The Future of Human Enhancements - Prof. George Church & Experts

[Applause] so I just want to point out that Jamie and I are tasked with creating a realistic vision of the future enhancements and whenever I'm asked that I have I feel I have to point out that when the scientists describes what could happen or answers a question from a journalist is what that doesn't mean that they're advocating that happening so just please keep that in mind I will make futuristic statements now my past is slightly different from Jamie's I think in that in that what I do in my lab is a as I take we take science fiction we turn it into science fact and and that's and that kind of anchors some of my comments so maybe I will be a little less futuristic so when we come to enhancements I want to connect the dots for you so you can not only see what could happen the future but not some kind of leap of faith but how how it could happen there's not just the nuts and bolts but how politically it could happen and so when we have any new therapy any new technology we have a watchdog organizations which are actually quite good at evaluating safety and efficacy for example the FDA in the United States EMA and Europe see FDA in China and they evaluate that but they do not there are certain things they are not tasked to do typically one of those just equitable access worldwide so everybody benefits if there's sufficiently in equal access you cause unrest they do not consider very very long term effects of new technologies they do not consider cultural conflicts that might arise conflicts with our accepted way of life or a religion and so forth and you don't worry too much about the commercial pressures that could happen in the context of fads or just marketing which could make us accept something that is not necessarily our best interests think about you know 32 ounce sugary drinks which as I understood had an interesting history here in the city nevertheless we accept the radical enhancements on a relatively routine basis without even recognizing so and I think we draw too much of a distinction between DNA inheritance and cultural inheritance there both are surely inherited by our children or as surely I mean you can look at your children say that's not my DNA or you can or you can say I have a cell phone and so does my daughter and my daughter's daughter and so forth and and that is surely inherited maybe even more surely than than your DNA and so I think we need to broaden our scope and consider those those sorts of things but what are examples of enhancement you know people or what are we worried about you know what is it that we're really worried about sometimes people paint this really scary nightmarish scenario of people having blond hair and blue eyes for example as if that's that's really threatening our society and but what we have in in practice I mean most of the things that are enhancements that are inherited are physics and chemistry they're not genetics they're they're like if we want to like we're worried that we'll have super athletes will be able to run at 60 miles an hour well we actually have cars that'll go this modern jets and rockets and things like that and we have we can see the entire spectrum from radio waves to gamma rays and everything in between we're not limited to to read through blue anymore but that's not genetic and it doesn't need to be particularly I don't think I don't think we get a huge advantage from that when we talk about you talk about some of the exceptional people in the world that are like a certain sense naturally enhanced I think of people like Helen Keller and Stephen Hawking right so Helen Keller could not hear or see but nevertheless she was a remarkable individual she inspired my wife to go to Radcliffe for college Stephen Hawking inspires me that I will retire when I can move fewer muscles in my body than Stephen could so what are we talking about what is it that we're actually concerned that we're I think we're mainly concerned about inequitable distribution it's okay for us to be enhanced relative to our ancestors we are I mean we don't have to worry about smallpox it's one of the few technologies is actually equitably distributed through an entire planet because we don't even have to do vaccines anymore so we don't have to worry about the cost of distributing and getting healthcare workers to find every single village and find the very last person who has smallpox we're still worrying about that with polio we missed the last person with polio and now there's spread around the world again and we're gonna get rid of it again just like with smallpox that didn't happen right away we're almost done with guinea worm there are a lot of diseases that are human specific and so we really can make them extinct and that's an enhancement there is no question relative to our ancestry but it's not so much an enhancement um that differentiates us that gives one of us an unfair advantage over the others often enhancement we worry about we convoluted we mix it up with germline enhancement there is there is not a one-to-one relationship just be it could be inherited my daughter could get the same enhancement that I get as an adult when she becomes an adult and her daughter and so forth and and to me that's a little scarier because if I want to engineer let's say sperm or other germ cells it takes 20 years to debug that another 20 years for it to go viral and everybody to use it even if it's a big fad but if I if we come up with an adult enhancement that could spread as fast as the Internet as everybody says oh yeah I'll do that you know here's my $5 enhance my brain so let's not get too concerned about germline there cultural evolution and cultural spread is so much faster and so much possibly less expensive too as well so how do we get what happens how do we connect the dots what's what's a likely slippery slope if you will if you don't like it or a beguiling approach if you do I think one of the possibilities that we can't do in the machines that's early already I mentioned machines is a way that we get enhancement right now is a cognitive and an aging we should briefly mentioned I think that the way that we might get cognitive enhancement a lot of these things are going to be we aim for being a new normal or just normal and we overshoot we're looking to fix a disease and we overshoot slightly so somebody has muscular dystrophy and we and the physician says well I could give you the same muscles as your brother but how about I'll aim a little higher than that just in case I fall short then I'd fall short I'll hit the average so it's easy to overshoot and if if you consider a cognitive decline a disease which I think many people is one of the most heart-rending diseases to see your loved ones lose their ability to recognize you and and their ability to care for themselves then we might overshoot slightly or we might provide something that prevents cognitive decline but when used off-label by the younger members of the family provides cognitive enhancement and may not be the goal initially but it could be consequences another example is organ transplants so my lab this is one of my lab works on almost everything I'm talking about but this is one we recently made some breakthroughs in if you're going to transplant our organ you might want that organ to be better than average or at least better than average from the we're transplanting from pigs into humans that's the goal anyway and so you might want that organ to not get cancer when you transplant you might want it if you're dying if you're having trouble with some pathogen like say hepatitis virus you don't want to bring in a fresh liver and have it get hepatitis again so you might want that we're going to be resistant to pathogens resistant to senescence you know the pig might dig organs might die a little faster you want him to die a little bit slower you might want them to be resistant to freezing you might want to be able to cry out preserve these organs put them in so you have them readily available so you might want alternatives to anesthesia or to the kind of drugs that are reduce pain but are also addictive there are humans that actually have chronic insensitivity to pain this is a genetic problem they have but imagine you could turn that on and off wouldn't it be great then you could take the organs out of your freezer and just like put them in yourself without anesthesia since you're passengered resistant so I think I will finish I will give my remaining 18 seconds to Jamie thank you [Applause] and I'll use them wisely so thank you very much I'm thrilled to be here with all of you and three wonderful organizations George mentioned in his talk that what he does is take science fiction and turned it into science faction in fact and for me as a science fiction writer what I do is look at research that George and others are doing and say and ask myself what are the implications of these technologies as we move into the future and the reason why my new book is is a nonfiction book is what I found is that in my book tours for my sci-fi books I was spending 90% of my time explaining to people the real science and people's eyes were just so big and wide and then at the end of my talks I'd say oh no yeah and I wrote some sci-fi books and they're available in the lobby because we're at this incredible moment where science fiction what feels like science fiction is becoming real and it's becoming real at a much faster rate than we are than our limited brains are prepared to internalize and the reason is that when we think backwards in time about well what is a 10-year unit we say well today's 2018 we remember what the world was like in 2008 and so that is one X that 10-year unit of change but because all of these technologies are influencing each other and the information technology revolution inspires the biotech revolution which inspires the information technician they all fold together the speed of change is on an exponential curve and so that 1x unit of change of what we think backwards of 10 years that could be the next 5 years and then have the same amount of the next two years after that the next six months after that so we're all I really need to buckle up because the world where we're going is going to change in a much faster rate than what we're what we're used to and so one of the ways that I like to help people think about where we're going is if we had a time machine and we traveled a thousand years back into the past and we kidnapped a baby don't do this at home because it's not nice to the parents and we brought that baby back to today and placed that that child with the family that child would grow up and be pretty much indistinguishable from the rest of us but if we went a thousand years into the future and got a baby and brought that baby back that baby would a Superman or superwoman by today's standards they'd live longer healthier be immune to all kinds of diseases they'd have capabilities that now we associate with outlier humans of one sort or another they may have capabilities that we that now we animals have and they may have new capabilities to create created from bio bricks from synthetic is from synthetic biology components that we have yet to invent and so how do we get from here to there and I as a sci-fi writer it would be easy for me to say oh we're gonna invent some magical Gazzara and plot machine and that's what's going to do it but really all of the technologies that we would need and will need to fundamentally transform our species already exist and that's the sink with this incredible time that we're living in in because when Watson and Crick and rosalind Franklin and Willa khun's identified that the double helix structure of DNA what they were essentially doing was saying that the the code of life has a structure which is the double helix and sequencing the genome as George Church and others did culminating in 2003 with the human genome project that was figuring out how did the the first steps in how to read the code of life and now with CRISPR and other gene editing tools we're at the phase of being able to write the code of life so when we think of technologies that are readable and writable and hackable what we think of as our information technology I think the kind of the primary insight of where we're going is not only will our IT tools empower the genetics revolution of AI which we're going to talk about in a moment is the underlying empowering tool of the genomics revolution because we need to be able to discern patterns that are beyond the pattern recognition of our brains but we ourselves are going to see humans as what we are is made of code and I know that for people who are spiritual it's hard to think of ourselves as made of code because if we are if we have souls if we are these magical mystical beings it's hard to say well how did that that we are made of code but we are made of code and that code will be seen by us as writable and readable and hackable and as I see it where we're going I is going to happen in three overlapping phases all of which are already oh there are already happening so the first is everyone here is familiar with IVF and pre-implantation embryo screening PGD or or PGs and basically what that is you know with IVF is that a woman has her eggs extracted average woman has about 15 eggs extracted during IVF then usually a smaller number of those are fertilized by the male sperm and average male ejaculation has hundreds of thousands and sometimes hundreds of millions Masari of sperm and so then you have these ten or fifteen or whatever the number is fertilized eggs and so prior to implantation and the mother after about five or six days a couple of cells are taken from those early stage embryos and sequenced and right now because we're at a relatively early stage of the genomics revolution what are the things that we can look for and can find one is we can look for single gene mutation diseases so-called Mendelian diseases and then a small number of traits like hair color and eye color and things like that and gender and this is in some jurisdictions those those other traits are that you can allow for giving that information to people in other in other jurisdictions not but right now that's what we can do but with genome sequencing entering this new phase and when more and more people are going to have their genomes sequenced and that's going to happen as part of precision medicine which I know the people at the New York Academy have spoken about before but what precision medicine means is rather than having general medicine where if you're kind of you have a headache you take a Tylenol even though you may be the one in whatever a couple hundred thousand people who could get sick or die from a Tylenol you find out by taking the tylenol in precision medicine the foundation of your medical record is going to be your sequenced genome and some of the some other things and so the as we move towards millions and then billions of people whose sequence genome is in their electronic medical record we're going to be able to do the big data and lytx comparing their genotypes with their genes say to their phenotypes how those genes are expressed over the course of their lifetimes and because of that we're going to be able to understand more and more of what genes do and in this process of selecting pre implanted embryos it won't just be about selecting out Mendelian diseases but we're going to be able to predict with greater or lesser probabilities the likelihood of certain disease states and ultimately certain traits being realized over the course of a lifetime if any of those embryos are brought to term and become children so that's going to speed up the process the second phase is that we're going to be able very likely going to be able to create not just so right now as I mentioned the average woman has 10 to 15 or so eggs extracted during IVF but using induced pluripotent stem cells we're going to be able to take adult cells probably blood cells or skin cells and induced induced those cells into stem cells those stem cells into egg or sperm precursor cells and those cells into egg or sperm and so sperm we don't need because we as I mentioned before we have so much of it and it's being given away Willy knowing but eggs for four humans are actually very are actually very rare and more precious and so now rather than having ten to fifteen eggs we're going to be able to have for every woman or any woman hundreds or thousands of their own natural eggs and so rather than having 15 early-stage pre implanted embryos run implanted embryos from which to choose we could have hundreds or thousands and so if you are choosing and if we know about certain traits and the genetic patterns that underlies certain traits like intelligence like appearance whatever like personality style anything that has a genetic foundation or part genetic foundation we're going to be able to have a greater range of options and people will will do that on top of that it could be possible and maybe will be possible to mate pre implanted embryos with other pre implanted embryos so one person goes through this whole process and develop and then they have their one of a thousand we implanted embryos that they have selected and somebody else does the same thing now you have two unemployed embryos and then from this one you extract eggs you do this that same process and from this one do the same process to have sperm and then those embryos can then mate with each other and so now for a human average American woman has a first child at 27 and so that's a generational turnover is 27 years if you're doing it with this other way of mating embryos with each other that means that we could go through 50 for generations in that same 27 years and think about what that does to how different changes could potentially be pushed through a population if we were focused on what we were trying to achieve and just one one shorthand for this is that our ancestors who didn't know anything about genetics took the wild chicken which laid one egg a month and turned that into the domestic chicken which lays one egg a day so if we are focused and informed and have all this technology behind us and we want to do this and I bet most people won't want to do this we have the technology to push this kind of change through our population and everything I said up to this point doesn't involve CRISPR or gene editing at all these are all how we would breed our own natural children and then on top of that then there is all the technology for for gene editing I think you're all familiar with it and again already now there's a lot of very preliminary work for gene editing early stage pre implanted embryos of animals to make single gene mutation single gene changes and that's would be very useful so if somebody has a disease like Huntington's disease and their code or sickle cell disease there could be a single single change that could eliminate potentially the heritability of that terrible disease and so people are going to want all of these kinds of things and when we think about well what will be the phasing of these enhancements I think first will be embryo selection and people will do it first based on selecting out diseases first simple single gene diseases and then more complex diseases based on probabilities and then people will get comfortable with that and so the next phase will be we're already going to be comfortable with selecting embryos we're going to have more information about what the genes are saying and then when you're getting the information before selecting out diseases someone will also say well you know of these ten embryos this one is likely to have the highest at least for the genetic portion of IQ this one is likely to have a sunny disposition or whatever it is any any trait that has a genetic foundation people are going to be able to to select for and that is going to lead George talked about about social norms that that we are going to get more and more comfortable with bringing advanced technology into ourselves and into our reproduction process and over time that's going to change not just the way that we make babies but the nature of the babies that we that we make and so then over time as we get more comfortable with this kind of change then be because we are these hubristic humans that have never for any technology we've always thought yes we should really understand this technology before deploying it but we always deploy the technology that would be useful to us and it's just it's who we are it's our history of a species and there are just very big and fundamental and for many people myself included frightening questions there are ethical questions about what would what will happen if because if maybe even very well-meaning well-intentioned individual decisions we limit the diversity of our species because everybody thinks that they want to have higher IQ kids or kids who are more likely to live longer or be immune to diseases and it's not just that that diversity genetic diversity has been a protective strategy for our species it's been our sole survival strategy for our species so if we would limit our genetic diversity that could be potentially very dangerous dangerous for us equity issues are are enormous and many sci-fi riders and other writers have been thinking about this at least since HG Wells and and before but what happens if one group of people has access to a technology that makes them immune to diseases or healthy or have higher IQ or maybe none of those things but can convince other people that they have those advantages we only need to look at the history of colonialism to understand what one group of people with a real or perceived set of advantages can do to bring it to help themselves at the expenses at the expense of other people so there's very big ethical issues and one of the challenges that we're going to face is that we are humans and we are a very diverse species and we have many different individual beliefs communal beliefs and national beliefs George and I both do a lot in China and when you think of what's happening in China and the norms that people are bringing to these kinds of questions it's very different from what's from what's happening here in the United States in China there's a plan that half of all newborns will be sequenced by 2020 so we have these very big differences and then we are all locked in this competitive arms race both on an individual level and on a communal level that is going to drive us for it it's going to be very very difficult to put the brakes on all this and so what it all means though is that because this change this future is happening so quickly we can't just leave it up to the to the scientists or the experts or even our governments that everybody needs to be part of this conversation because it's a conversation about the future of our species thank you [Applause] thank you very much we're gonna go straight now into the second panel which is going to elaborate further on some of the ways that Jamie just left this conversation on the social and ethical implications of these technologies and we're gonna hear first from Josephine Johnston Josie is a scholar at the Hastings Center and she is also our director of research a lawyer by training she's an expert on the ethical legal and policy implications of a wide variety of technologies particularly Human Reproduction psychiatry genetics and neuroscience in addition to numerous scholarly publications her commentaries have appeared in Stata News the New Republic Time Washington Post and a scientist and she also is very frequently interviewed by the media she's appeared and been quoted in the New York Times The Wall Street Journal The Guardian you get the idea and ABC Nightline I don't want to not mention that her current projects address the ethical implications of new kinds of prenatal genetic tests and the potential use of genetic sequencing in newborns she's also a member of the Columbia University Medical Center's Center for Excellence and ethical legal and social implications which is looking at psychiatric neurologic and behavioral genetics last week we were very proud to learn that Josie's been named to the New York State SEM stem cell board ethics committee which advises about how the state spends its money on research that um and the degree to which that research should go forward based on its compliance with ethics guidelines Meredith Whitaker is a research distinguished research scientist at New York University where she co-founded and co-directs artist artificial intelligence now AI now which is an interdisciplinary research center dedicated to studying the social implications of artificial intelligence dr. Whitaker has over a decade of experience working in industry leading product and engineering teams she was the founder of Google's open research group and she co-founded in LUP which is a globally network measurement system which provides one of the primary resources for policymakers and researchers investigating net neutrality and network performance she's worked extensively on issues of privacy and security she's advised the White House the FCC the city of New York the European Parliament and many other governments and civil society and and she has done this consulting on artificial intelligence Internet policy measurement privacy and security okay hi everyone one thing Millie didn't mention is that I have a really great accent you might want one too it's a New Zealand accent just so you don't have to spend the whole time trying to figure that out and so I know that that might make me a little hard to understand I'll try to go slow and I have some slides and some with some words on it so that hopefully nothing that I say gets lost in translation so as Molly said in her introduction I work in bioethics which is an interdisciplinary field and in my field and in the relate in related areas in philosophy there's a lively and fairly long-running scholarly debate about the ethics of human enhancement and that debate those who are enthusiastic about enhancement see the widespread and frequent use of biomedical and artificial intelligence technologies as the pathway to a new utopia critics of the enhancement project expressed concern about a new set of haves and have-nots and worried that enhanced humans will converge on a problem ethically narrow set of norms or ideals my remarks today are going to draw on this enhancement debate but also on my experience of becoming and being a parent in fact I'm going to begin my discussion of the ethics of enhancement by reflecting on those personal experiences before expressing some of the same ideas in more scholarly or academic terms so I'm going to I was the pitcher this is my photo of me and my daughter on the day of her birth in July 2009 she was born as we all are susceptible to a large number of contagious diseases not long after this photo was taken she began to be immunized when she was two months old she had a particularly large set of immunizations she was vaccinated against diphtheria tetanus pertussis Haemophilus influenzae type B pneumococcus and polio she was a little baby just six pounds one ounce in this photo and even by two months old she was still pretty skinny yet my husband and I gave permission for the doctor to push a needle into her tiny little legs not once but four times she screamed we both whipped and I spent the rest of the day walking her up and down the small strip of lawn outside our apartment while she cried and grumbled and fussed she was in pain and out of sorts all day she was also enhanced on that day in just a few minutes my daughter acquired immunity to sex diseases that her grandparents and great-grandparents either did not have or that they had to develop the OP the hard way by getting and fighting off the disease in an instant she was as Peter Kramer the American psychiatrist and author of listening to Prozac wrote better than well and I would have it no other way these are horrible diseases if she had contracted them she would perhaps have learned something about suffering and patients and the value of life but they also could have killed her so we skipped those life lessons and suffering in my utopia we are all enhanced in this way we do not suffer from horrible diseases even in an age of anti-vaxxers I don't think this is a controversial position to embrace the kind of enhancement that vaccinations bestow or that particular kinds of gene edits might one day offer most people want this kind of enhancement and they see it as a valid part of medicine where things get more complex and a rightly so I think is where we are offered or promised biomedical interventions to eradicate or avoid traits conditions or differences that are closer to our identities our histories our culture's and our belief systems this is where the enhancement rubber really hits the road since my daughter's two-month vaccination bonanza she's received quite a few more shots further enhancing her immune system she's also learned to read write and draw her father determined to pass on his heritage has spoken German to her each and every day for the past nine years so she has also learned to speak a second language she can ride a horse and she's on a swim team struggling to coordinate her arms legs and head turns to breathe she's learned or perhaps I should say is still learning to share her toys and indeed to understand why sharing matters she learns these and so many other things the hard way there is no pill or shot to make her more empathetic or more athletic or to turn her into a myth was she must struggle practice make mistakes feel the pain of sharing and the smart of jealousy almost all this learning happens through engagement and in the company of other children whose struggles are both similar and different from hers her father and I are right there we try to help her become kinder and smarter and more attentive to the richness of the world she inhabits I'm literally working to enhance my child everyday but not by changing her biology at least not directly and quickly in the way that I gladly did to see her immunized against polio this kind of enhancement we are doing the old-fashioned way using methods tried and tested over decades if not centuries these methods are predictably fairly slow they involve repetition there are surprises she seems to have perfect pitch I'm where did that come from wasn't from me she struggles and myth where I breezed through I'm stretched all the time to parent her in ways that help her grow I really smile geez I read the literature unknown predictors of childhood flourishing I amend my goals my hopes my dreams for her I find freedom in not deciding just exactly how she will be and along the way I too AM enhanced as I never gate parenting's balance both of shaping her and leading her unfold according to some plan that comes from I know not where I would have this no other way as well so let me just transition now to some more I use sort of academics types of here are four terms to denote four arguments that caution against a full throttled embrace of biotechnological enhancement there isn't time now to go into any one of these in any detail and there are certainly both other ways to express the same arguments and concerns or questions that are missing from the slide all together the first is the means in this ins distinction which is just a fancy way of saying what I said earlier when I shared that I do not look forward to a pearler an injection or a tweak to an embryos genes that aim to make my child more empathetic without her having to experience the road to empathy one might say the same for musical ability or academic achievement she's on her way to those capacities through hard work engagement reason experience in some degree of suffering and I don't know how far she will get the second term I want to put on the table is hubris which has already been mentioned and I think it's especially important as we consider making permanent even heritable changes we are often clear and over-the-top long term correct about what counts as a disease or is what needs correction or change it seems safe to say that cancer will never be understood as just a different way of being but we have also been quite wrong in the past it's just one illustration think of the previous classification of homosexuality as a mental disorder where we now understand it as simply one of many valid ways of being in the world if we sometimes make mistakes about the distinction between disease or dysfunction in simple difference we might also make mistakes about what counts as a real improvement to the human condition complicit complicity were suspect norms the idea are familiar to so many of us that it can be very hard not succumbed to social pressure or narrowly defined conceptions of beauty or talent or worth different from the time at state-sponsored eugenics these norms are unlikely to be forced on people today but a poor men might choose to alter his child's genes for enhanced athletic ability as a way to secure access to college a black mother in America might choose to alter her son's genes so that he has paler skin and his least subject to racial discrimination this is sometimes called liberal eugenics where we freely choose to alter ourselves and our children the government doesn't compel us they don't need to but the norms themselves may still be insidious and the pressures to conform so strong as to make it almost impossible to say no finally I have here the term hyper agency which was introduced into the debate by the political philosopher Michael Sandel he uses it to describe what he takes to be a mistaken approach to parenting one in which parents seek to remake nature to serve their purposes and satisfy their desire as a kind of helicopter parenting on steroids literally maybe I would only add for now that hyper agency could be a kind of felt obligation something parents don't want to do but feel that they must in order to compete we can debate with a hyper agency misunderstands the true nature of parenting but it certainly seems to be one way to totally ruin the parenting experience altogether I literally think it would be horrible to have to be the kind of parent who selects and therefore controls her child's traits and talents so this brings me to my last slide in which I seek to replace some of these concerns these worries with some positive goals some things to aim for freedom diversity engagement and wonder I want to stress it in the three seconds I have remaining that people like me who are critical of the Utopia of the enhancement enthusiasts can still find some kinds of enhancement laudable even necessary I'm definitely for vaccination and I'm also I also think that we should continue to ensure that out children are kinder and smarter than their parents but it's crucial to remember that that kind of enhancement is less likely to come from biomedical intervention from changes to our genes or muscles or brains than it is to come from engagement with our children and from a much broader range of social and cultural changes to provide all children with environments in which they can thrive these kinds of social political and economic changes are not generally previewed under the bio technological approach to enhancement even though many have decades of rigorous scientific evidence to back them up in my utopia we focus on these social determinants of health and beyond health of human flourishing we address poverty poor nutrition literacy neighborhood safety parental health and we ask hard questions of those who bring us biomedical interventions that promise quick fixes and endless control thank you hello everyone thank you so much for having me as a song I am wearing one oh hello this one is on thank you so much for having me I am meredith whitaker and i just want to clarify I'm not a doctor but they let me in anyway and I'm also still at Google leading their open research group in addition to NYU so all sorts of biographical clarity there so why invite an AI researcher to speak and it event at an event on human enhancement some may be wondering well beyond the historical entwine Minh a AI and neuroscience which I'll say for another talk I'd say it's because AI techniques and the infrastructure required for AI comprised the foundation on which many of these enhancement technologies rely and I think Jorge touched on this using neural data genomic data other data sources a eyes ability to process and find patterns in large quantities of data is applied to create models of the human from our physical ability to our propensity for illness to our mental state and of course a model of the human is required for any serious discussion of the enhanced human so it's through this lens that I'm going to be looking at the current state of neuro technologies so a narrower frame alongside a eye and I'm going to be looking at that in the context of the economic social and political environment in which they're situated I'm going to explore some of the challenges we should be aware of as we move to enhance human beings and before I begin a quick disclaimer this talk is gonna get a little dark so I think we we started with a sanguine and we're gonna we're gonna end with the somber and this is not because I'm not optimistic indeed I think there's great promise for many of these approaches I'm focusing on the dark side because I don't think we stand a chance of realizing any of these benefits if we aren't clear-eyed about their potential downsides so with that let's start by examining the current AI landscape now AI isn't new the research field is actually over 60 years old but you'd be forgiven for thinking that it was new because suddenly in the last five or so years it's everywhere you can't pass a newsstand without seeing another magical AI cover story so we need to ask why are we hearing so much about AI now and in answering this we hit on something that's omitted from the shiny robot stories that often dominate popular narratives and I think it's something we need to remember as we discuss the risks and opportunities of AI based enhancement technologies because the technological breakthroughs that propelled the current AI gold rush from deep face to alphago are all contingent on the vast power and resources of the current tech business ecosystem they simply could not exist or even be imagined without it so it's no accident that the recent AI boom happened at the same time that we're seeing the consolidation of the tech industry indeed in my estimation there are only seven or so companies with the means to create AI at scale these are those that have three things all at once first they have vast computational power this is proprietary chips and supercomputing clusters that are effectively unavailable outside second they can afford to pay the LeBron James salaries to scarce and highly trained technical experts who develop AI algorithms these guys are sort of rockstars in the community and third they have massive massive amounts of data the kind and quantity of data that it's almost impossible to get without vast and pervasive market reach so given this we should not be surprised that those looking to shake up the neuro tech industry and bring such enhancements to market are some of the very same companies that already have power and resources to make AI indeed these are among the few position to store to process to interpret the necessary data and importantly to monetize such technologies so when we consider AI and enhancements today we need to think as much about power as we do about tech right now Facebook has a team of 60 engineers working to build a brain computer interface that promises to let you type with your mind they put it this way quote we want to create a digital assistant that can literally listen to your thoughts anywhere at anytime and privately and quote privately you might ask indeed now of course by privately they mean without you having to say anything out loud in your environment not without you giving Facebook the ability to literally listen to your thoughts now Microsoft is also in the mix it published a patent about six months ago titled changing an application state using neurological data so you wear some kind of headband and presto you can use your thoughts to close a browser tab same premise now never one to shy away from tech from futurist fantasies there's Elon Musk's neural link it's a company developing an ultra high bandwidth syringe injected neural mint mesh meant to embed itself in the brain this is one more step toward his dream of avoiding human or Elan obsolescence by connecting humans to computers neural link promises quote direct lag free interactions between our brains and external devices so you can control your phone your nest thermometer your smart home your dumb car whatever using only your mind now maybe this sounds appealing to some of you it could but of course such things don't just happen this requires a lot of centralized infrastructure and indeed if you dig a bit deeper you'll see that most of the neural link job ads are looking for infrastructure folks these are people able to build end-to-end development storage and compute pipelines able to scale to petabytes of data and hundreds of developers across multiple clouds translation your thoughts in the form of neural data measured by the embedded brain mesh will be sent to the neural link server server infrastructure where this data will be stored interpreted processed and almost certainly used to train and calibrate the next generation of brain models which neural link relies on to interpret your thoughts in the first place now if the current law governing tech companies still applies this data will also be owned by neural link who may or may not be willing to fight a National Security Letter requesting your thought log so there are many other examples but these should suffice to make the point enhancement technologies like brain computing interfaces require incredible resources especially if you want to bring them to market and these are the kind of resources rarely available outside of industrial industrial tech or you know massive DARPA funded labs so when we talk about neuro tech and enhancement we need to seriously consider the risks of a future in which a handful of private companies whose incentives may or may not align with ours own and monetize a map of our thoughts of ourselves and of how we respond and feel at any given moment where these companies could be positioned to interpret our thoughts our psyches and our bodies with more authority than we are so pause for a moment there and then turn to the thorny issue of AI bias and inaccuracy because in many many cases the determinations and predictions made based on AI models turn out to be flatly wrong they turn out to perpetuate and amplify systemic inequality and discrimination under cover of flashy and in fact infallible innovation and this too shouldn't surprise us from the least advanced algorithm to the newest convolutional neural net AI draws its insights from data and data comes from the world we live in it reflects the histories and practices of our present and past including its hierarch it's power dynamics and the legacies of marginalization compounding this AI systems are currently designed by a very very small very privileged sub strata of the population in the western context these are mostly white mostly men almost all affluent mostly located in the Bay Area now many of them are also my dear dear friends and many of them care a lot about these issues but that does not mean that their worldviews and assumptions aren't encoded in the technology they create best intentions or no so this leads to facial recognition systems that are over 30% less accurate for dark-skinned women than they are for white men to machine translation systems that encode gender bias changing doctor to him even when translating from a language with gender-neutral pronouns to voice recognition systems that don't hear women's voices as well as men's and on and on there's a litany of examples and this is an issue we need to foreground when we look at how we model the human this normative archetype that enhancement technologies are being created to improve about a year ago I was struck during a conversation with a well-known neuroscientist we were talking data collection we were talking measurements and he mentioned an EEG measurement Helmick used for collecting neuro data the helmet is designed for people with large heads he said and it doesn't work on some smaller headed folk which mean that it works better overall for men than for women now since the helmet cost $200,000 his lab only has one and by relying on this one helmet for data collection his lab is also subtly but very meaningfully centering men as the norm and women as a nice-to-have which gets to a really gnarly epistemological question at the heart of these issues who gets to decide what's normal because again you can't have enhanced without some definition of normal is this an affluent Western version of normal is this the normal of a San Francisco tech VP or of the shuttle driver who transports him how do we ensure that this concept can change over time that different versions of normal can exist between cultures and communities that the norm isn't constructed in ways that serve to justify and naturalize existing marginalization think phrenology physiognomy and this matters you know reference the earlier talk we have only got to go back to the 1960s to look at the polish political abuse of psychiatry when changes to the Diagnostic and Statistical Manual manual of mental disorders or the DSM led to it the increased diagnosis of black civil rights leaders as schizophrenic based largely on pathologizing activism or to a time before 1973 when as mentioned the DSM included homosexuality among its listed mental disorders or even just a couple years ago before late 2012 when it included being trans it is impossible not to see the high stakes here and impossible to ignore the immense damage that oppressive classifications of normalcy have done to those who fall outside now this is where we need to touch on politics because bluntly none of this is happening in a vacuum we are in the age of Trump we're seeing the shift toward authoritarianism and an erosion of civil rights and liberties at the same time that we're seeing AI technologies increasingly used to manipulate monitor and catalog people in ways that serve certain forms of centralized power we can only imagine the potential harm where the next Cambridge analytic to gain access to the intimate neural data of 220 adult Americans or if this data was shared with ice or with corrupt law enforcement or with our employers or our health insurance or tabulated as part of our citizen credit score these are all scenarios we seriously need to consider because too often when we hear about these breakthroughs and the hopes of tech Titans we hear only the story is crafted by their marketing departments and the promises of benevolent applications that go along with these stories so we need to bring our own critical thinking to the table well AI driven enhancements hold promise the economic social and political context in which they're created deployed and used is as important or maybe even more important than the capabilities of the tech itself we need to persistently examine how such technology might be used to manipulate surveil and control and we need to look at power asking who gets to define normal and who falls outside of this category thank you it's a wonderful set of commentary and now I'd like to invite Glenda to join us Glenda if you'd like to come on up as the president of the Aspen brain Institute and sort of my corollary here I've liked Glenda to have the opportunity to ask the first couple of questions and we're gonna do this sort of as a fishbowl with you listening in and then I promise you we're going to turn it open it to you as well well yes I think I think everybody deserves a round of applause Millie I was just telling you that I had several several questions so I wonder if I can just mention the questions and maybe you can decide which ones you want to take hold up one is about what is often called the posthuman future so I just would like to know or have you discussed when does that happen what is posthuman where is this defining mark of when we are still a human and when we've gone beyond human that's one thought to talk about another is should we lay out the case for political control or global monitoring of emerging technologies I think one of you brought up the fact that in China for example with very few regulations they're saying that it would be possible to raise the IQ of a whole generation of Chinese people by 10 to 15 points and to continue to keep doing that so it seemed obvious that we need some kind of monitoring of regulation of these new enhancements then maybe the big question is cool as I believe you said Meredith who was gonna make who are who is gonna make these decisions I went to a wonderful conference at Harvard genomics at a summit with dr. George Church and his wife dr. King Wu and dr. Wu gave us four simple average kind of dilemmas that might come up for all of us lay people to decide and as she gave these four dilemmas let's just take one average dilemma that you might have about any kind of genetic sequencing or whatever and she said this was a very intelligent audience of about 60 people and she put us in four corners whoever would agree with this decision on the dilemma go in that corner whoever's almost convinced go over there whoever hates this decision go over there and one that almost hates it go somewhere else and in these four dilemmas that she spoke to us about we were somewhat evenly divided in these four corners these were very intelligent thoughtful people but weak on the simplest of dilemmas we could not there was no consensus for a decision and it just showed to me how what are the needs going to be we talked about having an inclusive conversation and getting the public as all of us are ready for these dilemmas but this is a serious issue and I would like to have that discussion I think the final issue is what I think the Lauer last two speakers were talking about are we going to protect the rights of citizens to either maintain or modify our own minds and bodies well we have to create some new rights we don't all want to modify and enhance every part of our body and new babies etc could we do we have the right to maintain what we have or modify that's a terrific should start with your first question so Glenda use with you first then when do we arrive at this postmodern period most human post - I'm sorry post human periods when do we know we're there am i friendly amendment it - it is why do we want to go there there's been enormous exuberance a right around what many people call transhumanism and a certain sense of inevitability about it so let this question how do we know we're there my question why do we want to go there anybody who so you know what one thing I want a you need a definition of what transhuman of posthuman is and one of my working definitions I'm sometimes accused of being a transhumanist under I'm not but anyway is that that the person in this transhuman is unrecognizable to what was an average human of the past and I would say fry that definition we are already transhuman that is to say if we were to transport someone from the path with a far distant past to the president or vice versa or even take them from one of the islands where they they live in a Stone Age culture that said that then you could redefine it as we don't recognize we the current ones don't recognize the transience but that's a moving target where we keep progress we need to keep coming the the human of the future and we memorize the increment so big that we can't relate to them their words even though we're vastly different from my parents generation we can wreck it we can that they can identify us and we can identify that and that so it may never be a time that's rust why we might want to do it creates an interesting litmus test that we can apply which is imagine that it were standard practice would we want to take it away there was a you know we heard that we wouldn't want to take away our vaccines right or some of us would not let's pick over the vaccines that imagine that we have all these other enhancers we're talking about imagine very concretely that they're safe they're effective they're equitably distributed wouldn't want to take them back so um the New Zealand philosopher Nick agar talks about a distinction between radical enhancement and sort of more sort of ordinary enhancement I guess and in radical enhancement is pointing out that these this kind of post-humans situation now I think it's hard to define what posthuman is I mean who here read Chaucer okay how many hundreds of years old is there Shakespeare so like I don't know how I'm recognizable we really are like we have different clothes on but we can speak many of the same emotional language and time but if you imagine someone tried to think of something that's incredibly radically different from who we are now Nick would say well like what interest do we have in there like literally what interest do I have in some future persons who have nothing to do with me remember in her the movie we eventually she just isn't interested anymore right like what interest so it's like I think it's a it's I don't know that we do it before us you're like why would we work toward something that is so alien from us that we actually have no connection anymore to it so I think like that's that's where what we're I suppose go with that question I think like what is parsed human like is it something somebody who is sown who is so different from me that I really literally cannot connect and I don't think we really have had that in human history yeah we I mean I think if we if we asked a Stone Age person living in Stone Age culture to connect with you know why we're worried about going some populating Mars or why it is that we worry about taking our cell phone all the time they might not connect with that and they might wonder why did they go to that you know why did they go that far and that's why I think that this idea of these huge jump it's a it's a fake it's a fake possibility and maybe we'll get there but we are going to line up all of us and our children and our grandchildren are going to be are going to line up for these small incremental benefits that all of this technology is going to provide just like as George was saying we have in the past and certainly there will be new possibilities for people to do things that feel uncomfortable to us and some of those we will see are just an affront to human dignity and just to give an example I was just reading I think was in The Times an article about domestication of dogs and so they're identifying that there's just a small number of genetic mutations in dogs that make them so domesticated and docile compared to wolves and I could imagine a situation where there was some totalitarian government that was using some of these technologies to make sure that everybody in their country had those kinds of of mutations or variations so that they could control them and we would feel well that's maybe a bridge too far but I think that more often the 99.9% of this technology is going to be a bunch of relatively small incremental steps and we're going to want it and we have to have regulation because what we're talking about is the future of life and we all need to be part of that of that process but I don't see that it's going to be possible by the time it's possible to make these huge jumps so that everybody today is seeing some other person it's and they're saying oh that's not a human that's a post human will already have adapted to a whole new set of social norms well I think you're correct and I was thinking maybe one point post human might be the book you may have already called Homo Deus so when man becomes God and man becomes God when he can decide and design his own evolution but I think what you're all saying is even that designing our own evolution self customizing ourselves won't happen in large chunks where suddenly we say I am God like I am it'll just it'll happen - it won't feel like I mean we're doing it now we've refined 'mentally transformed our evolutionary trajectory as a species through agriculture and medicine and all these kinds of things so we won't feel like it's a step from one to hundred at once I don't think we all do agree though Glenda I think I hear a lot of different strands there's a sense of inevitability and a definition of progress that I hear from Jamie and and and George but I also hear a big cautionary note from yep from Josie and from Meredith so I think this is worth drilling into Meredith yeah well I I share the view that I never quite know what post humor in our trans human means but it seems to be always a very general concern as this sort of monolithic pronoun we will and then you look around the world and you know it appears that we may be failing on any meaningful climate regulation right so the idea of a post human seems much more present in that reality where we can't even get it together there let alone this idea of us all sort of tweaking a menu of bits until we become these like creatures that can unrecognizable so I you know and that's that's a bit flip but it is also like I whatever we build today is going to be the foundation on which what we do in the future sits and we have yet to figure out how to distribute food equitably so I don't you know these concerns often occupy rooms full of people who you know are very interested in sort of thought experiments but I don't you know I have yet to get a sort of what's the leap from where we are today to a material practical future in which these you know aren't either a hobby or something that sort of you know deeply define defines populations in a world that is you know has other troubles on those so you think we aren't getting it's not around the corner is that a fair oh yeah I still don't quite know what it is yeah well the general definition that you know if you in the literature is basically an effort to extend life of very far away almost close you know maybe a thousand years old oh and a general enhancement of cognitive abilities the ability to genetically change our ability to see in different in different spectra and even moral enhancements you talked about possibly the creating you know more easygoing populace which could be a very harmful political move but parents might also want to at least as as this narrative goes might want to find might want to tweak the personality of their children to make it easier okay and all of these are relative measures so if you're say someone's going to live line with the anyone's ever going to live in thousand years but if you say people are going to live longer its longer relative to the social expectations of how long a person should live and I definitely agree that these equity issues are huge but often when people are talking about transhumanism they're having this debate like should we live to 500 or whatever and then but I just think that if we are if we want to address those equity issues for some point in the foreseeable or distant future the best way to do it is to say listen look at the world today like we don't need transhumanism because everybody in this room is a transhumanist compared to the average person in the Central African Republic and so if we want to model our behavior for when technology is going to be more advanced we should start doing it and living those values today but there are also ways in which every one of us is also like less able than the only person in the Central African Republic right and that's what it's lost when the people driving the vision are the people in this room and their friends because there are things that people in the Central African Republic can do that I can't do for sure that are incredible and none of that counts unless and I wouldn't I wouldn't say that because I mean I am NOT a pure relativist because the average life expectancy of everybody in this room is probably somewhere around 85 the average life expectancy of the people in the Central African Republic well I'm sure our incredible people with all these other skills is probably 45 so we are at least in that axis of age we are transhumanists but well who do you think that is some sort of basic biological difference or the fact that you have affordances that come through yes colonialism and extraction and they're you know you know that's my point so I did it if we want to if we're thinking about equity in the future let's think about equity now and model our value system yes we're not going to do it but I do so we would you please find a microphone and we're recording this so microphones are really important yes first person there um yeah are there any red lines for instance something like human cloning or are they many red lines we can all agree upon that should not be crossed or do we just take the attitude that technology will always put progress sure I think there's a tendency to try to draw red lines around mechanisms and things that just sound funny rather than on outcomes so for example we might draw one at cloning or germline or or enhancements and when you try to either try to define those or think of specific cases you realize that that we've thrown the baby out with the bathwater we've we really should be focusing on what outcomes Sagara so do we you know are there cases the National Academy of Sciences said there are cases where germline would be preferable to current practice which is abortion in a vitro fertilization there there are cases I mentioned where enhancement might be the consequence of just curing a disease like I would to decline so anyway that that would be what would concern me about the red lines we do in in in the legal system we have to draw all our non distinction so far we don't intend not to make probabilistic distinctions when we can avoid it I would just say that better be black red lines we have the technology right now to build black mayor right we're seeing a social credit score implemented that I alluded to in China right like if we don't you know coming from my technological realm if we don't think quickly about what these red lines are we're going to be entering into what could be a very frightening future can you describe what a social credit score's um oh it's horrifying um and and it's actually I would say it's not that different from some things that are happening from a market-based approach in the u.s. so it is a sort of aggregate of your you know educational history your friend graphs your sort of other data your credit history your employment history that is you know calculated into a score in your record of being critical of the government right as a personal Yelp Yelp rating I said think about everything you've ever posted online as aggregated by the government to give you a score that ranks you know what kind of citizen you are this has already been used to bar twelve million people from regional chance travel in China so it's a very punitive measure and it shows speech and a number of other activities they have facial recognition cameras that can automatically ticket people for jaywalking at the moment of jaywalking that goes on your social credit score so it's happening and it's there you know I'm not talking about enhancements exactly because I don't know as much about that but you know we're talking about a world where the the capabilities are are there you know as George was saying so we need to be able to make choices if we can't make choices we're looking at you know again a very frightening future and the parallel in genetics would be state-imposed eugenics programs what's the parallel you said there's a market pair and we and now ways don't think that's going to happen in eugenics it's going to be more consumer driven yes Jamie just told us we're gonna be lining up so it's not the government that's gonna be promoting genomic modification so what's the market driven an analogue to what you described in China a very quick example is a company called fama it licenses itself out to you know employers to companies it offers to troll Facebook and other social media sites and analyze using AI as they advertise it the data of potential applicants to see if they're a good employee or a bad employee based on a proprietary algorithm one of the things they look for is the possibility of someone having debt that would bar them from employment so you're looking it's the same logic it's applied slightly differently and it's more piecemeal here okay I just have to add that um governments aren't going to necessarily impose eugenics or at least maybe this government on people but if you refuse to address poverty or racism or sex discrimination or any other kinds of discrimination in your country then yeah people will be signing up for things that allow them to control those capacities so you can you know you don't need to impose it if you create the social conditions where it's the only real valuable choice yeah over here yes thanks for organizing this conference a question a fun question and another comment on inequality the question is can we find in this genetic code something that human beings learn how to accept their mistakes and be more humble because because for 200 years we believe that the neurological change would take us to you know to paradise or living Keynes was telling us just 15 years ago that we will live in the age of Plenty but now that will be important for me to know if they genetically instead of worrying about physical characteristics we go to the values that we have if that is reproduced to the next generations so we could when somebody makes mistakes or makes music I work for the United Nations and I tell the same thing when we make mistakes you say sorry excuse me I made a mistake and not after 200 years that we will now recreate a climate change a drought it's a humility actually humble that we discover now but we have to be more humble when we discover new thing because we made a mistake of 200 years now when the immigrants just this is a this is a question a kind of comment in this room there are also new first generation immigrants whose life expectancy is much less than 85 and the problem of averages I am economists I like average to the program average is this is that distorts the diversity of distribution in the world I just finished with my colleagues the new technologies study what is the implications on economics at the United Nations where I work for 18 years and there are people who as you mention who are living perhaps without 1 billion people lives with our electricity still and 2 million people don't have secure electricity so one one thing that worries me about the humility is the loss of diversity which was kind of reflected Jaime said that we we might at some point the government might want us to be domesticated I would say we are already domesticated relative to our ancestors we used to be a bit more murderous than we are now a little bit less humble in the sense that we were convinced that that killing somebody that opposed us was the best way if you believe state Steven Pinker is maybe even more pervasive than that so I but I worry that that even though I like the fact that we're more domesticated I worry that if if we set that as a goal any goal that that causes to strength the bell curve from this wide diversity down to the middle of lker for purposes of say educational convenience Henry Ford meets the classroom could be a problem so I second the motion of having more humility if it's voluntary but that the voluntary runs in the problems of manipulation by corporate giants and it may be that limiting the diversity will follow our greatest values it may be that we decide this X is so wonderful it's so wonderful for everybody to be empathic and pick your highest value of disagreeing with George whatever that is if we all follow that that could be a tremendous danger it could be the it could be a huge threat to our species so we don't know and I think that what we need to have is a level of humility and a level of hubris because that's the history of our species and the challenges I mean this stuff is frightening but we're also talking about these tools that are going to make all kinds of wonderful new realities possible we're going to be able to eliminate all these terrible diseases nobody has seized a young person dying of cancer and says wow that's so natural like it's hubristic it doesn't say we're going to overcome these things and it's it's all woven together yes we need to apply that is similarly Keynes is an easy one we can all agree on that but what about deafness right like with this just this so much of a distance to travel which mean if it's a wonderful new possibility then sure we're all for it but when you're wonderful new possibility might be someone else's nightmare right like deafness or other kinds of disabilities that people don't actually think need to be eliminated we're probably most people think it but you know a lot of people don't including people who know what they're talking about because they have the disability so I just we should be so suspicious we should but not entirely suspicions they need to be cautious about okay one word when discussing genetic potential for reprogramming ourselves genetically in germ lines or not one of the words that was used was was hacking and that's really resonated Rett resonated with me since that term came up because we see for example with computer and Internet technology wonderful revolutions what you do and all this stuff but then if one country or one company or one smart enough group of individuals decides to hack the system of somebody else suddenly we don't have to talk about what we should do and shouldn't do because someone else gets to make the decision so what what is the potential for hacking the genome who doesn't express my viruses that's one way to have the system right what's the potential and and and how is it preventable or is it product thank you well we could well we're pretty good at and hacking agricultural species and almost everything we've talked about about behavioral changes involved in domestication and so forth we know how to do that it's been done in a variety of species including most recently silver foxes I think that the way we're gonna hack it probably is some somebody creates a commercial incentive and then everybody does it except for maybe the ruling class and they don't do it because they know what's what's wrong with that picture and so everybody becomes humble except for a few people very very powerful yes I'm over here okay thank you all question that's a little closer to present day in terms of the the struggles we're facing with technology today and enhanced humanity through smartphones and the internet and other things that are at our disposal right now when I think about technology I think about it in three rough categories one it's the category of technology that's helped us create distance between us in the harshness of nature in order to promote our survival which we've talked about a lot tonight there's also technology that promotes convenience and comfort but it seems like we've reached this this age of technology of mass distraction where when we think about the number of people on the internet right now somewhere like three or four billion people 2.2 billion of them on Facebook and other social networks of course and it seems like it's distracting folks from these hard questions even on the smallest scale of everyday lives or how they raise their children and all that issues facing us now how how should we be thinking about that because it seems like it's quite an epidemic when we have social networks that are these distraction machines and these smartphones that are keeping us away from that at a massive scale it seems like something we should talk about I show you a concern I mean I guess I would go back to the engineer and scholar of Technology Ursula Franklin who defined technology is the way we do things around here you know we're talking not about this sort of continuum from keeping us away from nature that naturally and deterministically evolved in to us having smart phones in our hand we're talking about a number of you know late capitalist incentives that created a business model that was very profitable monetizing networked technologies so that's the sort of social and political and economic forces that led to where we are today I think we need to take a clear look at you know what how do we how do we rein in ad tech in our society right how do we begin to have some kind of accountability and transparency around what these are doing how they're doing it and how do we make a choice around where these technologies are and where they aren't because where you do see them is on smartphones right you're scrolling and you're you know looking but you don't see them is in many many interactions you have on a daily basis where you're being catalogued and ranked and otherwise assessed by backend technologies that are using the same types of logics and the same types of infrastructures often in a feedback loop with the data you're feeding into your phone Jimmy has yeah so I'm you know I'm against everybody spending all their time watching cat videos but you what your question is embedded in the broader issue of connectivity and this connectivity is such an incredible mirakl when you think of everybody who's doing anything and is if you're if you're doing scientific research or inventing things you come up to a series of problems and you have to solve one problem after the next for every problem that you're going to face there's somebody out in the world some group of people who are already working on that problem and so being able to connect to those other people means that you can go all the way to the end of what's been done and then take a step further so that coming to me yes there's the downside of all this distraction but this is something that is so incredibly empowering to our species that I think that that if we're going to think about the distraction we should think about the huge benefits over here thank you so there were a couple of comments where you take an initial idea that was done for good and then of course once you do it for good it can be done for a whole bunch of things so you can imagine something with CRISPR Cass there was sort of an apocalyptic future mentioned of what do we do if we can actually read people's minds why would we want to do that I'm the chairman of the research committee for the ALS Association we just spent some money so that we could do just that because there are patients who cannot communicate at all with the world even there though they're still thinking so we're trying to read their minds so they can communicate with the world so the question is how do you take a technology that was done for good and limit its use so that it doesn't and it's been a huge problem in history so it's how really had thing to do oh sorry yeah okay good I mean because I think in some ways the term enhancement technology is a misnomer right because usually it's just a technology that can have enhancement uses and therapeutic uses and everything in between them so it's really about the use a lot of the time but one way that right now at least in this country the use is limited is that you know insurance won't pay for certain kinds of things right so it's just it's just very expensive if you want to use it we don't have a medical in the and medicine becomes this kind of door for a lot of these things to go through and medical and insurance companies not saying that's the best way to really use but it's one of the ways there that we have of doing it which obviously means that some rich person can buy that thing for themselves I'm not sure I care if I don't really think it's actually in an like a great thing and that we should all have access to I have a follow-up question to your question that I want to pose to George because I bet that the Broad Institute for example when it's licensed going to licensed crispr Cassadine is going to control it and draw some red lines by saying that they're not gonna license certain enhancement uses but think will license other therapeutic uses if that's true then companies themselves could have a very powerful influence over controlling how how their technologies are being used when they create those licenses is that true like am I right about that what a member of the Brodus - I did not vote in favor of that particular clause I think it's tentative I think it's tentative and a subject to definition I think it's be very hard to legislate that either at the academic or the intellectual property owner level at the corporate level or at the government level I think would be very hard to say what is an enhancement it'll be case-by-case I think they're just trying to show off their power quite frankly I don't think I don't think Christopher was that big a deal either yeah I mean I think Dana slow did something like that with non-invasive prenatal testing in Hong Kong where he wouldn't license if they used for sex determination because he was so worried about sex Lewton nevertheless there is quite a bit of that going on you know he was dealing with them I mean it's it's the thing about these jurisdictional solutions right they're limited by jurisdiction and that's why they answer in my mind to the question that was just asked is we have to develop norms and I'm one that I do believe that there need to be red lines but they need to be constantly renegotiated and they need to be based on norms and the way we get norms is by having engaged inclusive dialogues in order to figure out what is a bridge too far yes so you guys talked a lot about transhumanism but this is like a really old discussion on like HG Wells the time machine thought about this right it was literally I go to the future and I meet humans and the protagonist does not identify them as humans for three-quarters of the novel and the whole point is that everything that has happened to the future people at least one half of society is good like they have unlimited food they don't get sick they literally sit around and play and have sex all day then there's like the underbelly of the society the poor people who run the city and like have to eat the rich people oh that's how they they eat the rich but this is an old discussion but the whole argument he makes is that every step towards that was so small and always good or we always thought we were doing the right thing how do we see far out how do we see to the future HG Wells and make sure we're not doing the same thing that you described thank you we have to address the social conditions that would make us to make those kinds of mistakes I mean as long as we have people hungry and discriminated against an impoverished entrant will have all the seeds for all of the stuff leave you know we don't want to happen and what's so bad about eating the rich continuing we can have whatever ethical guidelines or rulings or even read lines we have but you and I know that people of power and money and means are going to blow right past that how do we deal with that or do we just accept that as something that's going to happen I mean I would use the word regulation I would say we need well-funded expert oversight we need to fund research into I mean or or not you seem to be reservation is local there are places in the world where people who means could do anything they want yeah well I I agree that this is a really difficult challenge but what I don't want to do is slip into a kind of somber determinism where we just figure that the worst-case future is inevitable because we have yet to have come up with the perfect solution for how to stop that kind of avarice look at biological weapons I mean there are things that are terrible things that were very frightened of and yet somehow we've developed imperfect global norms that are somewhat enforceable enforceable enough to prevent some of the worst abuses I mean some terrible things do happen and that doesn't mean we don't try to make sure they don't happen here right like whatever we can control so there's really I think quite disturbing practices around surrogacy in india that Westerners are exploiting the front and center but that doesn't mean that we should just give up on trying to regulates and manage surrogacy in a way that's humane and good here so I you know I just it's so depressing to think that because it we can't control it all and there'll be some people somewhere who do it anyway that we shouldn't bother this really speaks to one of Glenda's first questions and I want to be sure that you know I come back to that which was you really asked two parts of the same question who should decide but also what are the regular regulatory mechanisms that we want to adopt and I wonder if before we go to the next person over here if anyone on the on the panel wants to but ultimately now respond to what you've asked it's very much relates to what we heard from the floor what are some regulatory mechanisms that we might consider and you might want to in your answer you might want to compare us to some other countries for example the HIV yeah and in the UK so well we're looking at you because you were the first one brave enough to use the word oh I know okay you know I can talk about a very small you know AI specific case the a I know Institute recently published a framework called the algorithmic impact assessment and this is you know suggesting that in the case of public agencies where they're using algorithmic systems that are often licensed for vendors that are often untested to make you know determinations about who gets resources or you know where someone goes to school healthcare etc that those need to be opened up for review that we need to be able to look at the data and the source code and that we need to be able to bring in interdisciplinary expertise so not just you know a technical AI researcher but people who actually understand the context in which this will be working to look specifically at the impact on marginalized communities whose voices should be centered in any discussion around you know the potential effects here so that's a that is a very small-scale example but it is incredibly practical because what it shows is that what we don't even have yet is a list of what these systems are where they are applied and how they may or may not affect us so we're starting you know when we're using these terms we're starting at a place of almost zero where marginal improvements could be really meaningful and could lead to a better ability to answer that question more specifically which we won't have if we don't get that kind of insight in the reproductive space well I mean there are countries who have more who have reproductive have frameworks for each for oversight and regulation of use of reproductive technologies this is not one of those jurisdictions the UK almost many others have a system I'm not sure like I don't endorse everything about them but they you know they've figured out how to the UK's had one since 1990 and it's been mending it and changing it through parliamentary debate and public discussion route I mean they have a way of talking about stuff that I think we should envy you off you know and I really wanted to bring out the reproductive regulatory framework in the UK because the irony is that in that case regulation is enabling certain technologies to go forward like mitochondrial transfer which is not able to go forward here because we haven't created a regulatory system so it's a perfect example of how regulation actually is facilitative of the okay it's going forward in ways that it's been stopped here where we haven't had a public framework for discussing this and that's and that's just one easy thing somebody was here from the United Nations every country should have some kind of regulatory framework it can be it can be especially tailored for their values and their systems their political system but every country should have one I think I would be great if the United Nations set up a system to share best practices and this would work just as well for for AI and then we can come together and try to develop norms and figure out what are these changing red lines that are that are beyond the pale for now just mention the US does regulate reproduction I would look at the Ohio bill that almost oh yeah okay so I just moved out of a national system yeah yeah there's a other kind of recoup yeah all over yes sir oh thank you so much I didn't think I'd be able to stand I was in my my local coffee shop the other day and I was reading something that my father had typed in 1947 he was his green rider so it was a treatment for a movie and it had onion skin paper and it was a carbon copy and it was all bound together by you know those two metal things he put old over I'm just reading it I'm interested in seeing what it has to say they got a tap on the shoulder I turn around and there's a 20 mid-20s young man and he says excuse me sir I said yes he says what is that I explained to him but it's just a way of conveying information so I feel that with Professor Church isn't you know that I'm not so sure that we can't negotiate you know generational differences and somehow if we have a chance to have coffee we might be able to work it out but I asked him I said so you're young but you must have seen things change very fast in your life he said yeah I said I'm not even sure I believe in the future anymore and I and since then I must say I've talked to 20 year olds and they don't necessarily believe in the future that's predictable but I said well well what do you think is going to happen he said well there's one thing I'm afraid my girlfriend's likes the idea but I don't and I think it's coming in about 20 years and the way it's the end of history and he says it's the singularity and I mean I wonder do you think that that's kind of like the elephant in the room that we're not talking about it in that area there may be developments so fast that the whole notion of talking about human enhancement may be kind of moot thank you I mean I spent too much of my career managing broken servers backstage trying to reconfigure broken systems working with people who carried a pager so when something in the data center went down they had to get up at night and fix it I am very skeptical of any claims especially from people who haven't really touched infrastructure and in you know the last 20 years that technology will suddenly gain sentience at the bottom layer it still requires precarious workers to label data it still requires massive data collection pipelines in a it requires huge armies of workers to continually tends to what are always clucky and almost broken server infrastructure so I don't believe in the singularity the singularity is far yeah I can see that's for sure okay we are coming up against the time there's a few there's still so many people left so I do have a question but two two quick comments to the gentleman who mentioned his research into ALS look for me afterwards because I have a colleague in Austria at GTECH that has made some remarkable accomplishments in that area and anyone else was interested so the question is this now this may come across as an unanswerable question about an ephemeral concept but I don't think it is so I'm going to ask anyway so I've noticed that one thing when people talk about other people just in general about the human species they refer when they're looking at any kinds of behaviors well it's human nature even Einstein believed that human nature was fixed and said and believed that all of the changes that had to be made what had to be socio cultural and political so I'm thinking that's not the case because we have certain genetically expressed evolutionary neurobiological predispositions we have 20 or 200 biases oh we only know perhaps 20 of them or can experience them we have the unstoppable need to form in group out groups that often lead to much of our misery so there's there are many many behavioral characteristics that lead to symptoms and some of the symptoms of what we're discussing tonight so my question becomes then has anyone on the panel ever considered looking at applying genetic editing and Synthetic Genomics and everything else that has been discussed because that I'm involved with that in the various limbic and other structures as well as the cortical structures to address the fact that the individual that creates technology can do so for the betterment and the destruction equally we can talk about it later if you're interested I think there's a great deal about that but I don't think there's a lot being done on genetic engineering I disagree with that don't tell you I love you I think our speakers will will hang around during the reception and will be happy to answer further questions plenty do you have a final word that you'd like I'm good I can't wait to discuss all there's and fries all right [Applause] you

Channel: The Artificial Intelligence Channel

Views: 13,864

Rating: 4.6435642 out of 5

Keywords: singularity, ai, artificial intelligence, deep learning, machine learning, deepmind, robots, robotics, self-driving cars, driverless cars, Human Enhancements, gene therapy, crispr

Id: 3fYGdchv32w

Length: 99min 32sec (5972 seconds)

Published: Thu Jun 21 2018