Is the Singularity Coming?

were slacking. They say it's near. Well, yeah, depending upon whom you ask, it's gonna be um like any day now. So we thought it's high time we actually take a look at the singularity, talk about what is this idea of the singularity, What does it mean? What do people who are super duper small aren't have to say about it on one side or the other, And kind of, I guess, ultimately kind of give our own opinions about what we think. Keeping in mind that we are not

futurists in the strictest sense of the word. This is more about how we kind of feel about things because we're not computer scientists. Were futuri ins or future ologists is what we are future rangers. So um, yeah, so let's let's get started. Let's let's talk about what the singularity is. So, what the heck is it? You know? I think in the popular consciousness when you mentioned the singularity, they're aware that it has something to do with technology

and something to do with the future. But other than that, it's just kind of this vague mental piece of stock art where there's some wavy lines and a picture of like a brain and some electricity. Right, yeah, maybe maybe brains suddenly going up into disappearing into ones and zeros and joining a giant cloud of them or something. Yeah.

So I think the general understanding of it is fairly vague, and so I'd like to narrow that down and focus it on what does this mean in the most basic technological sense, And what I'd say is that it's usually used to mean a time when humans develop superhuman intelligence through some means, whether by creating artificial intelligence or by amplifying their own intelligence, or somehow or other stumbling on an intelligence greater than that of what humans are normally capable.

And often they also go further into saying that in this world where we have developed superhuman intelligence. We will we we in the present right now, will be unable to kind of predict what that future is going to look like because we won't even be able to recognize it. Yeah, by our very nature, by our limited human intelligence, we cannot conceive of what that future will be. It's impossible. It's beyond our our what our limitations are are around us.

The idea also tends to involve not just a stasis, like we reach a point of plateau of very high intelligence, but it involves this idea of a vast acceleration in learning and improvement, right, so educational specotification, Yeah, or a Moore's law that applies directly to how smart we are, so that we're constantly getting smarter than we were earlier.

And it just keeps except that instead of Moore's law, which is steady you have you know, you have this this uh, this doubling growth, but it's a steady doubling growth in the singularity, and at least some versions of it, it's an exponential or or lager eithmic growth to the point where, uh, you know, it's you're not only getting smarter, but you're getting smarter faster, right, and in that sense, it's sort of a double impossibility for us to understand

what it will be like. It's not only smarter than we can imagine now, but it's getting smarter faster than we can imagine now. Right, So, there are generally speaking four different agreed upon ways that this singularity could come about. These are are proposed by a science fiction writer and mathematician and mathematician Verne Buddy Verne, I don't I don't actually know Verner of into it anyway. He's a he's

got this idea. He had this idea where he kind of proposed four ways that we could reach this superhuman intelligent future. And these are four pathways that are not necessarily completely um isolated from each other or compartmentalized. We may see some weird um combination of different paths that lead us to this, assuming that the singularity is indeed something we are going to hit at some point time or another. Okay, what are they? So you've got superhuman

intelligent machines. This is the artificial intelligence route. I think this is like the most classic sense that people would imagine the most it's just saying, like developing computers that

are beyond human capabilities. Yeah, the idea of being that you would be able to build a computer so sophisticated that it could design the next generation of computers much better than any human could, and that next generation of computers could design the next generation of computers better than the predecessor could, and so on and so forth, and then creating a level way beyond anything that we can

even understand. Right, and faster too, so that you know, the time between generations gets to a point where there's no longer you can no longer say this is generation one, this is generation too. They'll just be meaningless to ask the question because it's constantly changing. Yeah, That rapid optimization is often called the intelligence explosion. Yes, uh. And so next we have the idea about computer networks themselves gaining

some form of sentience or intelligence. Now, this is one that I've heard a few different philosophers and computer scientists talk about the kind of idea that with any uh, network of information exchanging nodes that's complex enough, you could have consciousness arise. So, in other words, if you had enough information exchanging nodes in a complex enough system, then it would end up mirroring what our brains do with

our neural pathways. And because uh, in the argument of these people, consciousness is itself a manifestation of these physical connections that our brain emergent manifestation, Yes, that that would in fact emerge in these synthetic networks. One of the things here would be to consider whether or not a system like this could have executive control of some kind.

Like if you imagine the Internet today as a brain, well that is a brain or a mind, whatever you want to call it, that is both has more intellectual capability and has more knowledge than any other thing that exists, but it has no executive at the top of it. There's nobody controlling the Internet. To say, now, think about this, well, if you consider that the Internet is a crazy cat lady,

it makes perfect sense. It's the question of whether a flock of birds really knows where it's going, or just if each individual bird is uh doing what it does. Are a school of fish, this kind of thing where you see these behaviors that are are fascinating, uh, and on the surface they start to really perplex people. It's what you have to really dig down to understand how they work. Well, next, we've got the idea of computers

and humans getting all buddy buddy. Okay, so this is this is also ties into the stock art, right, it's like a human brain with like a computer chip on it and electricity shooting now a giant wind up key sticking out of the humans back come from a different school than you guys, So uh yeah, this is the idea that we have computer and human intelligence merging into

some form of new entity. So a lot of not all, but a lot of the singularity models that will talk about include this idea of trans humanism, where humans evolve beyond what we are now. We become a new species, either because we've merged with machines in this way where where we've incorporated machine intelligence into our human intelligence in a in a very organic way. I guess it's kind of weird to say organic, but that's kind of what

they're talking about. Or we create something that's uh like a superhuman intelligent machine that then changes us so that we're no longer human in the way we would think of now. And in that way, it's easy to see how some of these different pathways could overlap. Like first, you design say just a classical AI, like a super intelligent machine that's a computer, but that machine that's a computer tells you exactly how to modify your body with cybernetics that allow you to go in this hybrid state

to the singularity exactly. And this this is also the option I think where a lot of nanobots tend to come in our good old buddies nano bots. Yeah, the idea of like, uh, sort of fortifying your body. I'll make sure you have your vitamins, your minerals in your nanobots because the nanobots will continuously upgrade you. That isn't one of those popular ideas in science fiction. It's also one that you know, we've talked a little bit about nanotechnology in the past. We'll talk a lot more about

it in future episodes, but it's complicated stuff. Question, all these involved computers. Is there a way we could achieve super intelligence without computer? This is the Frankenstein method. So yes, through through pure application of medicine and science, we can create a superhuman. This is starting to sound a little scary in a way, but yeah, we're telling bioengineering as opposed to the rest of them. It's sky nets was

totally fine with you. Look, this one is scary. I'd like to say that I, for one, welcome our robot overlords, and I'm okay with them, all right. I have always treated my room bow with respect. This is just as a recorded account of my philosophy. M That's all I'm gonna say now. So, the bioengineering approach is the idea that we find out ways to maybe genetically modify humans.

It involves changing us on a biological level, where we attain this sort of superhuman intelligence through that pathway, as opposed through some form of machine intelligence. Okay, so before this was a cliche on the internet, what was it? Where did this idea come from? Well, you know, it's it's an old idea like we've had. We've had people talking about machines giving us the capability to to think

better for more than a century. So if you go all the way back to the Industrial Revolution, there were people who were saying, as machines were becoming more important in human activities, they were saying, look, this is allowing us to extend our ability to do to complete tasks, and just imagine a time where those tasks aren't just physical but also mental. And so it was pretty early on that people were starting to have the very seeds

of this idea. It wasn't so formalized as the idea of a singularity that would come later, but it was already starting to kind of trickle into the minds of people at the time. The actual term singularity, as it relates to a point beyond which humanity as we know it will change um was coined a nineteen fifty eight by John von Neuman. Okay, oh, that's kind of interesting, like a singularity as in like a black hole, or as in like at the beginning of the universe, like

you can't point right right. That's that's where. Yeah, I'm just thankful they didn't call it event horizon that that shows that movies had enough references on the show. So but yeah, so singularity as a point that we are not able to see beyond simply because it's it's beyond what we are able to conceive of. You don't get into technological singularity until about thirty years later, and that's when our our good buddy Verne came up with this idea.

He wrote an essay for Omni magazine, and he predicted we would quote unquote soon create superhuman intelligences that would transform our world in ways we can't anticipate because again, we have human intelligence. So if you're talking superhuman by definition, we cannot conceive of that. It's it's beyond our ken. Okay, where does Ray come in? Oh, you're talking Kurtswhile great kurts Will. I mean, he's the big name in the singularity, at least as far as I can tell. He When

you say singularity, who do you think of? Kurtswild, Like to the point that a lot of people on some futurists on the Internet are kind of derisive and go like, you know, other people aside from Ray Kurtzwile are working in this field, Thank you very much. Churtswild published a very popular book about it in two thousand five. He wrote The Singularity is New, or rather he got it published,

and that book was extremely popular. It it really brought this idea of the singularity beyond a kind of niche audience. I mean, uh, my dad subscribed to Omni magazine, but I'm not sure that that many other people outside of the science fiction world really read that much of Omni. So suddenly getting this into more of a kind of a mainstream mindset made the whole concept explode. Yeah, this took this weird nerd concept and in and had TV

interviews and stuff about it. He's also so charismatic. Yeah, yeah, he's I mean, he travels the world giving talks about, you know, futurism as well as other topics. And of course he's had his own incredible career. I've talked about him in the past on another show. We do text stuff, and Kurt Swile has had a very long career in uh everything from voice recognition to computer AI and it's

pretty cool stuff. So when he talked, people listened. And his prediction, which as far as I know, he has not amended any time recently, was that we would have the singularity by twenty five. So, in other words, when the year gets here, we will reach a time and human history where it will no longer be something we recognize. Yeah. The unfortunate thing about predictions and dates in the future is you can't really know whether they're right until you

get there. And that's a long way off, isn't it. That's deep Joe, Yes, that is true. Both of those things you said are true. Uh. Yeah. He actually also said that we would have computers that are as intelligent as humans. As to me, that's that's a really tricky prediction, because, of course, you can argue that there are many ways that computers are much better at performing certain types of tasks than humans are. I mean, it's it's undeniable. Yeah,

Like sorting through an incredibly long list incredibly quickly. Computers tend to be much better than people are sequencing a genome. Yeah. There are other things that humans are great at that computers are not so great at, like natural language. Natural language is a great example, or just being able to see. Like if I show you a cup and I teach you what a cup is, and then I show you a totally different cup, that's not that's not the same design.

It's different colors, different shape, different size. You still know what a cup is? You still oh, that's a cup. Computers don't necessarily know that. You have to really start building in this software. It doesn't matter how powerful the computer is unless the software itself is able to also

give the computer that that processing ability. Right. This is actually one reason that scientists and engineers are trying to teach computers to work like brains in terms of the hardware itself, because you can take a normal computer and with good facial recognition software. Like the computer being taught how to recognize faces and what faces are. It can sort of, it can do a good job. It can look at your face and say okay, but that's after

you've taught it. What's a lot harder is to release a normal computer onto the Internet and say, learn what these images are. I'm not giving you any help, right, Associations which are something that that humans do very naturally, are not something it's you have to program that, and programming that is one of the more difficult tasks. Very even even if you build the structure, it takes a lot of time. This is why you know, we've talked about the semantic web and this idea of meta data

and tagging this data and building ontologies. I mean, these are very important so that a computer can understand how to navigate a world that we can navigate naturally. YEA.

So even though we have agreement that computers are good at some things and that humans are better at other things, the thought of the people who who propose this technological singularity is that computers are rapidly going to catch up to the areas where humans are better at at those tasks, and they're still going to be much better at the

other task that computers are already ideally suited to. Yeah, perhaps the the things that the computers are good at will allow them to rapidly become better at the things they're not good at. Sure, but not everyone agrees with

the idea of the singularity. No, that We've got actually a bunch of people on on both sides of the issue, right, right, Let's talk about some of these main figures in in sort of singularity thought or singularity studies and what they've had to say about this idea because it's actually somewhat controversial these days. So we've got Kurtswile we've already mentioned him, he's probably one of the most visible UH thinkers and

speakers about the singularity. Yeah, and we should make clear when we talk about these people that being sort of on the pro singularity side doesn't necessarily mean you think it's a good thing. Oh right. A lot of these people actually think that it will be very negative and

are looking at ways of mitigating that potential. Yeah, right, But we're talking about people who sort of think that it's gonna happen, and think that it's going to happen pretty soon, like within our lifetimes, or if not our lifetimes, the next generation's lifetime. Okay, what about Hans Hans more of It. Well, Hans more of Ek is a Carnegie Mellon Robotics Institute adjunct professor used to be anyway, and uh he says, of course that the mind is the

product of physical processes in the brain. So uh, you know this is this is again something that's argued among philosophers and uh neuro specialists as well. But his his premise is that if you have enough power, enough computing power, and enough sophistication with software, that you would be able to replicate what happens in our our brains and thus

give rise to a synthetic mind. Okay, this would sort of make sense if you're saying that you're starting position is that the brain doesn't have anything magic in it, then why should it be that difficult to replicate what it does or even do better? And on one level, a lot of people say, yeah, that makes sense, but but on another level, you may be trivializing exactly how

complicated the brain and neurological processes actually are. Well, that's exactly why it seems like there might be something magic in the brain. Right, imagine that there's not something magic in it. Well, it's so complex physically that it seems to lots of people like there is. Well, it's just like just like the idea of any technology sophistical enough

will appear as magic. Exactly, same sort of thing here, any science that's complex enough and really really and that's sort of a testament to the fact that the brain may be a lot harder to replicate than we're giving it credit for. Right, And it's and it's really big. I mean, don't don't let the relatively small sides of

our skulls fool you. There's a lot going on in there. Yeah, when you're looking at how many neurons the typical brain has, how many neural pathways, the fact that neural pathways will form and then start to kind of every time you think back on that memory, the same sort of pathway starts lighting up, and then sometimes the pathway changes a little bit, which means that our memories are not entirely reliable, something that a lot of people depend heavily upon, Like lawyers, Um, So,

I mean, it's it's true, it's true. So, but it's it's an incredibly complex system, and it works in a very different way than your classic computer system. Now, one thing I want to say in defensive this idea is that even if the brain is very complex, if we can simulate the brain, I can't really see any reason

why a machine shouldn't go way beyond the brain. Yes, the brain is limited by all kinds of external factors like you know, the size of the head when you need to give birth and things like that, and competing with other organs for developing resources. I think a lot of objections to this are not based on it's impossible. I think most people who object to the idea of of simulating the brain aren't saying this is something that's

never going to happen. They're saying this is something that if it happens, it's going to happen a really long time from now. Yeah, the more optimistic neuroscience people that I've seen talk about this have estimated at least a hundred years, like absolutely crazy, minimum of I don't even understand what's happening anymore, is a hundred years. Because if you even look at IBM S Watson, which is one of the great examples of artificial intelligence, that we can

talk about today. You know, that's the IBM computer that went on Jeopardy defeated the two former Jeopardy champions. Seem to be really, really smart. It was relying on thousands of processors running at an incre dorble speed in order for it to even compete at that level. And it was it was a fairly specific, uh kind of task. I mean, yeah, you had you had the wonderful idea of this is going to encompass all human knowledge because you never know what the the categories are going to be.

But it still was specifically to compete at Jeopardy, not to hold a conversation and try and pass the Turing test for example. Um, if for us to have a fully functioning mind, not just a brain, but a mind, and looks like it's going to have to be even more complicated than what Watson was. And it's not just the hardware that we have to worry about again, it's

the software. If the software isn't sophisticate enough in uh, in the job of simulating the pathways that are in the brain, you can't really hope to simulate an organic mind. You might be able to come across and create some other form of mind that we have never encountered. I mean, that's still a possibility, and it's it's fascinating and scary

all at the same time. But it's not going to be the same as an organic mind unless you're able to really create an insanely sophisticated piece of software running on an equally insane sophisticated piece of hardware. So anyway, that's my own personal objection. So I think you already see where my mind goes. But next we have a

we have Verned again. We have our science fiction author and mathematician um and he said that he would be surprised if the Singularity doesn't show its head around twenty thirty. So he's even more aggressive in his timeline than Kurtswile is. That is sixteen years away. It's hard to imagine it was only a couple of years ago that someone talked to him and asked him again, like do you want to revise that? And he says, I stand by it. I guess as well. At a certain yeah, you know,

you're like, either it happens or it doesn't. I mean, so it's I guess it's at some point you just say, well, well, we'll just wait and see. Although I do want to say that his concepts of the singularity were largely inspired by a mathematician by the name of I Jake good Um, who is writing a lot about the intelligence explosion back in the nineteen sixties. So so you know, Verner. Verner

might be a little bit on the fantastical side. The arguments say aggressive, but as singularity is concerned, but his but his concepts are certainly based on mathematical concepts. Right though. He's great fun to listen to. There's an interview that Wired did with him a few years back. If you go and read that, he's just, uh, he seems to have this great fluidity talking about issues like this. A

published science fiction writer. I mean, all the people I know who are published science fiction writers are the most amazing people in the world. High dad. Uh. Then we've got Alizar yet Kalski. Yeah, he's um someone that was is. I always found it fascinating because of his thought experiment, which we'll talk about in another episode. Maybe we've already

talked about it. It's our friendly AI episode. Right. He's known in this field of of AI and sort of cautioning in the future of AI and discussing how to

make it safe. Yeah. Yeah, he's particularly concerned that the singularity, I mean A will occur, but but B will be a species extinction level event and uh, and so he's yeah, working towards creating strategies for a sustainable singularity, right, which again, very hard to imagine if you're talking about in a point where you can't really predict what happens next, I mean, just alone, how do you prepare for something that's unpredictable.

Also interested in creating the idea of friendly artificial intelligence is Nick Bostrom. Yeah. He also again, anyone who wants to prepare for a friendly kind of approach to this, you know, you kind of have to say this person is at least somewhat convinced that it's actually gonna happen. I mean, you wouldn't have a conversation about preparing for it. I mean, we're it's always better to prepare for something and not have it happened than to not prepare and

it happens anyway. Yeah, I'd agree with that last part. I could say that you could be very interested in preparing for safety in the case of the singularity without being especially concerned that it's very likely. I'm just not sure how many people would actually put forth the effort, you know, I just knowing just knowing humans the way I do. But well, anyway, he's given a prediction, right, Yeah,

he's a director of the Future of Humanity Institute. Boy, that's a heavy burden in it if you're the director for the Future of Humanity Institute, Like, so, what's the future of humanity? Look, I'm working on it, Okay, back off, all right, but he was quoted as saying that there's less than a fifty percent probability that will develop some sort of super intelligence by twenty thirty three. So this

sort of contradicts what Werner's prediction was. Assuming that the singularity also necessitates some sort of super intelligence, most of our definitions do involve super intelligence, intelligence explosion. That tends to be the way we think about the singularity. I guess you could argue that if you could find some other means to constantly have the entire world changing so quickly that's impossible to define it without using superintelligence, that

would also be a singularity. Yeah, and so there have been some people who have been sort of pro singularity in terms of thinking, it's very likely to happen, but at the same time very against it in terms of what should happen. Like Bill Joy, co founder of Sun Microsystems. He got famous in this line of thinking when he published an article in two thousand and Wired magazine called why the Future Doesn't Need Us, and it was very

pessimistic about the role of technology in the future of humanity. Yeah, he specifically was worried that we would develop technology that would be incredibly destructive and yet really easy to use. So imagine the equivalent of the button, right, the whole idea of that, the button that you press and it launches a nuclear strike against another country. Now imagine that essentially a sizeable percentage of the human population has access

to something that's it's equivalent in being dangerous. Does necessarily have to be nuclear, it could be whatever. Saying that, you know, as technology gets more advanced, it's giving more power to people, and with that, with great power becomes great responsibility, and so we have to be very cautious. And so he's he's advocated, let's not build those technologies. Let's back off and just build the stuff that's going to help people, and not the stuff that's gonna white

people out, which is easy to say exactly. I mean, a lot of the stuff that you build with the goal of helping people could eventually be used by accident to cause problems. I've built this death scorpion too health humanity and not destroy it or you know. I mean, if if you create an AI that wants to make people happy, um, but as it turns out, what makes a lot of people happy is exploding other people's skulls. Uh, got problems. That would be a bad example of technology

you want to push. Let's just say you create an AI that's designed to maximize the output of penicillin, but accidentally, what it does by maximizing the output of penicillin is it harvests people's bodies in order to turn them into the constituents it needs to make penicilla. So this is like Outbreak plus the matrix. I got it. Okay, Well, those are the folks, they're they're more obviously, there are a lot of people who fall on the side of

the Singularity is a thing. It's definitely gonna happen. It may be good, it may be bad, but it's coming. But they're not the only ones in this discussion. Oh yeah, there's a whole list of critics who say that either it's not possible at all or that it's very, very not near. Let's talk about the skeptics. So so, we had the co founder of Sun Microsystems in one corner. In the other corner, we got a co founder of Microsoft, Steve Balmer. No, not that Steve Bomber was not a

co founder. He was an early employee, but he was the in the sales department. I can talk all about Steve Bomber if you really want me to know what we're talking about. Paul Allen, Now I want to know what Steve Bomber. What about Bomber is a robot? Come on, we know this. Developers, developers, developers. He was on that loop for a long time. Took a while to get

him out of that. Okay, what about Paul Paul Alan his argument which he made with along with another computer scientists expert, He posted a an article that was kind of a rebuttal of of Kurtzwild. Yeah, and he said that, Yeah, Kurtswell ended up having a rebuttal of the rebuttal. But Kurtswall strikes me as someone who who rarely um rarely resists the urge to have the last word. But uh,

maybe we'll find out for sure podcast His Life. No, but Alan said that that if such a thing as the Singularity is going to happen, it's a long, long, long way off, And he made a lot of the arguments that I had kind of alluded to earlier. This idea that creating any kind of of machine intelligence that has intelligence in the way we think of intelligence, like on human level or beyond, is so far in the future as to be kind of a meaningless discussion at

the moment. Yeah, one thing that it brings up to think about the software limitations is that it's not just computers, Like you can make the most powerful computer that you could ever have, and if it didn't have some kind of guidance, if it didn't have instructions on how to create intelligence, I don't know how you would achieve intelligence. Could just run Tetris way better than the other computers

could run Tetris. Yeah, now that that's kind of Paul Allen's argument, And in fact, there's the goofy, snarky fake law saying Moore's law. You know you have right right, Moore's law says that technology will increase, it will double in power, right every two years, every two years or so, and pages Law in in contrast says that software will

bloat twice as much, twice as much every two years. So, in other words, your performance from one computer to flash forward two years in the future, the next computer ends up looking the same because the software ends up expanding to take advantage of all those up all those those capabilities of your computer has, but not in a way that would necessarily strike you as uh as as noticeable. Like we just think, well, this, this computer runs this

program just fine, but it's not software. It gets twice as slow for every unit of uh yeah, computer getting fasterware. So yeah, you preserve the stats quo. It doesn't you don't end up with this world of magical computers that

can suddenly do everything. Although you know that's clearly a very tongue in cheek approach, because if you do look around us the it's it's undeniable to say that the computers of today can run programs that are far more sophisticated than what like I mean, I think back to my very first computer, which was an Apple to E, and it would just fizzle and explode if I were to show it some of the stuff I can run

on my phone these days. But the point is kind of is kind of made where the software sophistication isn't increasing at the same uh space as More's law and I and I do have to say that sometimes web page is load not quite as slow as they did when I was on dial up, but but approaching that do to say ad software? Okay, So we had a sci fi author in the pro corner, do we have any sci fi authors in the anti corner? Charles Stross.

He believes that the development in a I will be focused on creating reactive technologies and environments that respond to our needs, rather than intelligent environments that anticipate things make decisions on its own. I mean, even these reactive environments could anticipate our needs, but they're doing so in a quote unquote dumb way, not a smart way, right. I think?

Is this sort of an argument against what's called a generalized artificial intelligence or artificial general intelligence, meaning simulating the whole way a human brain works, with the total freedom and range of thought, as opposed to a limited AI that's really good at being like a human brain at one thing. Yeah, sure, Sure, it's the difference between a video game character that can competently follow you around and shoot your enemies versus a video game character, you know,

taking that same grunt and having it cookie breakfast. It's right. Or to put it in in terms of like a smart home. You know, you might have a smart thermostat which can, based upon a certain algorithm, anticipate what temperature you're gonna want your house, but it's not going to be able to do everything else that you need to do. You have all these other little elements, and together they can create an entire environment that seems really reactive and intelligent,

but ultimately it's not really smart. It's just all following very specific sets of rules and not necessarily in a coordinated way. Just integrated stupidity. Yeah, which I think that that should be the new industry buzz term. I think that they should be calling, like nest to the integrated stupidity. Pretty sure. I pledged to a frat that had that as motto. At any rate, Uh, go dogs at any rate.

Next we have Ernest Davis, who's a computer scientist with New York University, and Davis's argument is that the human brain and intelligence is far more complex than computer scientists tend to give it credit, that it is something that is non trivial, that we have barely even scratched the surface. That even to have a conversation about intelligence usually requires you to focus in on a specific aspect of intelligence rather than just talking about general intelligence, because it's just

too broad a topic. Right, So this is sort of the other side of the coin that we already talked about with respect to Hans more Evc. Right, So maybe there's nothing magic in the brain. It's not we're not unable in principle to reproduce it. It's just so hard, yeah, right, right. And and the fact that some a lot of futurists have really good understanding of how computers work, but a really poor understanding of biology is is kind of an

issue here. One. I don't think it was this guy, but but I did read one neuroscientist who was talking about the genome. I mean, because you know, people have said, like, so we've uncoded the genome, so we know how to build a brain, so therefore we can build an intelligent and it's like, no, no, no, that's not what that does. If I gave that same person just a list of ingredients and say Lucky Charms cereal, and so you now have the ingredients needed to make Lucky Charm cereal, go

and make it for me. I doubt that they would come back with a bowl that looks anything like Lucky Charm cereal. It wouldn't even be ingredients. You'd be I guess giving them a list of atomic ratios. That's true, I could go. I was, I was oversimplifying, just to

show the the how absurd the argument is. But true, it's even more absurd than that you need to make some carbohydrates out of them, and and and even if you build, even if you build that wet where Again, like we said earlier, the emergence of intelligence from that wet where is a whole other question. Now, Joe, I know you spent some time looking over our next our next critic Lanier here who uh as, as you pointed out, looks like the villain and iron Man two. Yeah, he

does from certain angles. The computer scientist and and vr iluminary jarn Linear looks like Mickey Rourke in Iron Man two. This is true. It's a white guy with dreads. I mean, he's he's definitely got to look going. Uh he's and also really intelligent dude. Uh you know again vr illuminary like this is basically the guy who popularized the term virtual reality. He's had a lot to do in the industry down ever since. But he has some really interesting

things to say also about the singularity. He he calls the entire concept of it cybernetic totalism. He refers to it as a fanatic ideology and compares it to Marxism his His arguments here are that the idea of the singularity is flawed because brains and bodies and culture are actually pretty unlike computers. Um. Also because Darwin's theories of evolution don't apply perfectly to culture the way that a lot of futurists tend to speak of them as um.

And then also that whole thing with Moore's law that we were talking about before. He He furthermore goes on to say that the very concept of the singularity is dangerous.

This goes back to his comparison to Marxism um, because he says that that the concept of the singularity denies scientific skepticism, strokes the the the ego of the general computer scientists and um, furthermore encourages and kind of unrealistically rosy sense of predestination, which which all of which he thinks could lead to like possibly apocalyptic separation of the

haves and have not the technological and economic Yeah. Sure, I mean that that's one of those ideas about trans humanism, is that if this ever does come to pass, we can easily imagine that if such a thing or is possible, a very elite few are going to have access to

it initially. So assuming we've gotten past those testing stages where we've had the people stop exploding when we put computers in them and now they're fine, Um, then you could imagine that really it's just going to be the super duper rich people who are going to have access to this, which makes them even more further distanced from the rest of us. And uh, yeah, there there are a lot of science fiction stories that are predicated upon this.

In fact, there are several that movies that came out over the last like twenty four months that have the same premise. And I love his quote that arrogance is always a bad strategy and science and philosophy, I suppose it's fine kind of again, a little backhanded diss saying that you know this, this whole concept of the singularity is more in the world of philosophy than science. Okay, speaking of philosophy, we've talked about the philosophical question before.

Then this is slightly separated from the idea of the super intelligence in terms of behavior and capabilities. But is it possible that this machine or this set of machines or what where it is the super intelligent framework could achieve consciousness or the experience of mind and the way we think of it. So this is the idea of a machine knowing that it can think, knowing that it is what it is, and it is able to make uh decisions based upon this information. And we have John

Searle who's another critic of the singularity. This name might sound familiar to you guys. We talked about him about the in the Chinese room experiment discussion we had in a previous podcast. So this is the experiment where you say, imagine that you're sitting inside a room. Uh, there's a single door and it's got a little slot and it occasionally someone shoves a piece of paper through the slot and when you pick it up, it's got a Chinese character written on it, but you don't speak or read

or understand Chinese at all. What you do is you then you didn't consult an enormous book that gives instructions on what to do. When you get a sheet of paper that has any particular Chinese character on and you find the one that corresponds to the sheet of paper that was sent through the door, you end up scribbling something else down a sheet of paper. You push that back through the slot. Then the question is do you, as the person inside the room, actually understand Chinese? And

people would say, well, no, you don't understand it. You're following these instructions, but you don't have any comprehension of what it is you're doing apart from when this happens, do this other thing. And so his argument is that you have to extend the same thing to machines that they don't understand what they're doing. They're they're carrying out instructions and that's it. The idea is that instruction based computation can't by itself account for the idea of understanding

right now. Of course, we've also talked about criticisms of that particular thought experiment. The idea that if you take the system as a whole, as opposed to just the person inside the room, then would you say it understands Chinese. So, in other words, you're taking into consideration the book of instructions and the entire apparatus that's involved. Maybe then you would have a more questionable approach, like could you say

that that understands Chinese. Maybe not in the same way as a native speaker would, but maybe in a way that is semi analogous perhaps, uh, as opposed to the person in sight, because you could say, well, just think of the person is a microchip. That's not the full system, that's just one part of the system. Okay. Anybody else critical of this consciousness question? No, everyone else loves it. Well,

Roger Penrose doesn't. He's a physicist and professor of mathematics at Oxford, and he's thinking that we can't duplicate consciousness and machines because it depends upon non computational physical processes which we don't yet fully understand. So again, this is another approach of saying the human brain is so complex and we only barely know anything about it that we can't hope to simulate it or or even create another mind based upon that little bit of information we have.

We're just miles and miles away, so that if there ever does come a time where we can do this, it's going to require a new understanding of physics that we do not currently possess. And because it's going to require that and we don't, you know, you can't say like, oh, we've got a timeline in eight months, we're going to understand this. You can't say that. So he says, it's it's impossible to predict. It's probably not going to be

in the near future. Yeah. So I think something's interesting, which is that despite the fact that we've got lots of really really smart people on both sides of this debate, people say the singularity, it's both possible and near, and people who say possible, but not anytime soon, and then people who say not possible. Despite all that, you can see very clearly people are very quick to jump into one camp or the other on whether or not this

is a good thing. Have you noticed this that you just dive right in, Oh, yeah, utopia, you know, just this kind of Curtsewile kind of it's great vision. Probably even more common is the dystopian idea that oh yeah, okay, it means we will be destroyed. Yeah, I I mean my actual responses, there's no assuming that we really do in our world where we cannot predict what's going to happen next, then you can't predict whether it's going to

be good or bad exactly. That That's my point. What I was trying to point out there was it's it's such a hard question about whether or not it will happen and what form it will take. How could you possibly predict without knowing those things, the kinds of effects it will have. Now, I would say that it's very

smart to be cautious. Sure, oh sure, sure, Well, and it's sort of like the old joke of of you don't know, like that comma that you could get in from a terrible accident might be the happiest time of your life. Uh. And and so you should open up door number three because I got door number one is a goat behind it. Okay, So no one else is a big fan of let's make a deal? All right, Fine, that's fine. I know about the let's make a deal problem, yes, which we should probably do a podcast on at some

point because it's an interesting problem future of it. It's it's a it's an awesome it's an awesome thought experiment. But the uh, you know what I was going to say is that, yeah, you can't predict it. I I think one safe prediction you could make, assuming again, that would go to the The more fantastical version of the singularity is that human beings would ultimately cease to exist

as we understand them today. It doesn't mean that the species would be wiped out and that no, no, uh, other kind of remains of humans would ever know exist. It would just be machines from there on out. But they might be transformed in some fundamental way where we wouldn't really refer to them as human beings anymore. Human beings would be something we would use as a term to talk about those guys who lived before the singularity A right, that whole trans human space baby, super advanced

X men kind of concept. Yeah, yeah, well yeah. People take the idea of the singularity and then they attribute all different kinds of effects. You know, it'll mean immortality or near immortality. Oh, we'll be able to live forever.

And then the next big engineering problem will be how do we stop the heat death at the universe so that we don't have to die then, or there will be people who talk about how it means the end of work, say, it's going to lead to this this post scarcity world where there's absolutely nothing we need, right, this idea of going into sort of a Star Trek future where everything's idea like and no one has to no one has to work, they can pursue whatever whatever

they want to do. Not saying that it wouldn't mean those things. But I do think it's interesting that people are using this as as almost kind of a fantasy fulfillment. Oh sure, and I do I don't think that we said earlier. But but kurtswhile UM seems very personally motivated from a really intense fear of mortality, of its own mortality, UM, I think I think that's not hidden. He's fairly explicit about the fact that he wants to extend his life, Oh absolutely, and and that that comes from the death

of I think his father at a relatively young age. Yeah, he has UM. He has spoken many times on his hope that in within his lifetime obviously that we will have UH either conquered death or if not conquered it at least extended lie indefinitely. Uh, And that he thinks that that sort of breakthrough is like the Singularity right

around the corner. Now, obviously, if that breakthrough does happen, or if it comes part and parcel with the Singularity, then we will live to see this eventually come to pass. I mean, ultimately, I would say that if the Singularity is indeed a possibility, it will happen. So if it is possible, it will happen, because there's going to be someone who's going to do that next step that pushes it into that world. It's only if it's not possible

that we won't see it happen. I don't think there's gonna be a future we might not humans, humans not not we as in the three people in this room. I'm not gonna see it happen because I'll be washing my hair that day. I keep it in in a box. It's in my desk at home, take it out occasionally, try and wiggle out of it. I know exactly what you had in mind. You're like Kurts of all. You think that the Singularity is going to give you hair,

you know, a long, beautiful hair, shining streaming. Yeah, although actually that that brings up a good point. I mean, it's not like any of us are not personally motivated to see this kind of thing happen. It's I mean, it's a beautiful idea, and I totally understand that the seduction of that idea. Sure. Yeah, now, it's just when you start looking at you know, it is amazing the stuff we can do today compared to the stuff that

we could do twenty years ago. But I don't know that it's so amazing that it has us on track for this incredibly aggressive timeline that people like Werner and Kurtswile have in mind of. You know, by twenty we're going to have this uh, this this world that we

can't even begin to describe right now. Certainly, although if those kind of aggressive timelines are leading to people putting real thought into it and and real caution into it, then I think that it's completely worthwhile to have these kind of blowhard standing at the front of the stage. Of course, the caution is what we're want you to

talk about in our next podcast. Yeah, yeah, nice lead in. Yeah, So guys, stay tuned because we're going to talk all about superhuman intelligent artificial intelligence and how it will try to kill you. That will be our next episode. And if you guys have suggestions for future episodes of Forward Thinking, you should get in touch with us. Let us know what you think, tell us what you love about the show,

tell us what excites you about the future. You can email us our addresses f W Thinking at Discovery dot com, or drop us a line on Google Plus, Twitter or Facebook. Our handle is f W Thinking and we will talk to you again really soon. For more on this topic in the future of technology, visit forward thinking dot com. Brought to you by Toyota. Let's Go Places,