Prof. Colin Humphreys - Faith, Humans and Robots
Just a fantastic idea, this meeting, bringing people, you know, Jewish faith, Christian faith,
perhaps people of non-faith even, here together to discuss this really important subject. So it's
a visionary meeting, so thank you very much for doing this. So let me press the right button.
So this is just my background, so very quickly, so I got a degree in physics and then I moved
into material science, which is, you know, adjacent to physics, it includes some engineering
and chemistry as well. I got involved in machine learning in 2016, so I wanted to optimise the
design of LED chips, there's lots of parameters in this, and so I had a very bright research student
who did this, and it was great to hear James talk about it, the companies he set up, so I set up a
couple of companies on LEDs, we made a way, discovered a way of making them more cheaply,
and then a company in England took over our companies is now manufacturing LEDs and
selling the LED chips and selling them to Philips and GE and companies like this,
and then in 2017 published a paper towards the end of last year on predicting laboratory earthquakes,
so at the moment the acoustic signals emitted before an earthquake happens gives you 10 seconds
warning, and with the laboratory earthquake predictions we can get one to two weeks warning by
the special signal that the machine learning identified from all the different signals
emitted before an earthquake occurs, and now the key question of course is, does this happen in
real earthquakes? This has been tested at the moment, and the initial results are very promising,
so we may have a technique using machine learning to give one to two weeks predictions of an
earthquake, which of course could save a number of lives, so that's a positive application of machine
learning, and I've got a great love of the Hebrew Bible, and I probably spend at least, as I do,
at least as much time reading the Hebrew Bible as I do reading the New Testament, so
great love for that. I published a couple of articles on a number of Israelites in the Exodus,
I wrote a book called The Miracles of Exodus in 2003, and I'll just mention I wrote another book
in 2011 I think on the mystery of the Last Supper, and this tackles or tries to tackle what's been
called the thorniest problem in the New Testament, and that is that Matthew, Mark and Luke rather
clearly describe the Last Supper as a Passover meal, whereas John equally clearly describes the
Last Supper as before the Passover meal, and whatever you think about the nature of inspiration
of the Gospels, it's a mistake you wouldn't expect anyone to make, because the Passover meals are so
distinctive, and these are Jewish people that have Passover meal, and so it's a very curious mistake,
right, and I mention this because in my book I believe the clue to this is in the Hebrew Bible,
right, so I argue from the Hebrew Bible that in fact all four Gospels are correct,
but it's a detailed explanation, so I won't go into it here, but the Hebrew Bible just,
so it shows in many ways the Hebrew Bible is just so important in understanding the New Testament,
right, the New Testament builds upon this, and lots of symbolism and imagery,
Hebrew Bible is just critically important for that. I'm interested in biblical chronology,
and the dates, trying to use science in particular to reconstruct biblical events,
like the times of Joseph, when might Moses have lived, when was the conquest, and so on,
so I'm interested in these, on biblical chronology, which is quite tricky, so with science you have to
move very rapidly because there's lots of competition throughout the world, biblical
chronology you can move quite slowly because the competition is limited, you know, just a handful
of people probably doing this, right, so I'm going to start by saying God uses big data,
so in the beginning God created the heavens and the earth, and so however you may think the heavens
and earth were created, this must have involved a huge amount of big data, just think of the data
required to create this marvellous university, and then think of the data required to create
the heavens and all the earth, so this was God using big data, if the creation was a big bang,
then think of what must have been encapsulated within the big bang, right, all the laws of
physics must have been there so that the expansion of the universe was consistent
and underpinned by these laws of physics, for example. A verse from the Psalms, the Psalmist says
you know when I sit down and when I rise up, you understand my thought from afar, right, so
the Psalmist is saying God understands his thoughts moment by moment, the implication is
he understands your thoughts and my thoughts moment by moment, think of the big data involved
if this occurs to everyone on earth, which is I think the implications of this verse,
it's quite a staggering amount of data that God must be involved with. If we go to
Jesus speaking in Matthew 10, not a single sparrow can fall to the ground without your father
knows it, so of course this may be a figure of speech, but what does that mean if it is
a figure of speech? He's saying that God has a detailed knowledge in real time of our world,
and just one final quotation, the very hairs of your head are all numbered,
and then Luke is sort of saying just in case you know too much emphasis is given to sparrows,
you're worth more than many sparrows, right, so the very hairs of your head numbered, again
maybe a figure of speech, but it does mean great detail, God has great detail about all of us in
real time, and then this is looking at relative worth, of course you know some people will argue
that animals all have the same sort of rights, this is putting that correctly, you know we are
all much more worth than sparrows in the sight of God, so God has very detailed knowledge about us
and our worth, so this is something on which the Hebrew Bible, as in most things the Hebrew Bible
and the New Testament agree, and that God knows about each one of us in detail in real time about
even our thoughts, and so just multiply this knowledge by the number of people in the world
today, and then add to that the number of people in the world in the past, and if you like add in
the number of sparrows, right, so this is just huge, sorry, and the future, yes thank you very
much, and the future, that's right, and I'm sure the rabbi would say that God's beyond time,
outside of time, so God has this information simultaneously probably, yes, so just huge
amounts of data involved, so God uses extremely big data, huge data, and God is good and God is just,
so God uses big data for good, and I tentatively put this up, God is big data, right, is that an
attribute of God in the same way that God is love and God is good, of course God is love and God is
good, and God is good and God is just are in a sense more fundamental, but nevertheless there's
so much big data involved in what God does, is that an attribute of God. So let's move on to
robots, and let's start, we're looking at robots versus humans, and the present, what is the
situation at present, so we all know this, robots can today be very useful in surgery, they can be
better than the best surgeon, because there's no shaking of an arm, of a hand, when the surgeon's
performing very delicate surgery, so very delicate surgery can be performed better by robots,
car assemblies of course, very precise car assemblies now by robots, remember they're not
always useful, and this arose this morning, and maybe people won't agree with this, but the
automatic trading system was triggered I think, that did trigger this financial crash, so even
today big data you know can have problems, and then robots are beginning to form relationships,
so there's something you can buy now, it's called the Jibo robot, and the advertisement says it's
the first social robot for the home, and then it says the relationships he forms are the single
most important thing to him, and the cost if you're interested is $899, so
this you can buy now, and people are using this to form relationships, and this is just
going to increase in the future, as the robots just get more and more powerful artificial
intelligence. Robots are being used as therapists, it's called the Eli robot, it's being used now
by the US army for post-traumatic stress disorder, and the interesting thing I put in yellow is,
there's been a survey of servicemen who've used the robot and used people, and they actually
prefer the robot to the real person, right, they prefer talking to the robot, they think the robot
actually gives better advice, and that's happening now, right, with a, yes.
with people, he charges $25, $250 an hour, right, and people talk to him because much, much freer
than they talk to the, to my nephew, right, and he sees patients, his dog, the dog sees patients,
right, yes, yes, people open up to the dog,
that's right, and probably the same with the robot, yes.
Although it's a difference of degree, I don't think it's such a new thing in some ways,
I was thinking of the statues of Easter Island, yes, they all face inland, and if you're living
in the middle of an ocean, they've lost contact with the rest of the world, then you build
pictures of, you build statues of faces, yes, yes, to represent the rest of the world pointing
inwards, so I think in the Old Testament it was all eyeballs, you know, but it was,
I think the desire to relate is so strong, you can certainly do inanimate things, and we have
done for a very long time, right, yes, that's right, and of course the ancient Egyptians
related to the statues of their gods, yes, yes, but, yes, now that's true, I do think this
relationship is at a different level, perhaps, in, yes, in that the robots are talking back to
the person, you know, and giving them advice, and it's at early stages, but it's showing what can,
what can happen now, yes,
yes,
right, right, yes, yes, because they're very clever, that's right, yes, yes, right, yes, yes,
yes, but I think that, that, this is clearly going to improve in the future, as robots, you know,
get more artificial intelligence and are trained, this is just a final example, this is, if you like,
messages from the dead, so this is a Russian woman whose partner died, and he sent her lots
of text messages and so on, and so she fed all these texts into what's called an artificial
intelligence chatterbox, and this then composed new texts based on, you know, the information that
it had, and every day her and the man's mother received texts and tweets from him every day,
which they find to be a great comfort, and then an American woman has built upon this more recently,
where she had, you know, warning that her husband was going to die, because he had, I think, cancer,
and so he lived for six months before dying, and she just brought a robot into the room, into the
bedroom, and this robot just listened to their conversations, and now the robot is sending her
messages, you know, after he's dead, I mean, it's slightly strange, you know, and slightly scary, but
this is what's happening at the moment, and these people are deriving comfort from it, you know,
it's superman's relationship with his parents made real, it's the AI superman had,
and now we're building it, right, right, yes, yes, so this is what's happening at the present,
and so, okay, where are we heading for the future? I have an amazing statistic, Amazon,
they've currently got 5,000 people working on the next Alexa, right, 5,000 highly trained computer
scientists, so you can just imagine the sort of developments which will happen from this,
there's just this huge, really huge effort going into artificial intelligence, and it's really
going to be very important for all of our lives, I think that's why this conference has so much
vision that we're having it now, because this is really going to impact people in a major way.
Stephen Hawking said robots could evolve faster than humans and be hard to stop,
I think we can say with certainty robots will evolve faster than humans, you know,
that we're evolving slowly now, robots are going to evolve very fast.
Many companies, universities, huge amounts of money are being invested into robots,
and this range of super intelligent robots is being developed, so this is a key question which
I think we have to face, particularly as believers in God, are intelligent robots
fundamentally different, or will they be, fundamentally different from humans,
or fundamentally the same as humans, obviously with some differences.
Stephen has also been expressing some concern about potential dangers that they offer,
are there ways that we can watch and prepare for that?
Yes, well I'll come to that in a moment, but people used to say, you know, if you're a robot
you get scared of, you just pull the plug out from the wall and turn the robot off,
but the problem is, once you get robots communicating with each other, if you do that to
your robot, you may get this armory of robots walking down your road, you know, and come and
kill you, so you've got to be quite careful of these simplistic answers, including the plug as
it were, yes, right, right, sorry, so we need to ask ourselves, and I think it will raise
important questions on the nature of humans, so this is going to be very naive, and people here
will know a lot more than I do, but you know, a hundred years ago, people thought, many people
thought humans were body, soul and spirit, and then today, most Christians, and I think most Jews,
but you know, I take advice from you, because it'll be very interesting to hear it, are essentially
dualists, so they believe that humans are body and soul, or if you like, body and mind, which are
separate but linked, and I think we'll take your advice, because, you know, people here will
know more than I do about this, however, most scientists today are what are called monists,
so they believe the soul, and the mind, and the brain are just the same thing, so humans are just
mechanisms, so I think nearly all scientists, or psychologists, and other scientists,
consider myself a dualist, but I say that there's a difference between the heart and the mind, so why can't you just say that it's programmed in a way you're comfortable with,
I mean, that seems to be a perfectly reasonable scientific statement, isn't it?
Right, right, yes, yes.
If the software is generated by the hardware, then it's, well, they're linked, it's machine software, machine-made software,
it's different, so it's not a human chemical, it's carbon-made thought,
I mean, it's still a program, I mean, it's the hardware and the computer that's the program too,
I don't see any difference.
To reflect three points that philosophers would take an exception to, so one is the dualism, because
most of Christian history, certainly, it's not been two substances, it's more integral than that,
let's go back to the ancient Greeks, a living thing is different to a dead thing,
they call the form the soul, so that is formed in programming, no, a huge difference,
so the living and growing thing, all the matter is organized towards a final flourishing,
and here again is a huge difference between living things and machines, because machines
are lots of two-body systems put together, like cogs and wheels, and living things work in a
fundamentally different way, it's much more emergent towards final forms, and physics has
not really crossed that gap between the two-body system and the true multi-body system,
we just approximate the world, the machine-like parts of the world, so that's a second problem,
you've got the soul-body issue, the machine-living thing issue, that was a
firm one as well, I don't think I'll put it out there, but although, you see, it's interesting,
see, what a lot of, the mistake people are making today, they're taking the cartesian model of soul
and body, and they're translating it to the brain and body, it's exactly what is going on there,
and there's a whole book about that, so philosophical foundations of neuroscience,
well and this is the question that comes up in cloning so much, in not cloning as in take myself
and make just the other body, but create a new body and transfer the consciousness into it,
which you find in a lot of sci-fi, is that transferred conscience, the same soul,
is teleportation, not teleportation, is the teleport machine from Star Trek, is that the
same entity that's been recreated on the other side, or have you repeatedly killed the individual?
So there's a huge volume of controversy over those three issues, so just be aware of that.
Yes, no, no, thank you, I'm aware that there is a case, that's right, and so this is
very simplified at the moment, so, but most scientists would basically say we are mechanisms,
right, full stop, and certainly in the UK, many scientists who are Christians,
and I don't know about the gay Jewish people, so I'd be very interested in that,
are monists, right, so many psychologists certainly in the UK who are Christians and
scientists would be monists, I mean that we are, our brain is just, we are physical mechanisms,
right, and that's what we are, only physical mechanisms, and I'll just come, I'll come to
consciousness in a moment, because what about consciousness, right, so many scientists who
are monists believe as we evolved, our brains became more complex, and one explanation of
consciousness is that consciousness then emerges, so when your brain reaches a certain level of
complexity, when you have enough neurons, and particularly when you have enough complexity
of interaction between the neurons, then the consciousness emerges, so the scientists will say
a person certainly has consciousness, but that is in their brain, it is their brain, right,
again it's physical mechanisms that neurons are firing in a certain way, and that produces what
we call consciousness, so this is what the vast majority of scientists who are not Christians
believe, and, and in the UK a lot of scientists who are believers in God will believe this, and
I am, I'm here to ask these questions, right, I say which, are we believing a duality,
well the soul is something distinct but integral, or are we saying there's no soul, if you can see
that difference, and let me just continue, so I say for discussion, I'm not, I'm not proposing any
particular view, but shall we move on, and then we can, okay, so that's fundamental nature of humans,
fundamental nature of robots, so if humans are body plus soul, which is the traditional viewpoint,
duality, and robots are only bodies, machines, then I think humans are fundamentally different
from robots, and I think that was always, it was seen that would always be the case,
right, however, if humans are only bodies, and robots are only bodies, then in the future,
you can envisage a situation as robots evolve, and their complexity of their brains increases,
or the electronic brains increases, they could acquire consciousness, right, so if you believe
that we're just machines, robots are clearly machines, then if consciousness is an emergent
property of complexity of brains, you can, you can foresee the possibility that robots might
acquire consciousness, and robots could become fundamentally the same as humans, so that's
another point I raise, I'm not saying I believe this at all, but I raise this point, because some
people are saying this, and it's something which I think, you know, believers have to consider,
so it's for discussion, shall I move on, yes?
Do you think that, like, it seems to me that, that we don't know anything about consciousness,
you know, I mean, I've read entire books, yeah, right, like Consciousness Explained,
I get in it, and I think, like, I'm more confused about this than before, so, I mean,
it just seems to me that, that we, like, we don't have any insight whatsoever into exactly what
consciousness is, and that's right at the heart of this question, yes, it is, it is,
yes,
yes, it does, yes, that's right, yes,
it's a very interesting book, because he argues that consciousness did not exist in human beings
until a few thousand years ago, but we haven't changed in how to get, yeah,
no, and you're denying consciousness to non-human, well, yeah, that's right,
I think the evidence is that some animals have a limited amount of consciousness, yeah,
yeah, I mean, that seems to be the point, it's not necessarily a discrete on-off, no,
but it's consciousness of an evolving degree, yeah, which is consistent with,
as their complexity of the brains increases, you get more and more consciousness, well,
and it's also consistent with the development of humans, I mean, a new embryo has no consciousness,
but certainly another human doesn't, so somewhere along the way, yes, yeah, that's true, that's
right, yes, yeah, that's right, but it might not just be complexity, I mean, no, it might be some
other, we don't know, yes, maybe some other feature, complexity might be necessary, but not
sufficient, yes, have you considered within this context the issue of people who suffer from mental
degeneration of some sort, either through injury or disease, you have a, do you have a change of
soul as well as a change of consciousness? Right, I depend, I guess a monist would say it depends
with what part the brain is damaged, this would cause a medical ethicist in the early 70s to say
that someone was unconsciously dead because he lost his soul, and that's what they wrote,
once you're unconscious, you're permanently dead, but permanent coma, irreversible coma,
was actually death because you lost your soul, so that's what Christian Gardner,
when he gave his first talk, that's why people were in coma that day. Gosh, I didn't know that,
that's very interesting, yes, yeah, right, I'll just, so if intelligent robots are
humans, then some robots will be evil, just as some humans are evil, and that's something I think
we need to consider, right, for the future. Can a machine be evil? Right, that's what you said,
well, if, but if they're intelligent, if we believe that they have artificial intelligence,
then, well, can a machine be evil? Well, if it's got a limited consciousness, then presumably it
can be evil, but it's a good question. But you see, I'm not sure if Jewish people believe in
the religion, go on. So we've already seen with some of the algorithms that do linguistic work,
there was Microsoft's AI that became racist rather quickly, we've seen, I'm sure all of us at one
point or another have had the algorithms that credit card fraud companies use to fight us
because they make poor decisions, and so where do you take something from being poor decision,
from being evil, where does a racist algorithm fall in? How do we put our own morality on software
that is learned by watching human behavior? So you have to define people. Yes. For example,
but is an algorithm intentional?
So if you're right, if you define that killing all these people in this room by gas is evil,
but I don't think the machine can be evil, you could say the same thing would be done
in both. And these are the kinds of things where I don't think that there is a generally agreed
upon truth, and that's what makes conferences like this interesting. That's right. So I think
just that we need to look ahead, you know, 10 or 20 years and look at probably huge developments
in robots and where they may give the appearance of having consciousness. We need to consider that
as a possibility. Yes, right. Is your position that of something like vitalism, that there's
a difference between a machine and an organism which in some way is not subject to laws of
physics? I'm not sure I'm understanding. No, no, I'm not putting a position here because I'm not
sure. No, but I certainly don't believe in vitalism. No, no. What's vitalism, sir? It's a
search that, um, that, uh, Oh, yes.
I'll be a heretic for this one because I don't think, uh, atoms, whether they're embedded in
living things, have laws in the same way as free atoms. We write laws of physics,
two-body systems in the vacuum, um, and then we have perturbations. The atom on the tip of a horn
of a galloping rhinoceros is not following a law. It's moving as a rhinoceros moves in lots of
planes. It's part of the whole physical system that is subject to law. Here we will have to
disagree. I think, uh, I think once, once the thing's embedded in a living whole,
come back to the usual poles, then, um, then it was more than the sum of the parts.
Yeah, well, that's an interesting discussion. That's right. Yes. It's the whole grace of the
sum of the parts. Yes, yes, yes, yes, yes. Gravity and non-gravitational forces.
Here's a funny one. If you put three, three masses together,
but we get less than the sum of that part. Um, yeah, exactly, exactly. So you see,
even something as simple as mass and gravity, um, but we can predict that by
climate diagrams. Oh, yes, yes. But it's, it's an illustration that things, things as poles
already behave a lot differently. This is already covered in the laws of physics,
but if you're, if you're proposing that they, that they act differently, I mean, I think most
physicists would be skeptical of that, but I mean, it's an interesting, would there be any way to
test it? That'd be an interesting thing. If it really behaves differently from what, what we
believe the laws of physics would give for the situation, then in principle, you might say
there's so many ways to test it, but maybe in practice it's impossible. We don't have laws of
zoology. No, no, but I just, but I mean, if you just, if you say you've got some quantum state
for your system big enough to contain for the thing to evolve by the, by the laws of, by the
laws of physics, then it's a question, does it really evolve by those laws or does it,
or is it different because somehow it's in life? I mean, I, I mean, I think most physicists would,
would, would believe that it's the same laws of physics that apply. Of course the laws have to
be applied to a quantum state that has a huge number of particles. This is where we have a
philosophical discrepancy because we have to decide what laws of physics mean and are they
simplistic? Are they simple representations of the simplest systems? Are we dealing with the
universe of space-time and a box of laws or are we doing things that have powers?
But it's a, it's a many-body problem, isn't it? You're talking about, and yes,
and I, and we, I think we're saying that the laws of physics will,
would apply to that but it's just too complex for physics to handle at the moment.
Yes.
The, the whole is more than some of the parts. I mean, here's something we don't get in laws
of physics but we do get in complex systems. We get teleology. We get systems moving towards
final forms and that gives us an arrow of time which you don't get just from Newtonian laws
or two-body systems. It would be fantastic in new metaphysics where we get complex systems.
But I mean, all that can occur without, under the laws of physics. I mean, there's nothing in,
nothing requires you to advance the laws of physics. I think they're great for what they do,
the quantity of two-body systems. But as soon as you get complex systems, you get a much richer.
Right, right. But the laws of physics, I mean, of course, it gets too difficult to calculate but,
but I mean, the laws of physics are believed to govern. It determines what happens.
It's not always deterministic and it's not, it's not just a matter of two different factors.
So, can I suggest that we talk about this later because it's slightly off the robots, yes.
Yeah, that's what, it's a really good thing for discussion but it's getting a little bit
away from robots. So, this is just a finite, evil humans could use evil robots. That's
something to think about, yes. And then robots and humans, we cannot read human minds at the
moment and perhaps we will never be able to but we look at their actions and they do.
What do you mean by read human minds? Because we're now getting to the point that we can tell
what people are thinking. I, I'm sort of saying that at the moment I don't know what your next
action will be. Like you may put up that glass and have some water from it and I don't think
anyone at the moment can say what your next action will be. So, we can't read your mind
and, but we can look at what your actions are and get information about your mind.
So, if you kill someone we might think you're a murderer, for example.
Sorry, that's a terrible example, but yes, you know what I mean. But basically you take
behaviorism, basically. Yes, yeah, that's right. And I'm saying, and the same is for robots. With
artificial intelligence, so, so when, was it, was it Deep Blue? The, the, the, was it IBM computer
which, which, which beat the Kasparov's, Kasparov, I mean Deep Blue was programmed as we all know
and so the, the moves that Kasparov made, which were televised at the same time, which chess
players analyzed, we say we know why he made that move. I mean all, all expert chess players
programmed, were involved in the program of a Deep Blue and so Deep Blue knew all the classic,
many classic chess moves and, and, and beat Kasparov's just, he had just a greater memory
of all these moves. When AlphaGo, which of course played by itself to beat the world's best Go
player, people are still analyzing what AlphaGo did and still don't analyze, understand some of
those moves. So, you know, with artificial intelligence we just do not understand how
the robot mind is working, but again we can go on the behavior and we don't know how robots come to
their decision when based on big data and it's somewhat similar to humans. So this is a sort of
parallel, I'm suggesting, between robot minds, if you like, and human minds. And at the moment
robots are working alongside doctors, I think in a number of countries, but the situation at the
moment is they're helping the doctor, the doctor's in charge. So the doctor may have in his surgery
a robot, the patient comes in and describes the systems and the robot searches its great database
and so on and comes up with some diagnosis and a solution and the doctor listens to that and can
override robot decisions. But you can envisage in the future that when people analyze whether the
doctor overriding robot decision was a good decision to make, it may not be a good decision
in many cases. So you may in the future have a situation where the robot is actually in charge
of the doctor and listening to the doctor's advice and then making the final decision and that
that could happen if analysis shows the robot's making better decisions than doctors, right. So
and I can envision this sort of thing happening because I think there will be a time when robots
will make better decisions than doctors probably. One study was already showing that last week.
Okay, really, that's interesting, right. And then, okay, slightly different but...
Right. Professor came up. Okay, should I just hurry up and finish?
Okay, right, okay, okay, fine. So this is an interesting question, I think. Should robots
replace jurors on juries? And I'll give an example where I think this would be good. So I've been
involved with some patent cases and patent law and what the American lawyers have told me is that
juries in America just do not understand the patent cases, they're too complicated,
but what they do do, what jurors do in America is they listen and they like the most attractive
lawyers and solicitors. So the people who win cases, patent cases, are the people who have
either the most physically attractive women speaking or the most handsome men speaking,
right. And so that's what wins the cases. And so you can envisage in these situations
having robots would be rather fairer than having humans. But then you can say,
would this be good for any criminal case because the robots are taking the emotions out of the
situation? But you could also argue maybe the emotions should be there in the situation,
right. So it's quite a complex sort of situation. But these are questions which I think
are questions which will need to be asked in the future. And there's all sorts of situations
like this. So these are just two examples. This I think is the final slide I have.
Did AlphaGo actually enjoy beating the WorldGo champion? Right, so we're,
and we all say no I'm sure, but you know in the future again if robots become very complicated
and develop some sort of consciousness, might they enjoy this? I mean I don't know the answer
to these open questions. But I think they're, yeah, should be asked. Yes.
Right, well that's right. So scientists will say enjoyment is a chemical reaction in our brain,
that's right, that's right. Yes, but right, but we experience what we think is enjoyment,
that's right. Can robots have passion? And again we will say not, you know, not now, but
maybe, you know, we should envisage this as a possibility in the future.
Right, if a robot football team beats Manchester United will it be happy? So you can envisage,
you can envisage a robot footballer who would actually be better than any footballer,
right, so they'd always, you know, score goals and so on. So you can just imagine this will
happen at some stage in the future. So you can imagine a robot football team could beat
Manchester United, would that team be happy, robot team? Perhaps even more importantly,
would the football supporters enjoy watching? They might enjoy watching the first game,
but would you turn up to see a robot team play? Supporters are humans? Sorry? Supporters are
humans, sorry, right, yes, thank you, it's a good question.
Yeah, so would humans enjoy going to see, now I think they wouldn't because people,
there is football, and it's depressing to watch, but still people cheer because they move for 30
seconds around themselves, there is such a competition, right, okay, yes, but everybody
cheers, but what, okay, I didn't know that, so it's good to learn, but would they enjoy going
to see, you know, queuing up to see Manchester United and standing in the rain for an hour and
a half or something while the match is played, and week after week? I suspect not, but I may be
wrong, because part of watching football matches, the emotion is involved, and part of it's watching
people make mistakes. Have you seen, oh, I just wrote down her name because I keep forgetting it,
it's Kate Darling's work at MIT Media Labs. No. So one of the experiments she's been doing is
trying to understand if people can have empathy with the robot, so she built these darling little
dinosaur robots, and she has groups of people work with them, and then after they've been working
with the robot for a while, and you have to give it a name, and they have to speak with it,
she asks them to kill it, tear its head off, and they can't, and there's money involved,
they will get money if they kill their robot, and they still can't do it. Yes.
And then there's the TV show Junkyard Wars, where people would get behind one robot,
killing another in a WWF style robot gladiator game. Humans are able to empathise with the
serious events. Yes, that's right, and taking it slightly more seriously, a robot in an old
person's home, interacting with the old person, that person really has a relationship with that
robot. Yes, that's right,
we can sort of see a robot in an old person's home is very, very helpful, but there's even a
danger there, and that is that relatives may feel we don't have to go and visit, you know,
granny, because she's got the robot.
Right, yes. We've seen that happen with pets, so I don't see a reason that it won't eventually
happen with robots. In the United States first. No, it'll happen in Japan first,
because in Japan there's actually cases of marriage with robots. Right, okay, yes.
So this is just going to end on a slightly more serious note, I thought at the end.
So this is, you know, these are, I think again, important questions, and how does God interact
with us? And I guess if we're idealists, we'll say God interacts with our spirit,
which is an integral part of our brain, but nevertheless that's the answer people give,
and if you're a monist, you'll say God interacts directly with our brains, I think, but this is
all for debate. And then life after death is an interesting question, of course. So if you're a
dualist, then our soul survives to be embodied at a future date, depending on one's religious
beliefs. But if you're a monist, then I think you would need to say the big days that God
already has on us survives to be embodied. And it just occurred to me that, as I was writing this,
you could envisage an argument this is actually better, because if, let's say a person has
Alzheimer's disease, or some mental dysfunction, then when they die, your soul at the point of
death, you probably feel, and one suspects that God might feel, that that's not the person to
be embodied in the future. You want the person at the peak of their powers to be embodied.
And so you can imagine that then God uses his great database to have this person who goes to
heaven, who isn't, who's without the Alzheimer's, right, who's in his prime of life, whatever age
that might be. I'd encourage you to read Morlach 36 Wakes, it's a science fiction book that deals
with, instead of having the three laws of robotics, it becomes the laws of cloning,
where you can transfer your consciousness from one body to the next, and they have to deal with,
are you allowed to reproduce in your secondary body? And you decide no, you only get to reproduce
in your primary body. Then there is the question of, can you cure what ails your body when you get
passed into the next one? And it's determined that injuries you can, but birth defects you can't.
And so these are all the kinds of ethical questions. As we start to perhaps reach the
ability to not just recreate the liver, but recreate the leg, the arm, transfer the head,
what are the ethics, and what is simply the body, what is the soul? Can you kill one without killing
the other? Is it left to the book? Yeah. I hope that when I'm resurrected, it's better than I
ever was. Yes, right, yes, okay, so right, so and I think I would, you know, I would very tentatively
say that, you know, our resurrection bodies will probably not be carbon-based and DNA-based because
they are subject to decaying death. Our resurrected bodies would be subject to some different laws of
physics, where we are immortal and don't have any pain and no injuries. Does that answer our
question then about... No, I'm discussing an intermediate case. If your body is dead, you're
not resurrected, but you've got it flown. So option C. If we ask if consciousness emerges only from a
carbon-based or organic base, then that seems, that's answering the question about robotics,
or also if it's not necessarily carbon-based or need to be, then that also would suggest that,
you know, it's answered about robotics too, that a robot, given a large enough neural network,
should have self-awareness arise, if it need not be carbon-based, organic-based.
And that, I read way too much science fiction. The Bobverse books actually,
the Bobverse books actually take on this idea of you can download the human mind, and the idea of
when they first started downloading the mind was eventually in the future, instead of cryopresing
the brain, you download the brain and someday you'll be able to recreate the body. They never
figure out how to do that, but then they need to create advanced spaceships because the earth is
dying, and the only thing that can handle the spaceship is the human consciousness downloaded
into the spaceship. But then it's like software, this is the model that Frank suggested before,
that it's all software. And the question becomes, if you download the same neural network into
multiple computers, that's no longer the same person, is it? It's identical down to the quantum
state, it's the same person. But if it's different, it's the same as identical twins. Identical twins
are the same genetically, but they have different cells. They're different people. They become
distinct at the moment the blastocyst collapses. And the idea here is the second you download the
brain, you're going to have quantum differences between them. So every reproduction is now
creating something new, but does that something new have a soul? These are all interesting
questions that luckily we don't have the technology that requires us to answer,
but we can begin the discussion before we begin building the technology.
Okay, so talk of science fiction just reminds me that Douglas Adams in one of his books
has a monk with artificial intelligence who then prays for other monks.
Yes, so later I can devise. The electric monk.
Right, that's right, yes.
Are you done? Because I wanted to comment.
Oh, very quick. I just got one more thing. I think this is the end of it. So this is,
yeah, again, a highly controversial question. But I mean, I'm sure we'll all say no to this.
But you know, it's just worth thinking about.
But the goal is the big data of everything, the big data of robots also.
Yes, that's true.
And it's not dissimilar to the idea of people on other planets.
One of my greatest concerns is what will happen to our society when we do first find evidence of
life on another world? Because will religion be able to accept the idea of consciousness,
of knowledge existing in completely separate life that may not share any theological overlap
with us? And this is something that Mary Norris Russell deals with very nicely.
Not in our past lifetime.
Yeah, yeah, it's a very big number, very small.
As far as the whole body transplant, that's exactly why we were spying at C5 and so forth.
Because all the plumbing has already been done. That's easy to do. In fact,
the whole body transplants have been done with rats and with monkeys and with dogs.
They only get about 10% of the function because of the spinal problem. And that was the first
insertion point for why we did this. So then you'll be able to take a brain dead patient
and use their body and a person who's lost their limbs and put their head in their body.
As far as the information and matter, the information is primary, the matter is secondary.
When I first type on my computer, that's going into direct. When I hit save, that's either going
into a magnetic memory or it's going into a flash drive. And when I upload that, that is going
through a wave in the air through some electrical wire now that's going to end up in a server farm
and back in flash memory. So it's gone to all these different mediums, but it's the same material.
And then first wrote it on a piece of paper and then type it in.
It's the same information.
It's the same information. So the information is primary, the matter is secondary.
Jesus said, every careless word that a man speaks, he shall render account for in the
judgment. By your words you shall be justified and by your words you shall be condemned.
So the information remains. We believe the words of Jesus. The information remains
after we're dead. The information is there. It remains. And you get the same understanding
that Abel, though he was dead, he was still speaking. The body was dead. The words were
there. And so this whole idea that information is primary, the matter upon which it is saved
is secondary. So where does God save the information? We don't know. But you can put
all the information in humankind that's ever been in anyone's brain and you can add that up.
And you can store all of that in a gallon of water if you had each one of the water molecules
encoded in a bit of information. So it doesn't take much volume. However he stores the information,
but the matter is secondary. And you see the same concept. It says in the beginning was the Word
and the Word was with God and the Word was God. So information, the Word, was in the beginning.
And then in verse 14 of John chapter 1, and the Word became flesh and dwelt among us.
The information was primary. The matter was secondary. It took on flesh. And so then you
had this merging of the information with the physical matter. And the last thing I wanted
to point out was robots. There's all this presupposition that we understand the origin
of life. People make grand proposals that are all a bunch of nonsense. And I'll show you an example
of where this hits with robots. If we understand the origin of life and we understand molecular
structure, we understand how to do things with molecules, we have ubiquitous life around us.
It's all over. All over. We have never, never built a robot that is built out of carbohydrates,
proteins, lipids, and nucleic acids. Never. Why? No, no. This is, if you say you understand,
so this is a different topic. So if you're going to say that you understand these things,
why do we build our robots out of silicon and composites and plastic and wires? Why do we do
that? Why don't we just build it out of the four elements we know can make very good functioning
organisms? Why don't we build a robot that looks like us, that's made out of carbohydrates,
lipids, and protein? Way too complicated. So we've made the suggestion we understand how life forms.
We can do all these things. We can make artificial cells, which is also a bunch of nonsense. The
authors of that paper never even called it an artificial cell. But this idea that we never
build a robot that's out of the same material that we've chosen, how far we are from understanding
what life even is, even if you put all those constituents together in the same place. We don't
even know what it is that gives a spark of life. What is life? Ask scientists to define life. Very
hard. Is it just ionic potentials or is it much more than that? Very hard to even give it that
spark. So it underscores how far we are from understanding the physical matter, which is
secondary. The information is minor. We see that same sort of thing. Yeah, yeah. No, I very much
agree. I think that's right. It's a good point. I think there's a lot we don't understand. So
this becomes maybe just the beginning of something bigger in the future. Right. Thank you very much.

