WEBVTT - Ep3 "Could animals learn to speak human?" 0:00:05.760 --> 0:00:12.040 What exactly is intelligence? And why don't crows invent an Internet? 0:00:12.119 --> 0:00:15.800 And why don't gophers right novels? And what does any 0:00:15.800 --> 0:00:18.680 of this have to do with George Orwell's analysis of 0:00:18.720 --> 0:00:23.279 Adolf Hitler? Or why World War five could involve many 0:00:23.360 --> 0:00:29.920 other species besides Homo sapiens. Welcome to Inner Cosmos with me, 0:00:30.160 --> 0:00:35.640 David Eagleman. I'm a neuroscientist and an author at Stanford University, 0:00:36.960 --> 0:00:41.080 and I've spent my whole career studying the intersection between 0:00:41.360 --> 0:00:48.559 how the brain works and how we experience life. On 0:00:48.640 --> 0:00:52.479 today's episode, we're going to explore the future not just 0:00:52.560 --> 0:00:57.000 for humans, but for other species as well. The other day, 0:00:57.080 --> 0:01:00.000 my kids were in the car while I was pumping gas, 0:01:00.040 --> 0:01:03.639 and I was thinking about the future for them, and 0:01:03.760 --> 0:01:07.200 specifically what my kids would look back on in twenty 0:01:07.280 --> 0:01:10.360 years or thirty years or fifty years and think, wow, 0:01:10.400 --> 0:01:15.080 I can't believe we did that. So California has new 0:01:15.160 --> 0:01:18.560 legislation that all new cars will have to be electric 0:01:18.640 --> 0:01:21.400 by twenty thirty five, and so I was realizing that 0:01:21.440 --> 0:01:24.760 when my kids are the age I am now, they'll 0:01:24.760 --> 0:01:28.160 look back in moments like this and think, wow, I 0:01:28.200 --> 0:01:31.680 remember when dad used to pump fossil fuel into his car, 0:01:32.280 --> 0:01:35.440 and they'll see this as a surprising feature of their 0:01:35.560 --> 0:01:39.600 childhoods and not something that exists for them anymore. It's 0:01:39.680 --> 0:01:42.559 analogous to the way that I look back on black 0:01:42.600 --> 0:01:46.360 and white televisions, or pay phones, or cassette tapes or 0:01:46.480 --> 0:01:50.160 walkman's or any of the other things from my childhood 0:01:50.520 --> 0:01:52.960 that we just don't have anymore. So when I got 0:01:53.040 --> 0:01:56.360 back in the car, I asked them to brainstorm about 0:01:56.440 --> 0:01:59.640 things that would be different for them when they're adults, 0:02:00.280 --> 0:02:02.760 and they brought up more things than I expected. We 0:02:02.760 --> 0:02:05.640 were on a road trip and we were passing a 0:02:05.760 --> 0:02:09.520 ranch where we saw thousands of cows grazing, and what 0:02:09.600 --> 0:02:12.519 we ended up talking about was whether they might look 0:02:12.560 --> 0:02:17.440 back strangely in thirty years on raising animals to kill 0:02:17.480 --> 0:02:20.440 them for meat, because we looked at those cows and 0:02:20.480 --> 0:02:23.600 we understood that they would all be dead soon and 0:02:23.760 --> 0:02:28.400 on plates around tables. Now that might change, and not 0:02:28.639 --> 0:02:33.679 necessarily because society drifts towards vegetarianism, and not even necessarily 0:02:33.720 --> 0:02:39.200 because companies develop better and better plant based alternatives to burgers. 0:02:39.200 --> 0:02:42.960 Our bodies, after all, have been clearly carved by evolution 0:02:43.000 --> 0:02:46.080 to eat meat, So I think the more likely scenario 0:02:46.400 --> 0:02:51.560 will be the move toward lab grown meat will pop 0:02:51.600 --> 0:02:54.880 a single cell off a cow and reproducing the lab 0:02:54.919 --> 0:02:59.080 dish to grow the burger or the steak directly from 0:02:59.120 --> 0:03:03.200 that cell, rather than raise the entire animal and feed 0:03:03.240 --> 0:03:06.280 it for years and then slice its throat and process 0:03:06.320 --> 0:03:09.480 it and ship the meat, so we could eat meat 0:03:10.040 --> 0:03:13.520 free of cruelty and with a lot less ranching effort. 0:03:13.919 --> 0:03:16.120 And as time goes on, we'll get better and better 0:03:16.120 --> 0:03:19.680 at growing the proper ratios of fat for marbling and 0:03:19.720 --> 0:03:22.200 blood vessels to carry in the right kinds of nutrition, 0:03:22.800 --> 0:03:27.000 and you'll be able to grow a filet mignon without 0:03:27.040 --> 0:03:30.440 actually having to have thousands of acres and raise and 0:03:30.480 --> 0:03:34.480 slaughter a living being that seems to care about it's 0:03:34.560 --> 0:03:38.680 young and tries to avoid pain and presumably has a 0:03:38.720 --> 0:03:42.440 whole range of emotions. So there are several companies working 0:03:42.480 --> 0:03:45.320 on this already, and with time we'll be able to 0:03:45.360 --> 0:03:48.280 do this at a price that competes with traditional meat, 0:03:48.720 --> 0:03:51.640 and eventually that's far cheaper because you won't have to 0:03:51.760 --> 0:03:53.960 raise the cows and feed them and tend to them 0:03:53.960 --> 0:03:57.880 and shelter them for years. You won't drive your gas 0:03:57.960 --> 0:04:00.840 powered vehicle along the highway and look at these thousands 0:04:00.880 --> 0:04:04.200 of cows waiting for their own murder. And once we 0:04:04.200 --> 0:04:08.600 can grow burgers from single cells, we can enjoy barbecues 0:04:08.640 --> 0:04:13.960 with lion burgers or tiger burgers or elephant burgers or whatever, 0:04:14.120 --> 0:04:18.680 because we're just reproducing cells. Anyhow, as we talked, we 0:04:18.720 --> 0:04:22.040 got into the future of genetic editing, and there are 0:04:22.080 --> 0:04:27.360 tools like crisper Cast nine. This is a molecular tool 0:04:27.440 --> 0:04:31.880 that allows scientists to splice out a well defined section 0:04:31.920 --> 0:04:35.000 of a genome, in other words, a particular sequence of 0:04:35.080 --> 0:04:38.160 a's and seas and teas and geese, and replace that 0:04:38.240 --> 0:04:41.760 sequence with a different sequence that they want, so maybe 0:04:41.760 --> 0:04:46.440 it's one that doesn't have a particular genetic mutation or 0:04:46.480 --> 0:04:50.360 that causes a disease or a disorder. Now, this is 0:04:50.360 --> 0:04:54.160 a technology, crisper Cast nine, that started to come into 0:04:54.240 --> 0:04:58.120 dim focus in the mid nineteen nineties, but using it 0:04:58.160 --> 0:05:02.359 to edit genomes in alien cells that wasn't demonstrated until 0:05:02.600 --> 0:05:06.240 twenty twelve, and suddenly the technology blew up and it's 0:05:06.240 --> 0:05:09.440 an every lab around the planet. Now. Now, the idea 0:05:09.520 --> 0:05:14.400 of using that kind of technology on the human germline 0:05:14.920 --> 0:05:18.920 has been an area of really heated ethical debate, and 0:05:19.120 --> 0:05:22.360 it appears to have been used on humans illegally by 0:05:22.440 --> 0:05:25.800 a doctor in China in twenty eighteen, and it's an 0:05:25.839 --> 0:05:30.279 open question where this all goes from here. All countries 0:05:30.279 --> 0:05:33.680 have said this is illegal for human use, but it's 0:05:33.760 --> 0:05:37.279 just a matter of time probably before more examples hit 0:05:37.320 --> 0:05:40.640 the news. And by the way, nowadays, instead of cutting 0:05:40.640 --> 0:05:44.400 and replacing whole chunks of DNA like Crisper does, there 0:05:44.440 --> 0:05:49.120 are methods to change just a single base pair. Molecular 0:05:49.120 --> 0:05:52.960 biologists from MIT and Harvard made a gene editing technique 0:05:52.960 --> 0:05:56.599 that rewrites individual as and seas and teas and g's 0:05:56.839 --> 0:06:00.159 in the genetic code. This is called base editing. But 0:06:00.240 --> 0:06:06.000 what difference would it make if we could edit genomes? Well, 0:06:06.040 --> 0:06:09.000 the thing that interests me today is whether we could 0:06:09.160 --> 0:06:13.600 leverage these advances in genetic engineering to someday allow us 0:06:13.640 --> 0:06:17.960 to boost human intelligence. Now, there are several things to 0:06:18.080 --> 0:06:22.120 unpack about that prospect. The first is that every few 0:06:22.200 --> 0:06:26.480 years in the neuroscience community there's buzz about the possibility 0:06:26.520 --> 0:06:30.680 that someone discovered a noatropic drug, and that just means 0:06:31.040 --> 0:06:34.479 one that's going to boost memory or recall or some 0:06:34.560 --> 0:06:39.239 other cognitive function. But despite the occasional buzz, there's really 0:06:39.560 --> 0:06:44.040 nothing that has changed our cognitive function beyond just drinking 0:06:44.040 --> 0:06:48.480 a cup of coffee. But instead of a drug, could 0:06:48.520 --> 0:06:54.040 we improve intelligence by tweaking the genome? Now, one possibility 0:06:54.160 --> 0:06:57.479 is that nature has already optimized our intelligence and there's 0:06:57.520 --> 0:07:00.760 not too much room to improve it. So let me 0:07:00.760 --> 0:07:04.640 explain that by an analogy. Take life span. There's been 0:07:04.640 --> 0:07:07.960 a lot of research on how to increase human lifespan, 0:07:08.480 --> 0:07:12.320 but the length of a human's lifespan actually has not 0:07:12.360 --> 0:07:15.960 really changed. When we look back to a time before 0:07:16.120 --> 0:07:20.360 modern medicine, we see that people in history we're capable 0:07:20.480 --> 0:07:25.000 of living long lives. Take Benjamin Franklin, who is born 0:07:25.400 --> 0:07:28.840 in seventeen oh six and he died in seventeen ninety 0:07:28.880 --> 0:07:31.840 at age eighty four. That's a very ripe old age, 0:07:31.960 --> 0:07:35.280 especially for a man. So it's not that people couldn't 0:07:35.280 --> 0:07:39.760 live that long. What has changed is the average lifespan, 0:07:40.120 --> 0:07:43.960 which is known as life expectancy. So when researchers talk 0:07:44.000 --> 0:07:49.280 about the massive improvements in human life expectancy, they're talking 0:07:49.280 --> 0:07:54.080 about the average human lifespan. So as we've developed medicine 0:07:54.120 --> 0:07:59.280 and antibiotics and methods for reducing child mortality and even 0:07:59.320 --> 0:08:02.360 reducing from simple things like getting gang green when you 0:08:02.400 --> 0:08:06.400 get a cut. That means that on average, people don't 0:08:06.480 --> 0:08:10.040 die in their childhood or young adult lives anymore. It 0:08:10.080 --> 0:08:12.240 turns out that some of the most important steps in 0:08:12.320 --> 0:08:17.000 reducing mortality where the basic medications introduced last century to 0:08:17.120 --> 0:08:21.880 control diarrhea and vomiting. These were massively important in reducing 0:08:21.960 --> 0:08:25.120 death tolls. So people make it down the road of 0:08:25.280 --> 0:08:29.400 possibility to live until they're old age. So what's been 0:08:29.440 --> 0:08:33.600 improved upon is the average age of death. The point 0:08:33.640 --> 0:08:36.160 I want to draw here is that it's not as 0:08:36.160 --> 0:08:39.319 though any of our researchers said, oh, I got it, 0:08:39.480 --> 0:08:41.960 here's how to make somebody live until they're two hundred 0:08:42.000 --> 0:08:45.280 years old. And this is because the human body, at 0:08:45.320 --> 0:08:48.480 least at this point doesn't appear to be set up 0:08:48.520 --> 0:08:52.719 to live more than about a century. So by analogy, 0:08:52.840 --> 0:08:56.440 it could be the case that human intelligence has been 0:08:56.480 --> 0:09:01.400 carefully shaped by evolutionary pressures and is essentially maxed out. 0:09:01.920 --> 0:09:05.040 In other words, if you tweaked the brain circuitry, you 0:09:05.160 --> 0:09:08.280 might not be able to squeeze much more juice out 0:09:08.320 --> 0:09:11.720 of the lemon. Nonetheless, even if this is the case, 0:09:12.120 --> 0:09:16.760 we could in theory improve the intelligence expectancy of the 0:09:16.800 --> 0:09:21.199 population by pushing everyone up towards the ceiling. So everyone 0:09:21.320 --> 0:09:24.760 is essentially operating at the level of the best brains 0:09:25.240 --> 0:09:29.360 like Albert Einstein or Tony Morrison, or all the people 0:09:29.400 --> 0:09:32.480 who are super talented and domains like math or music 0:09:32.559 --> 0:09:36.760 or language. Those people are massive outliers from the average. 0:09:36.800 --> 0:09:40.800 So that tells us the possibility that on average, most 0:09:40.800 --> 0:09:45.760 people could get a bit smarter. However we define intelligence now. 0:09:45.880 --> 0:09:48.160 I'll come back to genetic editing in just a moment, 0:09:48.200 --> 0:09:50.720 but first I need to lay another brick in the 0:09:50.800 --> 0:09:56.839 foundation by addressing the issue of what is intelligence? The 0:09:56.880 --> 0:10:00.839 main challenge we have with understanding it. We don't have 0:10:00.960 --> 0:10:05.360 a precise definition for it. The interest in this concept 0:10:05.440 --> 0:10:08.319 of intelligence goes back to at least the time of Plato, 0:10:08.679 --> 0:10:11.600 but just over a century ago people started working on 0:10:11.679 --> 0:10:15.880 how to quantify this, and by nineteen oh four, the 0:10:15.920 --> 0:10:20.800 English psychologist Charles Spearman found that people who scored well 0:10:20.960 --> 0:10:24.880 on one sort of task like verbal skills, also tended 0:10:24.920 --> 0:10:28.640 to perform well on other sorts of tasks like spatial skills, 0:10:28.920 --> 0:10:32.480 so that led Spearman to conclude that intelligence could be 0:10:32.640 --> 0:10:37.520 generalized across tasks, and other researchers then found that this 0:10:37.600 --> 0:10:41.440 seemed to be true. In other words, tasks of different 0:10:41.520 --> 0:10:46.520 abilities tended to have high positive correlations. So what that 0:10:46.600 --> 0:10:50.600 means is a person with a high general intelligence score 0:10:50.960 --> 0:10:55.520 tends to do better at language and memory and tackling 0:10:55.520 --> 0:10:58.880 new sorts of problems and real life activities and so on. 0:10:59.440 --> 0:11:02.280 And in nine twenty three a researcher wrote a famous 0:11:02.360 --> 0:11:07.240 paper in which he concluded tests tested by which he 0:11:07.280 --> 0:11:11.120 meant that whatever the heck intelligence is, there are ways 0:11:11.240 --> 0:11:13.960 to put a number to it. Now, I'll just mention 0:11:14.040 --> 0:11:17.080 this has been an area of debate where people point 0:11:17.120 --> 0:11:21.040 out that there can be cross cultural issues with testing intelligence. 0:11:21.080 --> 0:11:23.520 But that's not the point I'm making here. I'm talking 0:11:23.559 --> 0:11:26.000 about if you have a hundred people from the same 0:11:26.040 --> 0:11:28.480 culture in the same place and they all sit down 0:11:28.520 --> 0:11:32.080 for the test, you'll get a spectrum of scores, and 0:11:32.200 --> 0:11:36.920 these scores generally correlate with success in many domains, including 0:11:37.160 --> 0:11:39.800 how they do later in life. So with these early 0:11:39.880 --> 0:11:45.080 researchers noted is that something is getting captured. But what 0:11:45.400 --> 0:11:51.440 exactly is intelligence. Well, it's proven difficult to nail down 0:11:51.480 --> 0:11:56.240 with a single definition because it's not one thing, and 0:11:56.280 --> 0:11:59.640 it's certainly not defined by any single area in the brain. 0:12:00.480 --> 0:12:02.600 So what do we know about it? Well, first we 0:12:02.679 --> 0:12:07.719 know it's not defined by brain's size. At the end 0:12:07.720 --> 0:12:11.480 of the eighteen nineties, the Spanish neuroscientist Ramoni ca Hall 0:12:12.280 --> 0:12:15.800 was blown away by the fact that an elephant's brain, 0:12:15.880 --> 0:12:19.480 let's say, is so much larger than a mouse's brain, 0:12:19.800 --> 0:12:24.839 and yet they're both equally capable of strategies and clever 0:12:25.000 --> 0:12:28.720 moves and finding mates and so on. So a bigger 0:12:28.840 --> 0:12:33.120 brain is not necessarily any smarter. Or take somebody like 0:12:33.160 --> 0:12:37.200 Andre the Giant. He probably wasn't eight times smarter than you, 0:12:37.679 --> 0:12:39.960 even though his brain may have had eight times the 0:12:40.040 --> 0:12:44.360 volume of yours. So Ramoni ca Hall said, think about 0:12:44.360 --> 0:12:48.640 it like the giant clock tower of Big Ben compared 0:12:48.679 --> 0:12:52.479 to a wristwatch. They both keep time with the same accuracy. 0:12:53.320 --> 0:12:57.640 And so it goes with brains big and small. It's 0:12:57.640 --> 0:13:00.600 not the amount of stuff, but the al rhythms that 0:13:00.600 --> 0:13:03.680 are running on it. It may be that what we 0:13:03.800 --> 0:13:07.720 call intelligence is about how many things you could hold 0:13:07.720 --> 0:13:11.040 in your short term memory at one time, or maybe 0:13:11.400 --> 0:13:15.440 it's the idea that you can store stronger associations between facts, 0:13:15.920 --> 0:13:20.280 or maybe it's the ability to resolve cognitive conflict, or 0:13:20.320 --> 0:13:25.040 maybe it's about being able to better squelch distractors. So 0:13:25.160 --> 0:13:28.559 all these things have been suggested as the basis of intelligence, 0:13:29.080 --> 0:13:31.720 but it's going to involve all of these in some 0:13:31.760 --> 0:13:35.280 degree or another, and probably a bunch of other things, 0:13:35.720 --> 0:13:40.080 like whether you can better structure the information that you've stored, 0:13:40.720 --> 0:13:44.640 or do more parallel processing, or i'll talk about this 0:13:44.679 --> 0:13:51.120 in the future episode, whether you're better at simulating possible futures. Okay, 0:13:51.160 --> 0:13:54.839 So all this is to say that intelligence doesn't represent 0:13:55.240 --> 0:13:58.120 a simple single thing that we can point to in 0:13:58.160 --> 0:14:01.880 the brain. Instead, in eligence is one of those words 0:14:01.920 --> 0:14:05.720 that carries a lot of semantic weight, meaning there are 0:14:05.960 --> 0:14:08.920 all kinds of things that are being lumped together in 0:14:08.960 --> 0:14:14.240 that one word. But whatever intelligence is, it sits at 0:14:14.240 --> 0:14:18.640 the heart of what is special about Homo sapiens. Other 0:14:18.720 --> 0:14:24.080 species seem to come out more hardwired to solve particular problems, 0:14:24.520 --> 0:14:29.080 but we are more live wired, which means our brains adapt. 0:14:29.560 --> 0:14:32.960 So when we drop into the world, we can in 0:14:33.000 --> 0:14:36.760 our first several years learn all the most important things 0:14:36.800 --> 0:14:39.600 that our species has figured out before us, and then 0:14:39.640 --> 0:14:42.880 we can springboard from there. And that means we can 0:14:42.920 --> 0:14:47.040 solve open ended problems, and that can be surviving in 0:14:47.080 --> 0:14:50.960 the Arctic Circle or surviving at the equator, whether that's 0:14:51.000 --> 0:14:54.640 figuring out quantum physics or how to distill nitrogen from 0:14:54.680 --> 0:14:59.240 the air for fertilizer or whatever. Other species do smart things, 0:14:59.680 --> 0:15:03.920 but none can even come close to our capacity to 0:15:03.920 --> 0:15:09.040 tackle abstract, open ended problems. And all of this leads 0:15:09.040 --> 0:15:12.360 me to the main point of today's podcast, which is 0:15:12.400 --> 0:15:15.560 one of the most interesting possibilities for how the world 0:15:15.960 --> 0:15:19.880 could be different for our children. They've grown up with 0:15:20.000 --> 0:15:24.640 Homo sapiens being the only intelligent species that is doing 0:15:24.720 --> 0:15:27.760 anything remarkable on this planet. And that's not to say 0:15:27.760 --> 0:15:30.400 other animals aren't intelligent in their own ways, but they 0:15:30.440 --> 0:15:33.080 are not as far as we know, building an internet, 0:15:33.200 --> 0:15:38.720 or vaccines or quantum computation, or certainly not building satellites 0:15:38.720 --> 0:15:42.200 and rocketships and getting off the planet. For example, did 0:15:42.200 --> 0:15:44.680 you know that at any given moment in time, there 0:15:44.680 --> 0:15:49.440 are approximately one million Homo sapiens up in the atmosphere 0:15:49.440 --> 0:15:53.840 above the clouds, sitting in comfortable leather chairs. So we 0:15:53.960 --> 0:15:59.000 might argue about animal intelligence, but what's clear is that 0:15:59.440 --> 0:16:02.400 other animals are not building airplanes, or for that matter, 0:16:02.760 --> 0:16:07.080 fleets of electric vehicles, or neutron bombs or smartphones. And 0:16:07.160 --> 0:16:10.840 it's an interesting question, right, why don't we find wolves 0:16:10.880 --> 0:16:16.200 constructing great libraries, or kangaroo's building hospital systems, or just 0:16:16.400 --> 0:16:20.800 universities where say, Koalas from all over the world come 0:16:20.920 --> 0:16:24.200 loping over and they spend four years to learn everything 0:16:24.240 --> 0:16:27.480 about Kowala history and the future and new technologies, and 0:16:27.840 --> 0:16:31.720 they study all night to take Koala qualifications to make 0:16:31.760 --> 0:16:35.040 them experts in law or medicine, or study their own 0:16:35.360 --> 0:16:41.480 Koala brains. So what I want to consider is whether 0:16:41.640 --> 0:16:45.040 we will be able to leverage our growing knowledge of 0:16:45.120 --> 0:16:53.120 genetics to uplift other animal species. That means, to help them, 0:16:53.240 --> 0:16:59.440 presumably genetically, to have much higher levels of intelligence. The 0:16:59.560 --> 0:17:03.720 idea is that as we better understand genetics, will be 0:17:03.800 --> 0:17:08.520 able to tweak the circuitry to optimize things. After all, 0:17:08.520 --> 0:17:11.520 the important part is that every single organism on the 0:17:11.560 --> 0:17:16.240 Earth is made of exactly the same letters ACTG, and 0:17:16.359 --> 0:17:19.520 there's something just slightly different about the way that our 0:17:19.640 --> 0:17:23.080 letters structure the human brain, so that we not only 0:17:23.080 --> 0:17:27.879 have bipedal walking in hairless faces and acne and really 0:17:27.960 --> 0:17:33.960 useful thumbs, but also we have this massive intelligence. Could 0:17:34.000 --> 0:17:38.640 we crack the code on that part and give intelligence 0:17:38.760 --> 0:17:44.119 to our animal cousins. Now that might sound like pure fantasy, 0:17:44.160 --> 0:17:46.119 but in fact, there have been a growing number of 0:17:46.160 --> 0:17:51.200 experiments that make this easier to envision. For example, there's 0:17:51.200 --> 0:17:54.919 a gene called fox P two, which is thought to 0:17:54.960 --> 0:17:58.360 have something to do with the sudden explosion of speech 0:17:58.440 --> 0:18:02.520 and language in humans. Now, the interpretation about fox P 0:18:02.680 --> 0:18:05.240 two is a bit controversial, but the thing I want 0:18:05.240 --> 0:18:08.360 to point out is that colleagues in mt took a 0:18:08.480 --> 0:18:10.639 version of this gene that was like the human gene, 0:18:10.960 --> 0:18:13.399 and they put it into mice and they were able 0:18:13.440 --> 0:18:16.720 to show that the mice could learn mazes faster than 0:18:16.800 --> 0:18:19.600 normal mice. Now, there's still a long way to go, 0:18:20.200 --> 0:18:22.359 and it won't just be this gene, but hundreds of 0:18:22.359 --> 0:18:25.520 other genes that will be required to make cells in 0:18:25.560 --> 0:18:28.199 the brain hook up in the right ways, and it 0:18:28.320 --> 0:18:31.560 probably won't happen in our lifetime. But the point is, 0:18:32.480 --> 0:18:37.239 animal uplift is a possibility in the distant future, and 0:18:37.280 --> 0:18:41.160 It's something that's important to start thinking about now. As 0:18:41.280 --> 0:18:45.040 Charles Joy said in a recent Scientific American blog, if 0:18:45.080 --> 0:18:49.000 we cannot find aliens in the stars, we might create 0:18:49.280 --> 0:18:54.639 alien intelligences on Earth. Imagine our children looking back on 0:18:54.720 --> 0:18:58.440 this time and saying, Wow, that was the time before 0:18:58.680 --> 0:19:02.720 dogs could get elected as senators, or crows ran certain 0:19:02.840 --> 0:19:07.679 university courses, or zebras were winning the Nobel Prize. What 0:19:07.960 --> 0:19:26.600 a world we might be heading into now, as you 0:19:26.680 --> 0:19:30.920 might imagine these sorts of studies and the imagination that 0:19:30.960 --> 0:19:35.679 gets opened about animal uplift. This sparks massive debate among 0:19:35.720 --> 0:19:40.560 bioethicists and philosophers. The idea is, possibly we will get 0:19:40.600 --> 0:19:43.920 to the point where we can uplift in animal species, 0:19:44.600 --> 0:19:49.040 but should we The author David Brynn, who writes great 0:19:49.040 --> 0:19:53.240 science fiction, wrote a series of novels called the Uplift Series, 0:19:53.280 --> 0:19:56.480 and he explores this idea. And in these books we 0:19:56.560 --> 0:20:00.240 come to live with all kinds of intelligent animals who 0:20:00.280 --> 0:20:03.800 live among us, and they're working parts of society. And 0:20:03.920 --> 0:20:07.840 in a interview, Brynn pointed out that he felt the 0:20:07.880 --> 0:20:12.159 benefits of animal uplift could be amazing and really change 0:20:12.200 --> 0:20:15.399 the planet. And in fact, others argue that it is 0:20:15.440 --> 0:20:20.840 a moral imperative, meaning we're obligated to do this. We're 0:20:20.840 --> 0:20:23.320 like the first ones to get up the ladder onto 0:20:23.359 --> 0:20:26.119 the roof, and we need to extend our hands down 0:20:26.160 --> 0:20:30.159 and lift up all our cousin species who happen to 0:20:30.200 --> 0:20:33.119 be behind us. I think if it, like if my 0:20:33.240 --> 0:20:36.720 dog hurts her leg, I will use human discoveries to 0:20:36.800 --> 0:20:40.480 help her, like antibiotics in a cast, and if she 0:20:40.560 --> 0:20:44.359 gets cancer, I'm going to use imaging technology like MRI 0:20:44.960 --> 0:20:49.720 and oncogenic drugs. My colleague George Davorski, who's a futurist 0:20:49.800 --> 0:20:53.399 and an ethicist, he said in an interview with The 0:20:53.440 --> 0:20:57.640 Boston Globe, quote, there are other creatures on this planet 0:20:57.680 --> 0:21:01.040 that may be in need or deserving of also getting 0:21:01.080 --> 0:21:05.280 these sorts of interventions. We should always be considering the 0:21:05.359 --> 0:21:09.680 larger family of sentient organisms on this planet, not just 0:21:09.880 --> 0:21:13.600 human beings. And in fact, Dvorski argues that if you 0:21:13.800 --> 0:21:17.640 deny this future technology of making a brain better, if 0:21:17.640 --> 0:21:21.280 you deny this to an animal species, it's just as 0:21:21.400 --> 0:21:25.359 unethical as if you deny the education as some group 0:21:25.359 --> 0:21:28.200 of people based on their nationality or race or whatever. 0:21:28.760 --> 0:21:33.520 In other words, we're morally obligated to help. Now. On 0:21:33.560 --> 0:21:36.959 the other side of the argument, philosophers argue that the 0:21:37.000 --> 0:21:39.880 problem is that our long line of experiments to make 0:21:39.960 --> 0:21:44.359 this happen in animals won't be instant because it never is, 0:21:44.920 --> 0:21:47.480 and this will lead to lots of suffering among the 0:21:47.560 --> 0:21:50.720 animal species while we try to uplift them but don't 0:21:50.800 --> 0:21:54.359 quite get it right, and especially if we make them 0:21:54.359 --> 0:21:56.720 smarter but we don't get the rest of the thing 0:21:56.840 --> 0:22:00.879 perfect right away, such that they have meant illness and 0:22:00.920 --> 0:22:03.399 are smart enough to know it, or they have a 0:22:03.480 --> 0:22:06.560 ticking clock of a short lifespan and they're smart enough 0:22:06.600 --> 0:22:09.440 to know it, and this could get really terrible if 0:22:09.440 --> 0:22:12.879 we have to keep experimenting on them once they are 0:22:13.040 --> 0:22:19.160 cognitively enhanced like us. The difficulty is that uplift seems 0:22:19.200 --> 0:22:22.200 like an easy task. Perhaps we just make an animal's 0:22:22.280 --> 0:22:25.520 brain bigger, but it's going to be way harder than that. 0:22:26.000 --> 0:22:30.320 One quick example, in Sweden, some researchers tried to breed 0:22:30.840 --> 0:22:34.600 bigger brains in guppies, these little fish. So they took 0:22:34.600 --> 0:22:37.240 the ones with the biggest brains and kept breeding them 0:22:37.240 --> 0:22:40.159 together for several generations, and they did the same with 0:22:40.240 --> 0:22:44.640 the smallest brains too. Now, this sort of artificial selection 0:22:44.880 --> 0:22:48.399 is what agriculturalists and dog breeders and horse breeders have 0:22:48.440 --> 0:22:52.680 been doing forever. So eventually they had two lines with 0:22:52.760 --> 0:22:56.919 a nine percent size difference in their brains, and they 0:22:56.920 --> 0:22:59.640 were able to show that the bigger brained guppies could 0:22:59.640 --> 0:23:03.920 perform a little better on some simple cognitive tasks. So 0:23:03.960 --> 0:23:07.199 that seemed great, But it turned out the fish with 0:23:07.280 --> 0:23:10.920 the big brains were incapable of producing as many offspring, 0:23:11.000 --> 0:23:15.600 and they had smaller guts. So brains are massively energy 0:23:15.720 --> 0:23:19.399 hungry engines. And the conclusion the scienceists came to is 0:23:19.400 --> 0:23:22.159 that if you want bigger brains, the body needs to 0:23:22.240 --> 0:23:26.520 scale back somewhere else. In other words, big brains come 0:23:26.720 --> 0:23:30.159 with a cost. Now, the study doesn't really tell us 0:23:30.240 --> 0:23:33.959 much about intelligence anyway, because we already saw that brain 0:23:34.200 --> 0:23:38.760 size is almost entirely uncorrelated with intelligence. So there are 0:23:38.800 --> 0:23:41.240 far better studies that need to be done on this. 0:23:41.320 --> 0:23:43.480 But I just want to make a single point here. 0:23:44.000 --> 0:23:48.560 Changes in biology never come by themselves. They're always tied 0:23:48.600 --> 0:23:51.880 to the rest of the functioning of the machinery. So 0:23:51.920 --> 0:23:54.879 if you make a heart bigger or legs longer, or 0:23:54.920 --> 0:23:59.239 a head bigger than other. Things change as well, So 0:23:59.280 --> 0:24:02.920 when the con text of uplifting animals, it might mean 0:24:02.960 --> 0:24:05.680 that we need to figure out genetic changes not only 0:24:05.720 --> 0:24:08.119 to the brain, but to all kinds of aspects of 0:24:08.160 --> 0:24:12.160 their body, which could make the whole endeavor more challenging 0:24:12.400 --> 0:24:14.480 than we thought it was when we were just looking 0:24:14.480 --> 0:24:17.160 at the brain. And again, we know it won't just 0:24:17.240 --> 0:24:21.159 be about making brains bigger. For example, horses have brains 0:24:21.200 --> 0:24:25.800 as big as humans, but they're not discovering mathematics. Presumably 0:24:25.840 --> 0:24:29.720 the issue is that they're not running the same algorithms 0:24:29.800 --> 0:24:32.879 in their forests of neurons, or if you look at 0:24:32.920 --> 0:24:35.760 the great apes are nearest neighbors, they seem to have 0:24:35.840 --> 0:24:39.199 everything that you would need, but they're not building the 0:24:39.240 --> 0:24:43.879 Internet and writing novels and doing molecular biology. Why not, Well, 0:24:43.960 --> 0:24:46.760 it's true their brains are a little smaller, but presumably 0:24:46.800 --> 0:24:50.840 the issue is a more microscopic detail than that. It's 0:24:50.840 --> 0:24:53.800 almost certainly something in the algorithms that are getting run 0:24:53.880 --> 0:24:56.880 by the billions of neurons, So it's more than size. 0:24:56.960 --> 0:25:00.520 It's what programs are running, and there are decades of 0:25:00.560 --> 0:25:03.359 work ahead of us to even scratch the surface of 0:25:03.400 --> 0:25:08.199 that scientific problem. And there's another challenge here. Typically, when 0:25:08.280 --> 0:25:12.320 we think about animal uplift, we picture an animal speaking 0:25:12.400 --> 0:25:15.800 English and having a conversation with us. But it's possible 0:25:16.200 --> 0:25:18.480 that it's going to be very difficult to do animal 0:25:18.560 --> 0:25:22.840 uplift because the intelligence of our species is so tied 0:25:22.920 --> 0:25:28.040 to other things like our opposable thumbs or our larynx, 0:25:28.119 --> 0:25:30.520 which allows us to do things like make podcasts and 0:25:30.600 --> 0:25:33.640 sing songs. And so the argument is, if you give 0:25:33.640 --> 0:25:37.720 a squirrel a better brain or the right algorithms to run, 0:25:38.359 --> 0:25:41.080 it still won't be able to play a flute. And 0:25:41.200 --> 0:25:43.840 even if it can think of some new solution to 0:25:43.920 --> 0:25:47.480 quantum electrodynamics, it can't simply tell it to you because 0:25:47.480 --> 0:25:50.600 it doesn't have a larynx to produce language. So there 0:25:50.640 --> 0:25:52.639 are a whole series of things that may need to 0:25:52.680 --> 0:25:57.520 be put into place so that we can translate languages. 0:25:57.560 --> 0:26:00.240 And I'm not talking about English to Chinese like I 0:26:00.280 --> 0:26:04.240 have on my Google Translate app, but human to dolphin 0:26:04.920 --> 0:26:08.720 or gopher to human, which takes into account the evolutionary 0:26:08.880 --> 0:26:13.440 history of the species and things that would have meaning 0:26:13.520 --> 0:26:17.040 to that animal. For example, a dolphin lives in a 0:26:17.160 --> 0:26:20.640 three dimensional world of movements and patterns, and so it's 0:26:20.720 --> 0:26:24.760 picking up on different things, and it presumably cares about 0:26:24.880 --> 0:26:29.199