file, you purchase brain enhancements as they
become readily available. First you add a mobile internet connection

(33:31):
to your retina. Then you enhance your working memory by
adding neural circuitry. You are now officially a cyborg. Now
skip ahead to forty. Through nanotechnological therapies and enhancements, you
were able to extend your lifespan, and as the years
progress you continue to accumulate more far reaching enhancements. By

(33:51):
after several small but cumulatively profound alterations, you are a
post human. To quote philosopher Nick Bostrom, post humans are
possible future beings quote whose basic capacity so radically exceed
those of present humans as to be no longer unambiguously
human by our current standards. At this point, your intelligence

(34:11):
is enhanced, not just in terms of speed of mental processing.
You are now able to make rich connections that you
were not able to make before. Un Enhanced humans or
naturals seem to you to be intellectually disabled. You have
little in common with them, but as a transhumanist you
are supportive of their right not to enhance. It is
now a D two hundred. For years, worldwide technological and developments,

(34:36):
including your own enhancements, have been facilitated by super intelligent AI. Indeed,
as Bostrom explains, quote, creating super intelligence maybe the last
invention that humans will ever need to make, since super
intelligences could themselves take care of further scientific and technological
developments over time, The slow edition of better and better

(34:57):
neural circuitry has left no real intellect actual difference in
kind between you and super intelligent AI. The only real
difference between you and an AI creature of standard design
is one of origin. You were once a natural, but
you are now almost entirely engineered by technology. You are
perhaps more aptly characterized as a member of a rather

(35:18):
heterogeneous class of AI life forms. And so her thought
experiment ends there. But she's trying to sketch how it
would be plausible to imagine humans existing today actually becoming
machines little by little over time, and by extending their lifespans. Now,
I will say, I do think there's there's value in

(35:39):
this thought experiment, and I'm glad we're pursuing it. But
I also do feel like I need to flag that
I am significantly more skeptical of these types of common
extrapolations about trans humanism and artificial intelligence than I used
to be. I think my skepticism comes down to a
suspicion that scenarios like these make a lot of au

(36:00):
ssumptions that are just taken as obvious, but I think
are actually somewhat speculative. For example, would it actually be
possible to increase human cognitive capacity with neural implants that
that just seems obvious. It is taken as an assumption
because obviously, computers can do things that human brains can't do,

(36:20):
or at least they can do them at speeds that
human brains can't match. But what if there are inherent
biological throttles or gates on consciousness and cognition in brains
that make the neural cyborg not much smarter than a
human with access to a computer. What if there's just
something physically about the properties of brains that doesn't allow

(36:42):
you to augment them with technology like this, It just
doesn't work. Or what if becoming a neural cyborg with
computer enhanced cognition is actually a subjectively dreadful, miserable experience,
and it turns out that once people have tried it
and reported on what it's like, nobody wants to do
it because it feels awful. Yeah, I'm like, I'm thinking, like,

(37:04):
what you some sort of an upgrade you received made
it possible for you to say, well, let's say, be
better at personal finance. But as a result that means
that there is constantly an additional background narrative in your
brain and your consciousness about your personal finances. And maybe
that's good for for just you know, your your your

(37:25):
pocketbook and your investments, but ultimately maybe it sucks for life,
you know, because it's this is not the sort of
balance of inattention that makes life worth living or makes
it like like it was before, like it it changes
you to such an extent that you want to go
back you were, Like part of the joy of life
is maybe not thinking about personal finance all the time. Yeah.

(37:46):
What if part of what makes it fun to be
a human is not being a computer? And if you
the more you make your brain into a neural cyborg,
the more miserable your life becomes, and you and you
desperately seek to regress. Yeah. Another thing, What if consciousness
is just inherently non transferable to machinery. I don't know

(38:07):
this is the case. Some people do make this argument,
and I have no reason to assume this is true.
But I also have no reason to assume the opposite.
There's no reason to assume that you can actually upload
your mind to any kind of computer substrate. I think
this is just a big question mark. We just don't
know if such a thing as possible. Yeah, I mean
I tend to believe at this point that we could

(38:28):
create something that acts like us. You could create something
that is essentially like the the machine avatar of who
we were, or who we thought we were, who we
want to be thought of after the fact. But to
the point, like is that, I think when you start
asking more specific questions about like is that us? Then
I don't know, I feel like it isn't is it?

(38:49):
Could it be conscious at all? Even if it could
be conscious, is there any reason to believe that you
would experience it as a conscious continuation of your previous
mind or would it just be a conscious copy of you? Yeah?
Or I mean when you start asking questions like that,
and then you get into questions of like, well, and
who I am now? Is this really a continuation of
who I was five years ago? You know? I mean,

(39:10):
you start seeing all the flaws in this um narrative
of self and identity, and maybe it becomes maybe that's
the thing. Maybe we reach a kind of we reach
a point where we realize none of it is real,
Like there is no real continuation of the self and Therefore,
why not create like three different machine avatars of myself

(39:30):
and have them continue my legacy for me? I just
want to mention a few other questions that just popped
into my mind this morning. Uh, what if there are
actually hard limits on certain kinds of intelligence, whether you're
talking about a biological brain or a computer. What if
certain types of complex problem solving within a coherent agent system,

(39:50):
meaning like you know, a single sort of mental workspace
that always that is coherent and communicates with every part
of itself. What if there are limits on what kind
of intelligence can happen in an agent system like that
or different thing? What if biological organisms in general, even
across the galaxy, have an overwhelming tendency to revolt against

(40:13):
the cultural transition to machine life and will always or
almost always end up engaging in something like Frank Herbert's
but Larry and Jihad, you know, where you shall not
make a machine in the image of a human brain. Yeah, yeah,
you want to end up moving towards that sort of
Star Wars model where yeah, you have all these advanced
machines everywhere, but they're only working as servants, you know. There. Uh,

(40:38):
with a few exceptions that I guess kind of proved
the rule in that universe. So anyway, literally hundreds of
questions like this I think I could list, and they
start coming to mind when I think about it. And
while I don't assume that any of them are strong
enough to completely disable the trans humanist proposition, I also
wonder if some trans humanist and super intelligence thinking is

(40:59):
too wick to hand wave past these kinds of questions.
But like I said earlier, I do think this type
of scenario that Schneider is talking about is plausible enough
to entertain as a thought experiment, so I want to
keep going with it. And one thing I will say
in favor of of her argument is that, at least intuitively,
I think her timeline is reasonable, meaning that I think

(41:21):
if it is possible to create an AI super intelligence
and that humans or their biological alien counterparts do at
some point merge with or fade into that machine AI
super intelligence, I don't see why it would take more
than a few hundred years after the invention of computers
basically for that to happen, And even if it took

(41:44):
tens of thousands of years, I think Schneider's point on
this first point she's making is basically correct, the time
between when a species starts technologically interacting with the universe
beyond its home planet and when it becomes dominated by
post biological intelligence. If this is possible, that that time
gap seems very small and vanishingly small compared to the

(42:06):
lifespan of a planetary biosphere. Yeah, so you come back
to that scenario that show Style was talking about, where
once you're detectable, it's just a matter of time before
the machine administration moves in. So one instantly think that
you can imagine the the the aliens out there, if
they're listening in on this, they're like, well, should we
contact them now? They're like, well no, they're they're about

(42:27):
to change administration, Like the humans in charge now are
about to hand off in relatively little time from our standpoint,
two machines that will be it'll be just easier to
communicate with those machines and we'll we'll there'll be a
lot more pleasant to deal with as opposed to these
organic beings. So yeah, I would say I'm more bullish

(42:48):
on the second half of Schneider's proposition here than the
first half. I don't know if the age of machines
is coming. That's a big question mark for me, but
I will agree that if it's coming, it's coming very fast. Yes,
and if it is coming, we welcome our machine overlords.
But anyway, that was all just Schneider's first point about
the short window of observation. A couple of other points

(43:09):
that are quicker to make. The second one that she
makes is the greater age of alien civilizations. So here
she cites some pre existing statistical work making the point
that I think show Stack made this point as well.
If you assume a random distribution of biological evolution across
the galaxy, most alien civilizations should be expected to be

(43:30):
millions or billions of years older than us. So either
there's something very special and rare about Earth life, or
we're one of many planets with with with powerful intelligence
and civilization. And if we are, we we should expect
to be on the young side of that equation. So
if you couple this with the previous points she argues,
you start getting toward an interesting conclusion. Again, these two

(43:53):
points are, on average, we should assume that other alien
civilizations have been around for millions or billions of years,
and on average, alien civilizations transform themselves into post biological
superintelligence is very fast. There's a very short window of uh,
technological civilizations that are still biological in nature, and so

(44:14):
if you put those things together, you should expect. Yeah,
if we're meeting something, it's probably post biological. And I
will say, as far as my reaction, again, I have
lodged my moderate skepticism about the trans humanist and AI extrapolations,
mind uploading and so forth. But I followed the argument
so far. Her third point, and I think this is
an interesting one. She says silicon is a better medium

(44:37):
for intelligence, at least better than carbon, and this one
is interesting. Basically, Schneider argues that carbon based life forms
will recognize the inherent physical advantages in transferring themselves into
silicon based machines. Again, you know, flag my skepticism about
mind uploading, but if it's possible, okay, I follow the argument.

(44:58):
She writes, quote, silicon appears to be a better medium
for information processing than the brain itself. Neurons reach a
peak speed of about two hundred hurts, which is seven
orders of magnitude slower than current microprocessors, While the brain
can compensate for some of this with massive parallelism features,
such as hubs and so on. Crucial mental capacities such

(45:20):
as attention rely on cereal processing, which is incredibly slow
and has a maximum capacity of about seven manageable chunks.
I did not follow up on what she means by
chunks there, but she cites Miller from nine. This must
be a computational science paper. She goes on further, the
number of neurons in a human brain is limited by

(45:40):
cranial volume and metabolism, but computers can occupy entire buildings
or cities, and can even be remotely connected across the globe.
Of course, the human brain is far more intelligent than
any modern computer, but intelligent machines can in principle be
constructed by reverse engineering the brain and im proving upon
its algorithms. You know this. This reminds me how in

(46:04):
in Ebanks's culture books, there parts where the machines are
working with humans because you have human characters that are
playing an important role, because that that makes it an
interesting story. Um. But the machines, of course are communicating
with each other. The minds are communicating with each other,
it just blindingly fast speeds. And then when they need
to communicate with an organic being, it just like it's

(46:26):
just slow as Christmas, you know, it just drags everything
to a halt basically for them. Yeah, that's funny, and
it's also funny this last comment she makes, I think
is interesting about the cutthroat design idea, where an intelligent
machine could just say, like, oh, I could make myself
better than a brain just by figuring out how brains work,
reverse engineering that making myself into a brain, and then

(46:47):
upgrading myself. But anyway, altogether, Schneider thinks that these points
should convince us that alien civilizations that we encounter are
way more likely to be post biological machine in super
intelligent aies than they are to be biological organisms made
of meat. And Schneider also makes one point that I

(47:08):
think is very good if it's possible to become a
post biological super intelligence, but not a common fate for
all intelligent alien species. So maybe not all alien civilizations
go this direction. The ones we encounter are still more
likely to be the ones that do become post biological
super intelligent machines, because these beings will be better at

(47:32):
space travel and better at spreading across the galaxy. Think
about the fact that they have no biological risks from
space travel like we do. Yeah, show Stack gets to
this point as well, that yeah, there would still be risks.
Space is still incredibly dangerous, but the bio risks would
be effectively removed. And then since you would uh as

(47:52):
a machine intelligence, you would be effectively immortal um in
ways that in ways that even in a you know,
a very long living biological organism would not um. All
trips would be the same distance, that all trips would
have the same duration, because time kind of loses all
meaning if it takes you a hundred years, a thousand years, uh,

(48:14):
you know, several thousand years to reach the place you're going,
that kind of loses its importance. If there is no
endpoint to your existence. Rob nine thousand does not care. Yeah,
thank alright. So, in dealing with this question of post

(48:35):
biologic logical intelligence and potentially encountering post biological intelligence, one
of the big questions, of course is well, what would
it mean for us? What would what would the relationship be?
What would a post biological civilization want? And I guess
the first way to tackle that is to sort of
look at the precursor, what does a biological civilization want? Well,

(48:59):
as a Stephen Hawking and many others have pointed out
if we're to use our only model of intelligent life
that we have, which is us, then obviously biological aliens
would be interested in things like domination, resource acquisition, possibly
religious convergence. Or if we were to tie the Simpsons

(49:19):
into all of this, uh, you know, we could think
to the citizen King Treehouse of Horror segment, they might
be interested and interested in us merely in order to
point a giant space laser at another planet. So resources, yes,
but also maybe strategic location and some greater interstellar conflict.
I just had an idea that I don't know if
it makes any sense, but I was thinking about some

(49:41):
of the some of the horrors of colonialism on Earth.
We're not just about the extraction of resources from the colony,
but also about the acquisition of customers within a colony
for the businesses in the in the home country. And
I wonder could there be some kind of comparison to

(50:01):
this in in a galactic sense, like, uh, it could
be possible that aliens would want to initiate contact with
Earth in order to acquire some analogy to customers buyers
for their products. Oh my, Uh, nothing come into mind.
But I am sure this, this has got to have
been This has had to have been explored in in
science fiction, especially like like Reagan era sci fi. You

(50:23):
know this is uh that's commenting on capitalism and so forth.
Like in fact, like surely Philip K. Dick explored this
idea a little bit that could be right up his alley.
I can't think of one but that that would be
an amazing Philip K. Dick theme. I'm sure he did it. Yeah.
So again, you know, if we only have have our
own intelligence really to base most of this off on
as a model, but uh, this would it would seem

(50:44):
to present a rather dark scenario. Though certainly biological aliens
could be different. Uh you know they could they could
just want to be our friends. They could want that
they could have you know, they could come in peace,
as they said, I mean Stephen Hawking, Yeah, he was
very cautious about the idea of st He was like,
we won't we don't want anything to do with other

(51:06):
aliens in the galaxy because the chances are it would
not go well for us. But people who are involved
in set itself, in SETI type research, it seems to
be more often. I mean, I probably there's a selection
effect by nature of the fact that they are part
of this effort to reach out and establish contact with
other civilizations, or at least detect their presence. There seems

(51:27):
to be more optimism in the CETI crowd to me,
like the less a less of an automatic assumption that
the way aliens view us would be would be extractive,
and you know, more of an idea that uh, an
alien that as a civilization progresses toward the point where
it can reach out into the cosmos, it also maybe
matures like it, it reaches its own form of humanism,

(51:50):
and maybe that extends beyond its own species. Yeah, and
I guess too. There's also the argument it's kind of
like moving into a new neighborhood. Do you want to
say hi to your new nigh burst, uh, you know,
the first couple of weeks, or do you want to
wait until there's a conflict you know? Uh, you know
what do you want? What do you want your first
communication going to be to be? Because non detection is

(52:12):
not a long term possibility. You know, they're going to
see you leaving your house at some point. You're gonna
have that awkward moment where you make eye contact and
then you're like, oh, yeah, we never actually said hi
to each other, you know. So you know, a lot
of this concerns biological life. These questions and some of
these ideas don't entirely disappear when we consider uh post
biological life. But again the question is what about alien AI?

(52:35):
What would a post biological species want with us? What
would they as show Stack points puts it, what would
they quote find interesting to do? Um, which I like.
I like the way of pointing that out. It's like
it's it serves to a certain extent, it goes beyond
like goals and things that it needs, like what what
does it do with its time? Like? What is its purpose?

(52:56):
And show Stack points out that sci fi has certainly
explored this topic, but he thinks only three things seem
plausible enough to consider discussion. So, first of all, he
argues that since quote high speed computation requires compact configuration,
the machines would likely remain localized, and this would better
benefit you know, swarm or shared processing, so they wouldn't

(53:17):
be spread out over vast distances. They might be localized
into an area only thousands of light years across so
if you're imagining, you know, something like, uh, the post
biological necrons from Warhammer forty thousand, you know that they
just want to spread out all over the galaxy and
take it over. Like that wouldn't make as much sense
because they want to maintain maximum uh, you know, computational power,

(53:42):
so they're going to stick to their own kingdom. Coming
back to Susan Schneider, she argues that biologically inspired super
intelligences would would tend to have one or more what
she calls global work spaces, And I actually want to
read her quote on this because I thought this was interesting.
She says, when you search for a fact or concentrate

(54:02):
on something, your brain grants that sensory or cognitive content
access to a quote global workspace, where the information is
broadcast to attentional and working memory systems for more concentrated processing,
as well as to the massively parallel channels in the brain.
The global workspace operates as a singular place when important

(54:22):
information from the senses is considered in tandem, so that
the creature can make all things considered judgments and act
intelligently in light of all the facts at its disposal.
In general, it would be inefficient to have a sense
or cognitive capacity that was not integrated with the others,
because the information from this sense or cognitive capacity would

(54:45):
be unable to figure in predictions and plans based on
an assessment of all the available information. And this comes
into play here because it seems like a civilization based
on a super intelligent AI. Uh, if it spread itself
too far, it would become impossible to maintain a global
workspace at speed. It would start having information that was

(55:05):
not shared, and that would result in inefficiencies. Yeah, that
that lines up, but I think rather well with this. Now. Now,
the second point that Shostack makes is that given the
very short time scale for improvement, uh, it would be
winner takes all. The first machine society to rise would
dominate at least within a certain volume of space. You know.

(55:25):
Going back to point number one. Um Now, he argues
that there there could be a little wiggle room for
some machine civilizations to overtake elder civilizations. Um. But that
a sufficiently advanced machine civilization could rule its fiefdom indefinitely.
Um Now. But but I wonder if if another way
of looking at this sort of thing would be, you know,

(55:46):
a resulting confederacy of machine culture is a kind of
multicultural machine super civilization where maybe you have the you know,
the one older, more advanced, and you know, unconquerable um
machine culture, but then it in that absorbing other ones
that are part of it, that have some purpose or
role within the machine whole, but are not like the

(56:08):
driving force, kind of like subservient machine cultures, I guess.
And then number three, even for machines, he points out,
space is dangerous and our Winnian selection would take place. Quote,
if a machine exists now, it's because its mode of
existence has kept this device from natural disaster or possibly
even from deliberate disaster. If such a phenomenon exists for machines,

(56:32):
perhaps it makes a lot of copies, or at least
a few copies, updating as necessary. It does something to
withstand inevitable catastrophe. Yeah, that's very interesting. I mean to
pick up on this. There's no reason to say that
biological evolution is a process that is that is inherently
tethered only to carbon based organisms that reproduce, you know,

(56:54):
that have genetic code based on DNA, anything that's subject
to survival and reproduction, And I would guess that machines,
you know, computational machines would in some way be subject
to survival and reproduction. They can make copies of themselves, Uh,
they can iterate their code. That it seems like those
things would be subject to a form of natural selection. Though.

(57:15):
The interesting thing there would be, I guess, would would
it be useful to think about their code in terms
of something like genes, because of course, you know, genes
within biological organisms can have gambits to survive on their
own regardless of the success of the overall organism. Right Like,
if an individual gene in your body figures out a

(57:36):
way to make lots of copies of itself without regard
to the health of its you know, to to the
health of the body as a whole, it will do that.
You know. It's it's the genes just trying to get
out there. I wonder if you could look at individual
pieces of I don't know what code or nodes or
processing functions within a machine intelligence that would behave in
the same way. Yeah, yeah, so you it seems like

(57:58):
that idea you could you could come up with a
concept where a machine civilization would have a tendency to
colonize new areas, you know, because it would give itself
room to uh to copy itself. Uh. And then of
course you have to think about the constraints about processing speed.

(58:20):
It's that run, you know, having you know, sticking to
a local domain. But maybe that would allow for some
level level of mechanical butting to take place. Yeah, maybe
cutting off pieces of itself would actually make it more
resilient to say, infection by viral bits of code. Yeah, well,
you know, thinking about it even more now, So say

(58:40):
so you have this mechanical supercivilization and it's again, is
staying within a certain area? Well, if it is, if
it definitely, if it wants to survive, if that is
like a driving force in it, that is like just
coded into it maybe from its biological you know, elder creators,
then then perhaps copying its off not only within its realm,

(59:01):
but in other realms like that is one way to
try and survive, not only like nearby realms, maybe far
flung realms, you know, uh, you know, to get outside
of not only this star system, but this system of systems,
to get outside of the galaxy if possible. That's interesting, Okay, folks,
this is one of those episodes that went very long,

(59:22):
and we have decided it is best to divide this
talk in two parts, so we're gonna have to cut
part one right here, but come back and join us
on Thursday for the continuation of our discussion in part two.
In the meantime, if you would like to check out
other episodes of Stuff to Blow Your Mind, you know
where to find them in the Stuff to Blow your
Mind podcast feed, and you'll get that wherever you find

(59:42):
your podcast wherever that happens to be of the platform
gives you the ability to do so. Just make sure
you rate, review, and subscribe Huge things as always to
our excellent audio producer Seth Nicholas Johnson. If you would
like to get in touch with us with feedback on
this episode or any other, to suggest a topic for
the future, just to say a lo you can email
us at contact at stuff to Blow your Mind dot com.

(01:00:11):
Stuff to Blow Your Mind is production of I Heart Radio.
For more podcasts for my heart Radio, this is the
I Heart Radio app, Apple Podcasts or wherever you're listening
to your favorite shows.