WEBVTT - A Critical Examination of The Singularity 0:00:04.120 --> 0:00:07.160 Get in touch with technology with tech Stuff from how 0:00:07.200 --> 0:00:13.960 stuff works dot com. Hey there, and welcome to tech Stuff. 0:00:14.000 --> 0:00:17.560 I'm your host, Jonathan Strickland. I'm an executive producer with 0:00:17.600 --> 0:00:20.280 How Stuff Works, and I heart radio and I love 0:00:20.480 --> 0:00:24.320 all things tech. And in my last episode, I advocated 0:00:24.480 --> 0:00:28.320 a critical thinking approach to the subject of driverless cars. 0:00:28.760 --> 0:00:31.360 Today we're gonna do sort of the same thing, but 0:00:31.440 --> 0:00:36.120 with a different subject. This time it's the singularity, specifically 0:00:36.159 --> 0:00:41.600 the technological singularity. I've done an episode about the singularity before, 0:00:41.640 --> 0:00:43.960 but I think the last time I really talked about 0:00:44.000 --> 0:00:47.960 it was in and I think attitudes have changed a 0:00:48.000 --> 0:00:50.720 bit in the five years since I did that episode. 0:00:50.760 --> 0:00:54.760 So I have argued and will continue to do so, 0:00:55.280 --> 0:00:59.640 that the technological singularity is still a very vague term. 0:01:00.000 --> 0:01:03.240 It means different things to different people. It can even 0:01:03.280 --> 0:01:06.800 mean different things to the same person depending upon the situation. 0:01:07.600 --> 0:01:12.080 In fact, if you'll allow me a little allusion to literature, 0:01:12.160 --> 0:01:16.640 since I do have a background in scholarship and English 0:01:16.680 --> 0:01:19.760 lit that I rarely get to use. This actually reminds 0:01:19.800 --> 0:01:23.120 me of Humpty Dumpty and Alice through the Looking Glass 0:01:23.520 --> 0:01:27.520 Humpty Dumpty in that talks about how words only mean 0:01:27.640 --> 0:01:31.720 whatever he wants them to mean, and all it takes 0:01:31.880 --> 0:01:34.840 is the will to use a word however you want, 0:01:35.120 --> 0:01:38.360 and the word has to obey. That seems to be 0:01:38.959 --> 0:01:43.720 the general feeling around the technological singularity. But if we're 0:01:43.760 --> 0:01:46.960 to explore this concept, we're going to have to define 0:01:47.000 --> 0:01:52.760 it at least a little bit. So here goes my attempt. First, 0:01:52.800 --> 0:01:57.240 let's go with the singularity in cosmology, right, what does 0:01:57.280 --> 0:02:00.120 it mean in outer space? What is it in physics? 0:02:00.600 --> 0:02:03.240 In physics? In outer space, it refers to the one 0:02:03.480 --> 0:02:06.680 dimensional point that is at the very center of a 0:02:06.720 --> 0:02:11.359 black hole. This is the point where density and gravity 0:02:11.400 --> 0:02:16.760 are infinite. It's the point where spacetime curves infinitely, and 0:02:16.960 --> 0:02:21.679 most importantly for the concept of the technological singularity, this 0:02:21.760 --> 0:02:25.560 is the point where the laws of physics no longer apply. 0:02:25.840 --> 0:02:31.919 They break down. So the technological singularity is ultimately a 0:02:31.960 --> 0:02:40.600 metaphor based off this cosmological notion. So with the technological singularity, 0:02:40.720 --> 0:02:44.280 the idea is that we arrive at a moment where 0:02:45.160 --> 0:02:49.919 technology is able to iterate and evolve at a rate 0:02:50.040 --> 0:02:53.680 so fast that it is impossible for us to envision 0:02:53.840 --> 0:02:58.720 or predict what happens next. Or it's about superhuman intelligence. 0:02:59.280 --> 0:03:01.760 Those tend to be the two big ones. So generally 0:03:01.800 --> 0:03:06.640 we imagine that superhuman intelligence actually brings about this era 0:03:07.040 --> 0:03:11.920 of rapid technological iteration and evolution, and the nature of 0:03:11.960 --> 0:03:14.960 that intelligence is still a matter of conjecture, because as 0:03:15.040 --> 0:03:19.120 far as we can tell, it hasn't happened yet. Some 0:03:19.240 --> 0:03:23.880 days we struggle for normal human intelligence. But there are 0:03:23.960 --> 0:03:27.440 some big, broad categories that we can talk about as 0:03:27.440 --> 0:03:32.760 a possibility for this superhuman intelligence scenario. There's the artificial 0:03:32.840 --> 0:03:37.360 intelligence version, in which computer scientists design a system of 0:03:37.440 --> 0:03:41.840 such sophistication that it operates on a human like level 0:03:41.920 --> 0:03:47.400 of intelligence. This AI, in turn begins to design its successor, 0:03:47.800 --> 0:03:51.400 which is an attempt to improve upon the previous generation. 0:03:52.600 --> 0:03:57.720 This next AI ends up being smarter, faster, stronger than 0:03:57.760 --> 0:04:01.360 its creator was, and it goes on to create the 0:04:01.400 --> 0:04:06.640 next generation of artificial intelligence, which by definition is going 0:04:06.680 --> 0:04:10.400 to be better designed than anything we humans could accomplish, 0:04:10.440 --> 0:04:15.080 because generation two is already better than human level intelligence, 0:04:15.480 --> 0:04:18.560 and bam, at that point you have the intelligence explosion. 0:04:18.680 --> 0:04:23.720 You have a superhuman intelligence beyond any capability of human 0:04:24.279 --> 0:04:28.040 thought here and because it's operating at a level beyond 0:04:28.160 --> 0:04:32.159 human capability. We cannot possibly guess at what will come 0:04:32.320 --> 0:04:36.800 next because we are only human. We can't conceive of 0:04:36.839 --> 0:04:40.400 what will happen with superhuman intelligence because we don't possess that. 0:04:41.200 --> 0:04:44.400 The vision of the future is sometimes but not always, 0:04:44.720 --> 0:04:47.840 accompanied by the notion that these machines are going to 0:04:47.920 --> 0:04:51.680 gain sentience, that they will have some form of consciousness, 0:04:52.360 --> 0:04:57.240 and therefore they are consciously making these improvements so they'll 0:04:57.279 --> 0:05:01.480 have some sort of sense of self. That also suggests 0:05:01.480 --> 0:05:05.839 that there's the possibility that they might develop motivations and 0:05:06.000 --> 0:05:10.279 wants and needs. So this leads us to some scenarios 0:05:10.279 --> 0:05:15.280 where this all develops into a terminator or matrix like dystopia, 0:05:15.560 --> 0:05:18.039 in which these machines have come to the conclusion that 0:05:18.080 --> 0:05:21.120 we human beings are a blight upon the earth and 0:05:21.160 --> 0:05:24.719 that we have to be wiped out. Or it could 0:05:24.800 --> 0:05:28.840 lead to machines subjugating humans in some way, that we 0:05:28.920 --> 0:05:33.960 are the ants to its colossus, and we can either 0:05:34.040 --> 0:05:36.560 be ignored or we can be put to work in 0:05:36.680 --> 0:05:40.560 some menial task. Or maybe the machines would just be 0:05:40.640 --> 0:05:44.880 super nice, high tech guardian angels that are granting us 0:05:44.920 --> 0:05:48.120 wishes whatever we want it they make it possible because 0:05:48.120 --> 0:05:51.839 they're intelligent enough to do it. But the point is 0:05:52.240 --> 0:05:55.240 the machines are the ones making the decisions, and they 0:05:55.279 --> 0:05:58.480 would be doing so beyond what we could do ourselves, 0:05:58.480 --> 0:06:01.560 and therefore we have no way of predicting what would 0:06:01.600 --> 0:06:04.160 come next or what the future would be like, because 0:06:04.200 --> 0:06:09.000 it's by definition, outside of our own experience. Another version 0:06:09.400 --> 0:06:13.640 of this superhuman intelligence that will bring on the technological 0:06:13.680 --> 0:06:19.240 singularity comes about due to biological or technological boosting of 0:06:19.279 --> 0:06:23.560 our own intelligence. So instead of making computer super smart, 0:06:23.760 --> 0:06:28.760 we make ourselves super smart. Possibly with the use of computers, 0:06:29.200 --> 0:06:32.840 we've discovered a way to augment our intelligence beyond what 0:06:32.920 --> 0:06:36.480 was previously thought possible. Now, in a way, you could 0:06:36.520 --> 0:06:40.680 argue that the history of humanity has been one of 0:06:40.800 --> 0:06:46.640 using technology to augment our intelligence, though the ancient philosophers 0:06:46.680 --> 0:06:51.479 like Socrates would disagree. Socrates would say, don't write anything down, 0:06:52.080 --> 0:06:54.680 because if you write things down, you don't have to 0:06:54.760 --> 0:06:57.520 keep that information in your head, and if you let 0:06:57.560 --> 0:07:02.200 information out of your head, then you're less intelligent as 0:07:02.200 --> 0:07:04.720 a result, you didn't have to keep it of there 0:07:04.880 --> 0:07:08.320 in your noggain. Similar arguments were made in the wake 0:07:08.440 --> 0:07:12.280 of the Guttenberg printing press when that was invented, and 0:07:12.360 --> 0:07:16.960 more recently, much more recently, Nicholas Carr wrote about the 0:07:17.000 --> 0:07:21.400 same idea, only his target wasn't the book or the 0:07:21.440 --> 0:07:24.480 printed word, it was Google. He wrote, is Google making 0:07:24.560 --> 0:07:28.360 us stupid? So since we started writing things down, there 0:07:28.360 --> 0:07:30.880 have been people who have said this technology is going 0:07:30.920 --> 0:07:33.240 to make us all dumb, because you don't have to 0:07:33.280 --> 0:07:36.960 think if you have this stuff to rely upon. But 0:07:37.680 --> 0:07:43.160 the counter argument is that the technology can augment our intelligence. 0:07:43.360 --> 0:07:47.200 We can offload some of the cognition onto the tech 0:07:47.440 --> 0:07:50.360 or that we're carrying or that we're depending upon, and 0:07:50.360 --> 0:07:54.400 that frees up more of our capacity to reflect or 0:07:54.440 --> 0:07:58.840 to do free association of seemingly separate and distinct ideas 0:07:59.160 --> 0:08:01.520 to bring them together there and come up with innovative 0:08:01.560 --> 0:08:04.560 solutions to hard problems. So I think it all just 0:08:04.640 --> 0:08:07.480 depends upon your point of view and your approach to 0:08:07.600 --> 0:08:12.480 using technology. Anyway, in this vision of the singularity, human 0:08:12.520 --> 0:08:16.920 beings become endowed with superhuman intelligence, which might very well 0:08:16.960 --> 0:08:20.120 mean we would no longer really be what you and 0:08:20.160 --> 0:08:23.480 I would consider human at this point. This is the 0:08:23.560 --> 0:08:28.640 trans human vision of the future. So the end result 0:08:28.760 --> 0:08:32.080 would again, be that we're talking about reaching a point 0:08:32.160 --> 0:08:38.240 where we, as humans, flawed, limited in our intelligence, are 0:08:38.320 --> 0:08:42.840 incapable of imagining what would come next because we cannot 0:08:43.040 --> 0:08:47.080 put ourselves in the place of a superhuman intelligence. We 0:08:47.240 --> 0:08:53.400 lack that intelligence, whether it's machines or us, or some 0:08:53.480 --> 0:08:56.520 combination of the two. The basic idea of the technological 0:08:56.559 --> 0:09:01.360 singularity hinges upon the emergence of this superhuman intelligence in 0:09:01.480 --> 0:09:06.200 almost every case. So where did that idea even come from? Well, 0:09:06.240 --> 0:09:10.080 back in seven there was a publisher named the Reverend 0:09:10.240 --> 0:09:14.640 Richard Thornton who wrote a piece about a mechanical calculator 0:09:14.760 --> 0:09:18.360 that illustrates my earlier point about how people have talked 0:09:18.360 --> 0:09:22.040 about how technology can make us dumb or it can 0:09:22.120 --> 0:09:26.760 be a hindrance for thinking. He wrote that such a device, 0:09:27.080 --> 0:09:32.080 if placed in a school, would do quote incalculable injury 0:09:32.200 --> 0:09:35.360 end quote. Don't think he was necessarily intending to make 0:09:35.400 --> 0:09:38.360 a pun about a calculator, but he did go on 0:09:38.440 --> 0:09:42.160 to say that these machines might get better and better, 0:09:42.280 --> 0:09:46.560 and what might happen next? He said, maybe they would 0:09:46.559 --> 0:09:49.640 be able to calculate ways to improve their own design 0:09:49.800 --> 0:09:54.600 and quote grind out ideas beyond the Ken of mortal 0:09:54.920 --> 0:10:00.000 Mind end quote. So he's saying, these machines themselves might 0:10:00.040 --> 0:10:04.000 even come up with the method of making a better machine, 0:10:04.400 --> 0:10:07.600 which in turn might make even better ones, and eventually 0:10:07.679 --> 0:10:10.000 get to a point where they could produce information that 0:10:10.040 --> 0:10:12.880 we humans can't even dream of. This is back in 0:10:12.960 --> 0:10:18.280 eighty seven, before the electronic computer. Alan Turing, whom we 0:10:18.360 --> 0:10:22.280 all often associate with this idea of machine intelligence, wrote 0:10:22.280 --> 0:10:25.400 a paper in nineteen fifty one at which a hypothesized 0:10:25.400 --> 0:10:28.880 that once machines do level enter a level of like 0:10:29.040 --> 0:10:33.160 human like thinking, or at least appear to think like 0:10:33.280 --> 0:10:36.280 humans do, it wouldn't take long before they would outpace 0:10:36.400 --> 0:10:40.439 our own capabilities. And he cited the nineteenth century writer 0:10:40.559 --> 0:10:44.160 Samuel Butler in this Butler himself had said there is 0:10:44.240 --> 0:10:49.160 no security against the ultimate development of mechanical consciousness in 0:10:49.200 --> 0:10:53.920 the fact of machines possessing little consciousness now, essentially saying, 0:10:54.200 --> 0:10:56.720 just because machines are dumb right now doesn't mean we 0:10:56.760 --> 0:11:01.040 can discount the possibility that they will one day be intelligent. 0:11:01.559 --> 0:11:06.120 He was making an observation about how quickly machines were advancing, 0:11:06.600 --> 0:11:10.320 and he was comparing that against the very slow process 0:11:10.480 --> 0:11:14.080 of evolution in biology. He was saying, you know, it 0:11:14.160 --> 0:11:20.800 took millions of years for lower organisms to evolve two 0:11:21.400 --> 0:11:27.600 higher thinking organisms. But look how quickly technology is developing. 0:11:28.240 --> 0:11:30.640 It won't be long at all before those are thinking 0:11:30.720 --> 0:11:34.800 before machines can think. That seems like a somewhat faulty 0:11:34.960 --> 0:11:38.439 reasoning to me, but we'll move on. John von Neumann, 0:11:38.440 --> 0:11:42.160 who I talked about in episodes in Tech Stuff this year, 0:11:42.760 --> 0:11:46.439 reportedly talked about a singularity that would happen once technological 0:11:46.480 --> 0:11:50.079 process a progress, rather would occur so quickly and become 0:11:50.160 --> 0:11:53.200 so complicated as to become incomprehensible. So this is more 0:11:53.200 --> 0:11:57.880 in that technology is rapidly iterating and evolving, but not 0:11:58.000 --> 0:12:01.520 necessarily related to a superhuman intel aligence. Uh. He told 0:12:01.559 --> 0:12:05.200 his buddy Stanislav Ulam about this, and it was Ulam 0:12:05.240 --> 0:12:10.199 who reported the conversation, so it's more from a secondhand information. 0:12:10.920 --> 0:12:13.640 Then there's I. J. Good, who was one of tourings 0:12:13.720 --> 0:12:16.719 peers at Bletchley Park when they were working on cryptography 0:12:16.800 --> 0:12:21.000 during the World War, and he wrote about a future 0:12:21.080 --> 0:12:24.880 intelligence explosion. He described the scenario that I mentioned earlier 0:12:25.280 --> 0:12:29.360 of machines designing ever more capable successors and leaving human 0:12:29.559 --> 0:12:32.839 ingenuity in the dust in the process. His work would 0:12:32.840 --> 0:12:36.920 inspire a science fiction author named Verner Venge, who presented 0:12:36.920 --> 0:12:42.840 the earliest recorded version of the phrase technological singularity. He 0:12:42.840 --> 0:12:45.080 wrote it in a piece that was published in Omni 0:12:45.080 --> 0:12:50.480 magazine in nineteen eighty three. A decade later, Venge predicted, quote, 0:12:50.600 --> 0:12:54.720 within thirty years, we will have the technological means to 0:12:54.840 --> 0:13:00.480 create superhuman intelligence. Shortly after the human era will be ended. 0:13:01.000 --> 0:13:05.320 End quote sounds pretty ominous, but honestly, it could just 0:13:05.440 --> 0:13:12.040 mean that we humans would be evolving into something new. Now, 0:13:12.080 --> 0:13:14.680 that was a prediction he made in and he said 0:13:14.679 --> 0:13:18.200 it was thirty years in the future, which means we're 0:13:18.240 --> 0:13:21.560 coming up on that deadline awful soon. It would mean 0:13:21.600 --> 0:13:24.360 that we would be there and let's see a two 0:13:24.440 --> 0:13:29.080 thousand twenty three by my math. Venge also suggested four 0:13:29.120 --> 0:13:33.040 ways this singularity could emerge. Computers could reach a certain 0:13:33.080 --> 0:13:37.480 level of sophistication and become awake and you know, therefore 0:13:37.559 --> 0:13:42.559 conscious and they would be superhumanly intelligent. Or computer networks 0:13:42.600 --> 0:13:47.680 would become suitably complex enough for intelligence and consciousness to 0:13:47.880 --> 0:13:51.840 emerge from that complexity, so the network itself and the 0:13:51.880 --> 0:13:57.080 associated users would become kind of a collective superhuman intelligence. 0:13:57.760 --> 0:14:01.800 There could be a computer human inter face future where 0:14:01.840 --> 0:14:05.839 these interfaces are so sophisticated that we can augment our 0:14:05.880 --> 0:14:11.319 intelligence seamlessly with machines, and together we are superhumanly intelligent. 0:14:11.600 --> 0:14:14.000 And finally, he said, we might figure out a way 0:14:14.000 --> 0:14:18.280 to biologically boost our own intelligence. Now, I would argue 0:14:18.280 --> 0:14:21.840 that Venge is one of the two really famous champions 0:14:21.960 --> 0:14:25.800 of this concept of the technological singularity. The other one 0:14:25.880 --> 0:14:28.400 I will cover in just a moment, But first let's 0:14:28.440 --> 0:14:38.680 take a quick break to thank our sponsor. I think 0:14:38.720 --> 0:14:43.200 the person who we most readily associate with the concept 0:14:43.240 --> 0:14:46.760 of the technological singularity, he has written numerous books on it, 0:14:46.840 --> 0:14:50.200 and has given lots of talks on the subject, and 0:14:50.280 --> 0:14:55.120 has promoted the idea extensively, would be Ray Kurtzwild. Kurtswilds 0:14:55.160 --> 0:14:59.080 definition is a little different from Vinge's definition, and the 0:14:59.160 --> 0:15:02.600 singularity is near. He talks about the era of rapid 0:15:02.640 --> 0:15:06.040 technological development that will make defining the future or even 0:15:06.080 --> 0:15:10.200 the present at that point impossible. And I think this 0:15:10.240 --> 0:15:13.320 is closer to what Stennis law Oolam said John von 0:15:13.400 --> 0:15:17.360 Neumann had told him. Kurt Swell, I should point out 0:15:17.680 --> 0:15:22.280 has made numerous contributions to technological development. I don't want 0:15:22.360 --> 0:15:24.760 anyone to think that I think Kurt Swills a crackpot. 0:15:24.800 --> 0:15:28.960 I don't. He was a pioneer in computer pattern recognition 0:15:29.080 --> 0:15:32.520 and speech recognition. He did a lot of development and 0:15:32.560 --> 0:15:37.720 other areas, including electronic music. So kurts wells a really 0:15:37.760 --> 0:15:41.320 smart guy. His view of the singularity tends to include 0:15:41.760 --> 0:15:46.400 some pretty far out ideas, however, including the possibility that 0:15:46.880 --> 0:15:51.120 through this process we will discover some means of extending 0:15:51.200 --> 0:15:55.680 our lives indefinitely, including the possibility of porting our consciousness 0:15:56.280 --> 0:16:00.400 into technologies somehow, in other words, finding a way to 0:16:00.520 --> 0:16:05.960 take what makes us us and put that us in 0:16:06.080 --> 0:16:08.760 some sort of machine, and that would effectively allow us 0:16:08.800 --> 0:16:11.560 to live forever, though I don't know what sort of 0:16:11.680 --> 0:16:15.640 experience that would be, like, would you just be a 0:16:15.680 --> 0:16:19.800 consciousness inside a computer? And if so, would that be 0:16:19.840 --> 0:16:25.440 a satisfying existence? Also, there's no way to even guess 0:16:25.480 --> 0:16:30.440 how this would be achieved. Uh, I'm not entirely clear 0:16:31.440 --> 0:16:34.200 what he's basing these ideas upon in the first place. 0:16:34.240 --> 0:16:37.040 It seems to me, and this is just my own 0:16:37.080 --> 0:16:41.680 opinion that some of Kurt Swill's notions are based upon 0:16:41.720 --> 0:16:45.200 a general fear of death, and that the singularity is 0:16:45.280 --> 0:16:49.440 viewed as a sort of escape mechanism from death. It's 0:16:49.440 --> 0:16:53.960 almost wishful thinking, in other words, that maybe once we 0:16:54.040 --> 0:16:58.640 hit this point, will have conquered death itself and we 0:16:58.680 --> 0:17:02.800 will never have to worry about facing our own mortality. 0:17:03.480 --> 0:17:06.960 I still don't understand how this sporting of consciousness would work. 0:17:07.640 --> 0:17:10.440 Or maybe you would just make a copy of yourself, 0:17:10.920 --> 0:17:13.359 in which case you would have a digital version of you, 0:17:13.760 --> 0:17:17.040 and in the meat version of you, but the meat 0:17:17.119 --> 0:17:20.760 version of you was still gonna die, which means, yeah, 0:17:20.840 --> 0:17:23.399 there'll still be a version of you hanging around, but 0:17:23.440 --> 0:17:26.920 it's not the you that's experiencing this podcast right now, 0:17:27.520 --> 0:17:31.159 unless you're hearing this post singularity and you are the 0:17:31.240 --> 0:17:36.359 digital version. Where did you get this show? Okay, I'm sorry, 0:17:36.359 --> 0:17:41.240 I was losing my mind there. Again. While I think 0:17:41.800 --> 0:17:46.919 that his ideas are far fetched from a personal standpoint, 0:17:48.200 --> 0:17:50.840 there is no doubt he is a billion times smarter 0:17:50.920 --> 0:17:55.080 than I am, so I acknowledge I could very well 0:17:55.119 --> 0:17:58.520 be in the wrong here. It just my my critical 0:17:58.560 --> 0:18:03.199 thinking and skeptical nature are really starting to raise some 0:18:03.240 --> 0:18:06.080 red flags. Now. There are a lot of other people 0:18:06.119 --> 0:18:08.680 who are also important in this field. I don't mean 0:18:08.680 --> 0:18:11.120 to suggest that I have covered them all. There are 0:18:11.240 --> 0:18:15.800 proponents and there are skeptics. For example, there's Hans more Effect, 0:18:16.080 --> 0:18:20.719 who's a computer scientist and a roboticist famously working with 0:18:20.760 --> 0:18:26.080 Carnegie Mellon University. He generalized Moore's law to a broader 0:18:26.119 --> 0:18:29.920 application of artificial intelligence and the arrival of a super intelligence. 0:18:30.160 --> 0:18:33.280 And one of his predictions was that by twenty forty 0:18:33.560 --> 0:18:37.359 or so, robots will become so sophisticated at that point 0:18:37.640 --> 0:18:41.560 and so complex that they will effectively emerge as their 0:18:41.600 --> 0:18:47.200 own species. In he penned the paper titled the Age 0:18:47.200 --> 0:18:51.480 of Robots, and this is a quote from that piece. 0:18:52.280 --> 0:18:57.639 Computer less industrial machinery exhibits the behavioral flexibility of single 0:18:57.760 --> 0:19:02.840 celled organisms. Today's best computer controlled robots are like these 0:19:02.880 --> 0:19:07.960 simpler invertebrates. A thousandfold increase in computer power in this 0:19:08.080 --> 0:19:13.600 decade should make possible machines with reptile like sensory and 0:19:13.720 --> 0:19:18.399 motor competence. Properly configured, such robots could do in the 0:19:18.440 --> 0:19:22.399 physical world what personal computers now do in the world 0:19:22.560 --> 0:19:27.240 of data. Act on our behalf as literal minded slaves 0:19:27.880 --> 0:19:31.040 growing computer power over the next half century will allow 0:19:31.119 --> 0:19:36.680 this reptile stage will be surpassed in stages, producing robots 0:19:36.680 --> 0:19:40.800 that learned like mammals, modeled their world like primates, and 0:19:40.880 --> 0:19:44.840 eventually reason like humans. Depending on your point of view, 0:19:44.960 --> 0:19:50.320 humanity will then have produced a worthy successor, or transcended 0:19:50.520 --> 0:19:56.200 inherited limitations and transformed itself into something quite new. No 0:19:56.240 --> 0:19:59.520 longer limited by the slow pace of human learning and 0:19:59.600 --> 0:20:04.720 even slower biological evolution, intelligent machinery will conduct its affairs 0:20:04.800 --> 0:20:09.120 on an ever faster, ever smaller scale, until coarse physical 0:20:09.240 --> 0:20:14.840 nature has been converted to fine grained, purposeful thought. Now, 0:20:16.320 --> 0:20:20.760 his ideas are predicated upon the assertion that consciousness, which 0:20:20.800 --> 0:20:24.320 is a quality that's devilishly difficult to define. In fact, 0:20:24.320 --> 0:20:26.960 I would argue it's just as difficult to define as 0:20:27.080 --> 0:20:32.720 the term intelligence. He argues, this arises from the material 0:20:33.080 --> 0:20:36.840 that is, the mind is totally the product of our 0:20:36.880 --> 0:20:39.359 nervous system, or, if you want to be a little 0:20:39.400 --> 0:20:42.959 more generous, the combination of our nervous system and our 0:20:43.000 --> 0:20:48.240 interactions with our environments. So, in other words, consciousness emerges 0:20:48.280 --> 0:20:52.840 from a system if that system meets the physical criteria. 0:20:53.320 --> 0:20:57.000 If true and I happen to believe that this is true. 0:20:57.800 --> 0:21:00.840 That it then stands to reason that if you have 0:21:01.040 --> 0:21:07.359 a sufficiently complex system with powerful enough machines, we should 0:21:07.359 --> 0:21:13.320 be able to create an artificial entity that possesses consciousness. If, however, 0:21:13.920 --> 0:21:21.080 consciousness arises from some other scientifically undiscovered or even undiscoverable quality, 0:21:21.200 --> 0:21:24.199 then it wouldn't matter how complicated we build our toys, 0:21:24.480 --> 0:21:28.040 they would never become conscious. So, in other words, if 0:21:28.080 --> 0:21:33.640 consciousness were the emergence from some other thing that science 0:21:33.800 --> 0:21:41.000 cannot address, like a soul, for example, then there's no 0:21:41.080 --> 0:21:45.199 way that we could create a conscious artificial being. We 0:21:45.320 --> 0:21:49.760 can't create the soul. If that is in fact how 0:21:49.800 --> 0:21:55.000 it works. I personally feel that that's not the case. 0:21:55.080 --> 0:21:58.800 That our consciousness does arise from the material that it 0:21:58.880 --> 0:22:03.080 does come from our nervous system, the complexity and the 0:22:03.119 --> 0:22:08.000 electrochemical processes of our nervous system. The question I have 0:22:08.280 --> 0:22:10.520 is whether or not we will ever be able to 0:22:10.600 --> 0:22:15.840 replicate that in an artificial system. Not saying that it 0:22:15.880 --> 0:22:19.400 would be impossible, just wondering if we will ever figure 0:22:19.440 --> 0:22:23.560 it out. It remains an open question. Nick Bostrom, who 0:22:23.640 --> 0:22:27.080 served as the director of the Future of Humanity, Institute 0:22:27.480 --> 0:22:31.480 has written extensively about trans humanism. I talked about that 0:22:31.560 --> 0:22:37.840 second ago. That idea that we transcend being just humans 0:22:37.880 --> 0:22:42.240 through some process. Whether that means a computer augmented person 0:22:42.359 --> 0:22:46.760 or a biologically augmented person isn't really important, at least 0:22:47.000 --> 0:22:50.320 from this perspective. It's very important from an ethical perspective. 0:22:51.200 --> 0:22:55.400 But he's using trans human to indicate that this describes 0:22:55.520 --> 0:22:58.480 someone who has moved away from what we would define 0:22:58.520 --> 0:23:01.600 as being a human being today. And like kurtswhile he 0:23:01.640 --> 0:23:04.840 has hypothesized that the Singularity will bring along with it 0:23:05.240 --> 0:23:10.159 some means of extending our lifespans indefinitely, but he feels 0:23:10.160 --> 0:23:12.840 that some of the more aggressive predictions are a little 0:23:12.840 --> 0:23:15.400 too optimistic. He has said that he felt that there 0:23:15.440 --> 0:23:18.119 was a less than fifty chance that we will have 0:23:18.200 --> 0:23:21.600 developed any sort of superhuman intelligence by the year twenty three. 0:23:22.200 --> 0:23:25.320 He thinks it's going to happen, but it might take 0:23:25.520 --> 0:23:28.720 a bit longer than that. Some of the people who 0:23:28.720 --> 0:23:33.320 believe or have formerly believed the Singularity to be around 0:23:33.359 --> 0:23:37.320 the corner aren't convinced it's necessarily going to be good 0:23:37.320 --> 0:23:42.320 for us. Venture capitalist Bill Joy, who co founded Sun Microsystems, 0:23:42.600 --> 0:23:46.840 has expressed concerns about it, and it wouldn't necessarily take 0:23:46.880 --> 0:23:52.000 a superhuman AI to do damage to us. Joy has 0:23:52.000 --> 0:23:55.480 pointed out that technology tends to advance our capabilities in 0:23:55.520 --> 0:23:59.400 all sorts of areas, including destructive ones. In other words, 0:23:59.400 --> 0:24:02.520 it gets easy you're and easier for smaller and smaller 0:24:02.560 --> 0:24:07.240 groups to do more and more harm. And that's not 0:24:07.280 --> 0:24:11.400 true of all technologies, obviously, but that technology does enable this. 0:24:12.560 --> 0:24:16.000 So as technology gets more sophisticated, the potential for a 0:24:16.040 --> 0:24:18.639 person or even a small group of people to exploit 0:24:18.680 --> 0:24:21.760 it in order to do a lot of damage also increases. 0:24:23.040 --> 0:24:26.320 We certainly see the potential for this in some technologies. 0:24:26.400 --> 0:24:29.280 For example, we've already seen three D printers that can 0:24:29.320 --> 0:24:33.960 be used to create untraceable firearms. But Joy's concern is 0:24:34.000 --> 0:24:39.240 that we could have a much more critical existential threat 0:24:39.600 --> 0:24:43.960 as a result of these technological developments, specifically in things 0:24:44.040 --> 0:24:48.280 like bleeding edge technologies like nanotech or biotechnology, and he 0:24:48.320 --> 0:24:50.920 feels that we need to build in the systems or 0:24:51.560 --> 0:24:56.960 create social pressures to prevent those things from happening. Similarly, 0:24:57.760 --> 0:25:03.119 Alizer Yakowski has advocated for the development of standards and 0:25:03.160 --> 0:25:08.160 algorithms to guide AI towards a benevolent mindset. Now I 0:25:08.160 --> 0:25:11.920 imagine he's an advocate for computer scientists to include ways 0:25:12.000 --> 0:25:15.440 for humans to see how machines are arriving at answers 0:25:15.560 --> 0:25:19.440 or calculations, which I agree with. I think that is 0:25:19.560 --> 0:25:24.119 very important. Transparency is incredibly important with these machines and systems. 0:25:24.640 --> 0:25:27.520 One of the dangers I see with machine learning is 0:25:27.560 --> 0:25:31.160 an approach that would block off this process from view. 0:25:31.200 --> 0:25:33.840 We wouldn't know how a machine got to the result 0:25:33.880 --> 0:25:36.280 it got, and we would just be taking it on faith. 0:25:36.960 --> 0:25:43.200 This would turn machines computers into black boxes. They produce results, 0:25:43.200 --> 0:25:46.400 but we don't know how or why. Such a thing 0:25:46.400 --> 0:25:48.919 would be more of an oracle than a computer, and 0:25:48.960 --> 0:25:52.240 that's not terribly healthy. I think it would be better 0:25:52.240 --> 0:25:54.840 if we build machines that are capable of showing their 0:25:54.880 --> 0:25:58.600 work as it were, to explain how they arrived at 0:25:58.600 --> 0:26:02.600 a particular conclusion and based upon the input that they received, 0:26:03.080 --> 0:26:06.200 and that way humans could verify that the decisions the 0:26:06.280 --> 0:26:11.840 computers were making were actually sound and appropriate and not mistakes. Next, 0:26:12.280 --> 0:26:15.000 I'll talk about some of the objections to the concept 0:26:15.200 --> 0:26:19.480 of the technological singularity itself. But first let's take another 0:26:19.560 --> 0:26:29.919 quick break to thank our sponsor. So what do people 0:26:30.119 --> 0:26:33.440 who think the technological singularity isn't going to be a thing, 0:26:33.680 --> 0:26:36.000 or at least not a thing that's going to happen 0:26:36.040 --> 0:26:40.160 anytime soon. What did they have to say? Well, one 0:26:40.240 --> 0:26:45.160 counter argument is that futurists and singularity enthusiasts are far 0:26:45.240 --> 0:26:49.760 too quick to apply the results of Moore's law across 0:26:49.800 --> 0:26:53.280 the board to all sciences and technologies that would be 0:26:53.400 --> 0:26:56.919 necessary in order to bring about some sort of superhuman intelligence. 0:26:56.960 --> 0:26:59.960 And this to me seems like a pretty solid count 0:27:00.040 --> 0:27:03.239 her argument. So, first of all, Moore's law is not 0:27:03.400 --> 0:27:08.600 really a law, right, It's an observation. It's an observation 0:27:08.640 --> 0:27:11.720 and a prediction that Gordon Moore made in the early 0:27:11.840 --> 0:27:17.880 days of silicon based transistors. Namely, more stated the due 0:27:17.880 --> 0:27:24.080 to economic demand for electronics, the semiconductor companies would invest 0:27:24.240 --> 0:27:28.640 the resources needed to create more complicated chips. And now, 0:27:28.680 --> 0:27:30.600 at the time he was making this prediction, he was 0:27:30.640 --> 0:27:36.200 specifically talking about how many discrete components you could fit 0:27:36.400 --> 0:27:40.880 on a square inch silicon wafer. Essentially, this was really 0:27:40.920 --> 0:27:44.240 about how many transistors can you cram into that space. 0:27:45.080 --> 0:27:50.119 Demand would create the need to innovate. The economic incentive 0:27:50.320 --> 0:27:54.440 to innovate, so companies would spend money in research and development, 0:27:54.520 --> 0:27:58.080 and as a result, they would shrink those components down 0:27:58.200 --> 0:28:00.040 so that you could fit more of them on the 0:28:00.080 --> 0:28:03.720 same amount of space and have more powerful electronics, which 0:28:03.720 --> 0:28:08.040 would continue to create more demand. He traced this pattern back, 0:28:08.600 --> 0:28:10.960