Pushkin from Pushkin Industries. This is deep background the show where we explored the stories behind the stories in the news. I'm Noah Feldman. Welcome to this week's program, where we're going to talk about gene editing, designer babies, and the future of science. These are all topics of importance all the time, but they've become more pressing since the announcement that a Chinese scientist had actually used Crisper Cast nine, the leading cutting edge editing technology, to edit the genomes of two embryos to assure that they would not be able to be susceptible to the HIV virus. This was good for the embryos, but it wasn't necessary scientifically, and it's led to lots of intense discussion about whether the era of designer babies was too close and needed to be headed off by regulation. To discuss these and closely related issues were super fortunate to have with us George Church. George is the Robert Winthrop Professor of Genetics at Harvard Medical School. He also teaches at MIT. He's been a central actor in the development of the technologies of gene editing and in applying them to the creation of new genomes. He sometimes, in fact calls himself a genome engineer. George, thank you so much for joining us. It's a great pleasure to be here. Noah, let's just start with the headline in order to make sense of what it really means and whether there are dangerous associated with it of the kind that many imagine. We read in the paper that Jan Qui, a Chinese scientist, says, or has it said about him, that he used Crisper cast nine to engineer the DNA of a couple of babies, and that they were then subsequently carried to term. That's what appears to be new here, and he's been The Chinese government's not very happy with him, and he has been under house arrest and radio silence more or less since. From the standpoint of the state of crisper technology, was there anything remarkable about the accomplishment, assuming it was accomplished, or is it relatively a trivial step compared to the scientific advances that had already been in place? Yeah, I would think similar things had been done by other groups, including groups in the United States and Oregon on human embryos, and even more amazing things had been done on other mammalian embryos, including altering dozens of genes. So is here the claim was only the altering of onengenee. You know, there's some interesting questions about choice of genes. You know, one of the things that's challenging in this field is picking a gene that even a few other people would agree is a good choice of gene, because you know, a fair number of genes could be fixed by other methods, Like you have to be in in vitro fertilization or IVS clinic to do this at all, and if you're in that clinic, you could fix it by just IVF prenatal genetic testing EGT. So when it goes to question of choosing the target of the intervention, what are the right criteria or what do you see as the main criteria people would consider. I mean, you've mentioned now one could you get there by some other means? Could you get there without without changing the germ line? But what are other criteria that you think would be relevant here? So in addition to could you get thereout altering the germ line, there's also the criteria of is it a net positive for that person or in a public health sinse on average on that positive and environmentally dependent. So if you're in an environment where everybody gets HIV and dies young, and that's a very different environment than or nobody gets HIV and China is not ground zero for the highest level of HIV in the world, who is highly stigmatized there, but you know places in Africa have higher incidence of it and it would be better candidates. Another one is, you know, is costs is a consideration if you want something that's going to be of true public health benefit for a disease like this impacts a lot of impoverished individuals because you have the alternative is drugs or safe sex. There there are a few alternatives, none of what you're working perfectly worldwide because a million people die for years, it's about two percent of all human deaths. So to claim that this is a solved problem is just as Naives saying that this is the only way to solve the problem. Right, it's not a self problem in real world descriptive terms, even though in principle it could be solvable. So what about the criterion that asks whether something is presently an immediate threat to either the embryo or the local population as opposed to offering some background improvement for the potential person who might come to be created, or for the environment more generally, does that seem like a meaningful difference to you? And I just to show that I'm why I'm asking that question. It goes to the broader societal fear about what are sometimes called designer babies, you know, babies who are who might have their genomes edited to give them some advantage, which could be a health advantage to begin with, but could also be other kinds of advantages, whether in esthetics or intelligence, or athletics or what have you. I think this is something that is very interesting entanglement of concerns. So the concerns are that we will create a monoculture number one, where everyone is the same, the way that you might have a you know, many square kilometers of identical crops. The second is that it will be inequably distributed, in no words, that some people have some people don't. That's a very different concern than you don't want anybody to have it, as you want everybody to have it, okay, you know. A third one is you could make a mistake that it could be a very popular fad, but it was a mistake that has long term consequences, either for the people who had it or the people who didn't get it. There could be stigmatization issues direction. Yeah, it could beat not to have had the intention or to have absolutely and there are a few more things that get tangled up here. Sometimes it's phrased as blondere blue eye, which doesn't make any sense at all. You know, that's not a public health threat. It's unlikely to have long term consequences even if it were a monoculture. It's not a monoculture that threatens society if it's not. But it's not an accident that people use that as the example. What they're invoking when they say that is there associating genetic engineering with the eugenics movement of the eighteen eighties, nineties or early twentieth century, which was not only by any stretch of imagination, popular in Germany but all over the world, including very much the US. Yeah, the United States kept doing it long after World War Two. I think it was into early nineteen seventies, and so eugenics though this wasn't the only line of eugenic thinking at all, but it married itself up at least in the European context, even you know now long discredited racialized theories. That's what people mean, I think when they say absolutely absolutely. And furthermore, there's all kinds of body image issues, shaming and so forth the en conjures up, which some of which are racially independent. You know that within a race you'll have you know, obesity shaming and many other features that determine your status of society, your socio economic status, enhance your health status. So even though it's not health explicitly, that has ramifications in that direction. So some of those can be addressed. So you know, equitable distribution, we could bring down the price as we have for various technologies like smallpox. Vaccine has made it extinct and so it's basically free now. So there are technologies that come down pretty quickly to zero, so that you can take some of these things at least not necessarily off the table, but you can put them aside in a separate category. But another one is that you could create It falls a bit in the equitable distribution, but you could create people that it's not necessarily racial. It's just a new capability right now, most new capabilities. I believe that the human population at present doesn't have right, and that enhancement is something that we should feel very familiar with because we are enormously enhanced relatives to our enhancestors. We're not actually fearful of enhancement as far as I can tell. But that do you mean incrementally enhanced? I mean so if you look at I don't know the fastest one hundred meter dash times, they're getting faster, though presumably they're also going to a mutual limit. But I can beat one hundred meter dash person any day in my jet. But that's a capacity enhancement that is not embodied. We have developed tools that enable us to do things, but they're not quite the same as being incorporated. It's an interesting question why we treat them differently because they are hurtable in a certain sense culturally culturally inheritable. But culture is an inheritance. I consider many ways more threatening and more rapid than DNA obsessed inheritance. You know, it's much more likely that my daughter will have my cell phone then she will have my facial features. Thankfully, so we get hung up on this sort of DNA obsession. If they actually knew less science. Oddly, we might draw less of a line between these, Right, we didn't know how we inherited these things. We didn't really understand culture and technology and DNA they would kind of look similar. Oh then it's your fault for helping us understand it, so well. Sorry, Well, you're making a really interesting observation that culture has a wide range of effects. It's in some broad sense heritable. But I want to add that it's also subject to the full range nearly every one of the serious risks that you described with respect to genetic editing. In fact, in many cases, the way we come up with our fear is we look at the distortions or the injustices produced by culture and we say, well, oh boy, you know, genetic editing might have the same effect, so you know, societal inequity. Check on that. Right. It can also these target effects which we haven't yet gotten to. But you know, that's broadly speaking, the idea that if we edit one thing, there will be unforeseen effects somewhere else in the genome. Check on that. In culture, all the time, we improve one thing and we make other things much worse, so certain consequences. Check on that. Go ahead, Yeah, I mean the early adopters sometimes they're more exposed to the unintended consequences the off targets, and so they're essentially be signing up with their extra dollars to become the first guinea pigs. So it seems like there's some meaningful debate within the scientific community about how concerned we should be that particular desirable genetic edits. This is assume it's we're avoiding a disease, that there's no other way to avoid it, that we fit your other criteria. There seems to be some disagreement between one campus scientists who say, because we don't know for sure what the off target effects might be of a given intervention, we should proceed extremely slowly and carefully, and another campus scientists, again this is a continuum, who seem to say, we can measure off target effects like we can measure anything else, and if we reach the point that off target effects are less likely to occur from a genetic edit than they are in nature, then it's absurd to be so worried about it. And I'm fascinated by this question because as a layman, I have no idea how to go about answering it right. Well, the probably the most glib answer to it is this is the responsibility of food and drug administration in the United States and equivalence in other countries. And it happens with every single technology to something expose. Certainly the medical technologies were the medical devices, or molecule drugs or protein drugs. Every category has all target effects which are physiological, so you need to do as much as you can theoretically, followed by as much as you can with animal models or human cells in culture, and then you move on very cautlessly to one person to do phase one toxicity and then efficacy, and then make sure that it's really ready for primetime, and you scale up the size of the cohort cautlessly so you expose the minimum number of people to risk for good theoretical reasons. So this combination of theory and testing cautiously is what protects us from all our new technologies. Is the timescale particularly challenging in the case of genetic editing because you're editing an embryo, but it might be that the off target effect doesn't express itself, you know, typically until much later in life, so that it might take much longer to find out what potential left target effects are. Absolutely you know, preventive medicine is a nice buzzword, but it's very hard to develop powerful preventive medicines because you're testing them on people that are healthy. And it's not restricted to embryo editing, its fetal surgery with actual microscalpels of things and monus sickness drugs like Blood of Mind, which which failed disastrous effects. Extreme caution is required when you're dealing with prevented in medicine. For example, you take chemotherapy as an adult woman, you could be affecting your German line in ways that you also won't see until your babies are born, and maybe their babies are born. Right used sub interesting words now, which I think are maybe they are characteristic give you, although I don't think of them as characteristic of you, namely the words extreme caution. So I want to ask you about your headge about the FDA. Of course, institutionally we assign those decisions in our democracy to the FDA, but that doesn't necessarily tell us what criteria the FDA ought to apply, or, more to the point, how cautious the FDA should be, or how risk taking the FDA should be. I mean, real human beings. Ideally, scientists and statisticians sitting in the FDA have to make these decisions on the basis of what data is known. So I guess what I wanted to ask you is, if I'm right that there is this continuum of how much risk we should take in the scientific community around gene editing, where do you fall on that continuum? The words extreme cautions sort of put you sound like they put you among the Really, let's not rush this side. I think tend to fall on the I worry about everything, and I want everybody else. I don't want to reassure people necessarily, I'm not in a big rush for most things, even aging reversal, even though I'm I'm not in a big rush show, so don't worry about it. I'm not in a big rush. But that said, I I don't fall into the camp of we should put as many barriers in the way as possible so that it never happens, or be so vague about it that we can never be satisfied as to what the criteria are for letting it go forward. And it is a continuum, and I think the FDA does an admirable job of prioritizing so if you have a very serious disease is going to kill you tomorrow, they have a different threshold. Then if you have a healthy baby who has every expectation of being healthy, and you're going to give it something that might extend its life by ten years eight years from now, that barrier is very hard to get permission to do that study. And rightly so, there's a line of thought that says, using the FDA as one pole of you describe pretty admirable cost benefit weighing, and then on the other side, looking at China, where at least in one instance, there seems to have been a lot of risk taking, it says, look, it's all well and good to say that the FDA should be cautious, but out there in the world, outside the reach of more cautious and maybe typically democratic although that doesn't have to follow governments, we're going to get lots of innovation, lots of risk taking, and that we'll fall behind if we listen to the FDA or we allow the FDA to be cautious, or alternatively, that says we ought to be very aggressively going out and trying to export our own limitations by pressuring foreign governments to make sure that they crack down to a greater extent. How do you think about that phenomenon? The science is increasingly globalized, as prices come down, as scientists train across borders. There's no in principal reason that a country with a reasonable infrastructure scientific infrastructure can't make all kinds of innovations under a very different regular environment. Well, and first of all, I would say that the Chinese FDA to c FDA is very similar. I think they have a similar risk profile. I think their government whether you want to call it capital hyper capitalistic, or less than perfectly democratic, it doesn't really matter. The point is they are capable and often do crackdown better than we do. So in the United States it's very hard to interfere with the industry, not because they run Congress or anything, but because there's an obsession with freedom. And in the Chinese government they can crack down on things that are unsafe in ways that are very difficult to do otherwise. So for example, cleaning up Pijing for the Olympics, and that's something that I don't think we could have done. And they're very technocratic too. I meant I'd heard that a huge faction, maybe eighty percent of their top politicians have degrees in science for engineering, and not because they're failed scientists or engineers, but because as politicians they went back and got those degrees because I felt it was important for understanding the issues. That is so far from where we are. We're more in a culture of if I need some facts, I'll give them to you, right. You know, this is a question that I've been very curious to ask you today. To be a cutting edge researcher as you are, it's also perhaps expect it is too strong a word, but not surprising when one also holds lots of patents and starts companies that develop those patents or that try to develop those patterns. So the entrepreneurial side seems to coexist very naturally, easily, almost normatively with being a leading scientist. How do you think about that relationship, because that's been an important part of your practice. So if you have an invention where you've gone from the vast cloud of scientific discovery to something that you think might be useful, it's not sufficient to write up academic ivory tower paper and hope for the best because nobody's going to use it. You know, sometimes there's some concern that capitalism in that form will infect us, you know, that affects us by manipulating us, or as market scientists entrepreneur cannot be as pure as the scientist who's received grants and does basic research. I think a lot of people fear that. Yeah, and they fear that even after the sign us steps away and goes back the Ivory Tower, a company that has been created as a life of its own, and it will be motivated to use marketing and advertising and operative lobbying to give us our opinion, run listen to us. And I think these are all valid things. On the other hand, if you're talking about companies going rogue, there's certain challenges to that, meaning that the CEO reports the bord of directors and to the stockholders in general, and so to do something blatantly irresponsible result in a huge public relations problem, your stots value and rash and so forth. While an academic with tenure who has acted responsible, they can go off and do whatever they want. They don't really report to anybody. So there's a danger in other words, also in having the pure scientists who can do whatever he wants or she wants, and it's not responsible at all. So let's take this then back to the question of the future of Crisper or other similar editing technologies in terms of the state of the science. Now, how credible is it that existing editing technologies could, if they were allowed to by regulators, actually make interventions at population level in a significant way. Are there actually traits enough traits that matter that are addressable by editing one or two genes or a handful of genes to actually make it make an impact. How much of this is science fiction fantasy and how much of this is within the reach of reality. So there's an interesting phenomenon. I would say a huge fraction of my colleagues would reassure you that human genetics is so complex that we are so far away we don't need to worry about it. I think that is false to some of them. I think that is false reassurance. So I think that we have sort of the textbook cases of complex genetics or things like height. Right, thousands of genes involved, possibly all the genes are involved, and they each have a very tiny effect along with hundreds of environmental effects. You know, why is it over the generations we've gotten taller and tallerance better nutritions. So this seems hard to predict, much less to manipulate. And when your colleagues just to flush out the argument for listeners when they say nothing to worry about, they often do use height and they say, well, look, if we know right now that there are four or five hundred genes involved in just producing a time percentage of the variants right in thousands to get you to the whole variety, we're never, or at least not in the foreseeable future, are going to be able to edit all of those different locations. So therefore, you know, don't worry your pretty little head about it, or that's what they say to me. Sore. There's three things wrong with that. One is, there are single genes that are so impactful that even though in the natural population they hardly add anything to it, in the medical situation, they have a huge impact, in particular for height. There's about seven different medical situations where physicians and patients want to do something about height. It isn't necessarily because they're genetically defective, and it's some other reason, And the answer is one gene. One gene product was somatotropin or sometimes called growth hormone. To me, that just really nails the counter arguments. So the counter argument is the geneticists who don't want you to worry, say there's so many different genes they're involved, and you naturally right, And you say, well, that might be true naturally, but that doesn't mean there isn't an intervention we could make exactly on a particular gene that would actually have the effect of overshadowing overshadowing the natural variations. So, in other words, and it's a mistake to reason from natural variation to the capacities of technological intervention exactly. And this is not an argument, and this is not hypothetical, because there are seven different so called diseases or medical treatments that are approved and are in routine use to involve this one genes a metatropa. Second, they'll say, oh, we'll have unintended consequences. Well, the fact is in medicine we always have unintended consequences. We almost always deal with them. We just say the benefits outweigh the risks. So that's an ethical argument that they make. That The first argument is a don't worry because it's not going to be practical to engage in these kinds of systematic edits. This argument set is the is the ethical concern that says, we don't know there might be off target effects, and you're saying, there always are, and we were accustomed to that, and we we balanced the costs in the benefit. The distinction is blurred between practical and ethical very often because you'll say it's impractical because there's these off targets. The thing is it is totally practical because almost everything has off targets and you maybe take a second drug to deal with it, or you stratify the population, or there's all sorts of things. There's the tricks you can use, so don't they might be target effects. When you hear those two arguments, these are the grains of salt, you should take it with, okay, because they're pretty big grains. And the third one is the technology is so far from standing still. I mean, it is growing exponentially that that even if you say we can't do it today, it could be you know, somewhere between minutes in years, not centuries. They often use the word centuries. It's incredible, there's no centuries anymore. So. So for example, my blab's personal record, a pretty much world record for editing genes free Crisper or around the time of Christophers two, we could addit two genes, and we were very proud of ourselves. We edited more than one. Just just literally two years later, we had edited sixty two genes simultaneously to make a pig that didn't have any viruses and retroviruses in its you know. Just a year or two after that, we now have edited thirteen thousand genes simultaneously in a mammal in human human cells. Okay, not a human important to think. So the point is this is moving very quickly. What's great is we're having these conversations well in advance. But what used to be well in advance. We started talking about designer babies in the early days of our comment at DNA and in vitro fertilizations sort of in the early seventies. That was well in advance. But the conversations we're having now are not forty years in advance. They're more like a decade. When I look at the overall picture that you describe, you know, the extraordinary change, the speed of change. I wonder whether the institutions that we've talked about before the FDA or the Chinese FDA are really strong enough and capable enough to resist the temptations they're going to come with the tremendous technological capacities that are being developed. I mean, if we look at our history, we don't have a great history of using institutions, regulatory institutions to block the cutting edge. The cutting edge usually tends to emerge. And that's partly, I think because of the power of capital, you know, which you've mentioned here. It's also partly that once something is possible, there will be people advocating for it. If you have a rare disease, you will say, look, it's not that it would be unethical to try to cure this disease. It would be unethical not to try to cure this disease. So can we hold back against the potential risks here? You know, we say, now, well, of course we would never do this to give people purely esthetic advantages, but in reality maybe we would. It is true that capitalism can distort the FDA, and it's ilk around the world, but I think there is a there is still feedback mechanism that's bigger than all of this, which is that if it really doesn't work or isn't cost effective or desirable, we will stop using it. In fact, if it doesn't work. My worry is more, what if by work, I mean doesn't work for society that the long term benefits society. And they are examples. So first, what do FDAs around the world not do? So they do safety and efficacy, but they don't necessarily do equitable distribution or very long term risk. They have long term risks in mind, but the way they've set up the structure is if you can convince that on a short term risk basis that you've done all the experiments, you get to test in phase four, which is basically selling it, and then once capitalism kicks in, then it's very hard to regulate. So perfect examples of this are nutritional supplements and stem cells. Those managed to get into the marketplace without full FDA approval and now they're bigger than the rest of the pharmaceutical industry and so they're hard to regulate. But bad example will be the opioid crisis. I mean, it was just about gerty. It's a good one. So this is something where if you create something that's sufficiently addictive, people will try it and then they become lobbyists. They vote, They vote with their wallet, which is in many ways more powerful than voting with ballots. That said, there will be a feedback system. If any country manages to get a business model or a country policy model that prevents this, they will win because the nations that can't control their opiated crisis become impoverished, mismanaged, etc. They become the new cash. Can I push back on that? I mean, so everything we've talked about until now, I'm the purest lay person, But now when we talk about feedback and the governments, I have something to say. And here's what I would want to say. We've got lots of examples where societies inflict huge costs on themselves, but the costs are not so great that they turn them into, as you would say, third world countries, because they have other tremendous advantages. So tobacco would be a really good example. Yeah, you know, I remember my grandfather, who was trained in pharmacy school in the nineteen twenties, describing a film that they were all shown of the horrible effects on the human lung of tobacco smoking. And then the tobacco industry literally came bought up all the copies of those films, destroyed them and block that from, you know, from being publicly. No, it wasn't the same as a cause of a relationship to cancer necessarily, but they knew the health effects were absolutely terrible incomparable to what you got by working in a mind. Yeah, so you know, there's an example of total failure of regulatory institution. Absolutely, and the handful of countries that we're able to regulate smoking to a slightly greater degree did not enjoy any great In fact, there's an even an argument, it's a perverse argument, but the tobacco companies didn't make it at one point that it was good for the economy that people smoked because you didn't spend a lot of money on their latent life healthcare because they died of lung cancer. So it's not always the case that really bad social effects right impoverish a country. That's correct. So I was I was just saying that there is a feedback mechanism. Yeah, do we need other things? We probably do, But do we think that we can be a voluntary moratoria do something more powerful in capitalism that I think is a little naive. We need to think of all the ways we can do surveillance, for example, is one of the things I've advocated. Rather than being falsely reassured, let's have strong surveillance. You know, in the topic today of these Cristopher babies, there were plenty of people in the world, both in China and the United States who knew about it, but they weren't reporting it. We need to encourage whistleblowers, and we need to optimize surveillance. I mean this one form of surveillance. The other form of surveillance is computer surveillance. So we should have computers looking through all the orders that are relevant to synthetic biology broadly, and maybe other technologies as well, to see if people are doing things for which they do not have a license. So the privacy concerns interview are just outweighed by the tremendous dangers associate with lack of So I think that everybody who practices synthetic biology and it's close relatives has given up some of their privacy rights in the same sense that when you get a driver's license, you give up certain rights. There are radars that are looking to see what you're doing. You can get pulled over if you're weaving around and you can get an alcohol test, and you can get they can check your age and so forth. You have to have surveillance, and we do have more resistance typically those forms of surveillance in the US than they do, for example, in Europe. Here we think, you know, it would be a terrible violation of our civil liberties if when we got onto the highway they check the time, and when we got off the highway they check the time and saw whether we were speeding in between. But in you know, if you did it on the mass Pike, you know there would be a riot here in Massachusetts. But in Europe there are speed cameras everywhere. The fines are enormous. You know, you are meant to feel surveilled when you drive, right that you do feel surveiled when you drive there, And that's pro social, that's that's that's desirable, and that's why I think when we create boogeymen outside the United States, were completely ignoring how we have a huge fraction of the world's billionaires, many of whom are very pro libertarian. I can do whatever I want, or I think it's much more likely that the rogues are going to be us. George, thank you so much for your fear time, and thanks for sharing your ideas. I'm not sure I emerged less worried than when I started, but you've told me that worries healthy, so I guess I've got plenty of it. Thank you very much, thank you, thank you. I asked George Church to come on the show because I wanted to get a better sense than I had about just how worried we should be about the potential advent of designer babies. When George first started talking, I started to feel better. I began to think that maybe the FDA was a reasonable institution for regulating the possible bad effects of designer babies. And George was also reassuring about the thought that the Chinese government also is interested in blocking progress from going too quickly or getting out of hand. But as our conversation proceeded, I started to get more and more were nervous. In particular, George was pretty convincing in suggesting that those scientists who say we don't have to worry about gene editing leading to designer babies soon are actually completely wrong about that. Then, as he began to describe the potential effects of companies and capitalism on the development of science. I began to think that maybe they were going to be economic pressures driving us in the direction of greater and greater innovation and greater and greater risk taking. And finally, when we circled all the way back to those same regulatory agencies, George seemed to think that there was some possibility that in the long run we would make mistakes, but that we would be self correcting. And there he and I parted paths because I'm deeply afraid that we're not that good at fixing our mistakes, that there are too many examples of our regulatory regimes breaking down. And we ended with George saying that he thinks that if anyone's going to go rogue, it's going to be us right here in the United States, and that, my friends, leaves me more nervous than anything else. Deep Background is brought to you by Pushkin Industries. Our producer is Lydia gene Coott, with engineering by Jason Gambrel and Jason Roskowski. Our showrunner is Sophie mckibbon. Our theme music is composed by Luis GERA special thanks to the Pushkin Brass Malcolm Gladwell, Jacob Weisberg, and Mia Lobel. I'm Noah Feldman. You can follow me on Twitter at Noah R. Feldman. This is deep background

Pushkin from Pushkin Industries. This is deep background the show where we explored the stories behind the stories in the news. I'm Noah Feldman. Welcome to this week's program, where we're going to talk about gene editing, designer babies, and the future of science. These are all topics of importance all the time, but they've become more pressing since the announcement that a Chinese scientist had actually used Crisper Cast nine, the leading cutting edge editing technology, to edit the genomes of two embryos to assure that they would not be able to be susceptible to the HIV virus. This was good for the embryos, but it wasn't necessary scientifically, and it's led to lots of intense discussion about whether the era of designer babies was too close and needed to be headed off by regulation. To discuss these and closely related issues were super fortunate to have with us George Church. George is the Robert Winthrop Professor of Genetics at Harvard Medical School. He also teaches at MIT. He's been a central actor in the development of the technologies of gene editing and in applying them to the creation of new genomes. He sometimes, in fact calls himself a genome engineer. George, thank you so much for joining us. It's a great pleasure to be here. Noah, let's just start with the headline in order to make sense of what it really means and whether there are dangerous associated with it of the kind that many imagine. We read in the paper that Jan Qui, a Chinese scientist, says, or has it said about him, that he used Crisper cast nine to engineer the DNA of a couple of babies, and that they were then subsequently carried to term. That's what appears to be new here, and he's been The Chinese government's not very happy with him, and he has been under house arrest and radio silence more or less since. From the standpoint of the state of crisper technology, was there anything remarkable about the accomplishment, assuming it was accomplished, or is it relatively a trivial step compared to the scientific advances that had already been in place? Yeah, I would think similar things had been done by other groups, including groups in the United States and Oregon on human embryos, and even more amazing things had been done on other mammalian embryos, including altering dozens of genes. So is here the claim was only the altering of onengenee. You know, there's some interesting questions about choice of genes. You know, one of the things that's challenging in this field is picking a gene that even a few other people would agree is a good choice of gene, because you know, a fair number of genes could be fixed by other methods, Like you have to be in in vitro fertilization or IVS clinic to do this at all, and if you're in that clinic, you could fix it by just IVF prenatal genetic testing EGT. So when it goes to question of choosing the target of the intervention, what are the right criteria or what do you see as the main criteria people would consider. I mean, you've mentioned now one could you get there by some other means? Could you get there without without changing the germ line? But what are other criteria that you think would be relevant here? So in addition to could you get thereout altering the germ line, there's also the criteria of is it a net positive for that person or in a public health sinse on average on that positive and environmentally dependent. So if you're in an environment where everybody gets HIV and dies young, and that's a very different environment than or nobody gets HIV and China is not ground zero for the highest level of HIV in the world, who is highly stigmatized there, but you know places in Africa have higher incidence of it and it would be better candidates. Another one is, you know, is costs is a consideration if you want something that's going to be of true public health benefit for a disease like this impacts a lot of impoverished individuals because you have the alternative is drugs or safe sex. There there are a few alternatives, none of what you're working perfectly worldwide because a million people die for years, it's about two percent of all human deaths. So to claim that this is a solved problem is just as Naives saying that this is the only way to solve the problem. Right, it's not a self problem in real world descriptive terms, even though in principle it could be solvable. So what about the criterion that asks whether something is presently an immediate threat to either the embryo or the local population as opposed to offering some background improvement for the potential person who might come to be created, or for the environment more generally, does that seem like a meaningful difference to you? And I just to show that I'm why I'm asking that question. It goes to the broader societal fear about what are sometimes called designer babies, you know, babies who are who might have their genomes edited to give them some advantage, which could be a health advantage to begin with, but could also be other kinds of advantages, whether in esthetics or intelligence, or athletics or what have you. I think this is something that is very interesting entanglement of concerns. So the concerns are that we will create a monoculture number one, where everyone is the same, the way that you might have a you know, many square kilometers of identical crops. The second is that it will be inequably distributed, in no words, that some people have some people don't. That's a very different concern than you don't want anybody to have it, as you want everybody to have it, okay, you know. A third one is you could make a mistake that it could be a very popular fad, but it was a mistake that has long term consequences, either for the people who had it or the people who didn't get it. There could be stigmatization issues direction. Yeah, it could beat not to have had the intention or to have absolutely and there are a few more things that get tangled up here. Sometimes it's phrased as blondere blue eye, which doesn't make any sense at all. You know, that's not a public health threat. It's unlikely to have long term consequences even if it were a monoculture. It's not a monoculture that threatens society if it's not. But it's not an accident that people use that as the example. What they're invoking when they say that is there associating genetic engineering with the eugenics movement of the eighteen eighties, nineties or early twentieth century, which was not only by any stretch of imagination, popular in Germany but all over the world, including very much the US. Yeah, the United States kept doing it long after World War Two. I think it was into early nineteen seventies, and so eugenics though this wasn't the only line of eugenic thinking at all, but it married itself up at least in the European context, even you know now long discredited racialized theories. That's what people mean, I think when they say absolutely absolutely. And furthermore, there's all kinds of body image issues, shaming and so forth the en conjures up, which some of which are racially independent. You know that within a race you'll have you know, obesity shaming and many other features that determine your status of society, your socio economic status, enhance your health status. So even though it's not health explicitly, that has ramifications in that direction. So some of those can be addressed. So you know, equitable distribution, we could bring down the price as we have for various technologies like smallpox. Vaccine has made it extinct and so it's basically free now. So there are technologies that come down pretty quickly to zero, so that you can take some of these things at least not necessarily off the table, but you can put them aside in a separate category. But another one is that you could create It falls a bit in the equitable distribution, but you could create people that it's not necessarily racial. It's just a new capability right now, most new capabilities. I believe that the human population at present doesn't have right, and that enhancement is something that we should feel very familiar with because we are enormously enhanced relatives to our enhancestors. We're not actually fearful of enhancement as far as I can tell. But that do you mean incrementally enhanced? I mean so if you look at I don't know the fastest one hundred meter dash times, they're getting faster, though presumably they're also going to a mutual limit. But I can beat one hundred meter dash person any day in my jet. But that's a capacity enhancement that is not embodied. We have developed tools that enable us to do things, but they're not quite the same as being incorporated. It's an interesting question why we treat them differently because they are hurtable in a certain sense culturally culturally inheritable. But culture is an inheritance. I consider many ways more threatening and more rapid than DNA obsessed inheritance. You know, it's much more likely that my daughter will have my cell phone then she will have my facial features. Thankfully, so we get hung up on this sort of DNA obsession. If they actually knew less science. Oddly, we might draw less of a line between these, Right, we didn't know how we inherited these things. We didn't really understand culture and technology and DNA they would kind of look similar. Oh then it's your fault for helping us understand it, so well. Sorry, Well, you're making a really interesting observation that culture has a wide range of effects. It's in some broad sense heritable. But I want to add that it's also subject to the full range nearly every one of the serious risks that you described with respect to genetic editing. In fact, in many cases, the way we come up with our fear is we look at the distortions or the injustices produced by culture and we say, well, oh boy, you know, genetic editing might have the same effect, so you know, societal inequity. Check on that. Right. It can also these target effects which we haven't yet gotten to. But you know, that's broadly speaking, the idea that if we edit one thing, there will be unforeseen effects somewhere else in the genome. Check on that. In culture, all the time, we improve one thing and we make other things much worse, so certain consequences. Check on that. Go ahead, Yeah, I mean the early adopters sometimes they're more exposed to the unintended consequences the off targets, and so they're essentially be signing up with their extra dollars to become the first guinea pigs. So it seems like there's some meaningful debate within the scientific community about how concerned we should be that particular desirable genetic edits. This is assume it's we're avoiding a disease, that there's no other way to avoid it, that we fit your other criteria. There seems to be some disagreement between one campus scientists who say, because we don't know for sure what the off target effects might be of a given intervention, we should proceed extremely slowly and carefully, and another campus scientists, again this is a continuum, who seem to say, we can measure off target effects like we can measure anything else, and if we reach the point that off target effects are less likely to occur from a genetic edit than they are in nature, then it's absurd to be so worried about it. And I'm fascinated by this question because as a layman, I have no idea how to go about answering it right. Well, the probably the most glib answer to it is this is the responsibility of food and drug administration in the United States and equivalence in other countries. And it happens with every single technology to something expose. Certainly the medical technologies were the medical devices, or molecule drugs or protein drugs. Every category has all target effects which are physiological, so you need to do as much as you can theoretically, followed by as much as you can with animal models or human cells in culture, and then you move on very cautlessly to one person to do phase one toxicity and then efficacy, and then make sure that it's really ready for primetime, and you scale up the size of the cohort cautlessly so you expose the minimum number of people to risk for good theoretical reasons. So this combination of theory and testing cautiously is what protects us from all our new technologies. Is the timescale particularly challenging in the case of genetic editing because you're editing an embryo, but it might be that the off target effect doesn't express itself, you know, typically until much later in life, so that it might take much longer to find out what potential left target effects are. Absolutely you know, preventive medicine is a nice buzzword, but it's very hard to develop powerful preventive medicines because you're testing them on people that are healthy. And it's not restricted to embryo editing, its fetal surgery with actual microscalpels of things and monus sickness drugs like Blood of Mind, which which failed disastrous effects. Extreme caution is required when you're dealing with prevented in medicine. For example, you take chemotherapy as an adult woman, you could be affecting your German line in ways that you also won't see until your babies are born, and maybe their babies are born. Right used sub interesting words now, which I think are maybe they are characteristic give you, although I don't think of them as characteristic of you, namely the words extreme caution. So I want to ask you about your headge about the FDA. Of course, institutionally we assign those decisions in our democracy to the FDA, but that doesn't necessarily tell us what criteria the FDA ought to apply, or, more to the point, how cautious the FDA should be, or how risk taking the FDA should be. I mean, real human beings. Ideally, scientists and statisticians sitting in the FDA have to make these decisions on the basis of what data is known. So I guess what I wanted to ask you is, if I'm right that there is this continuum of how much risk we should take in the scientific community around gene editing, where do you fall on that continuum? The words extreme cautions sort of put you sound like they put you among the Really, let's not rush this side. I think tend to fall on the I worry about everything, and I want everybody else. I don't want to reassure people necessarily, I'm not in a big rush for most things, even aging reversal, even though I'm I'm not in a big rush show, so don't worry about it. I'm not in a big rush. But that said, I I don't fall into the camp of we should put as many barriers in the way as possible so that it never happens, or be so vague about it that we can never be satisfied as to what the criteria are for letting it go forward. And it is a continuum, and I think the FDA does an admirable job of prioritizing so if you have a very serious disease is going to kill you tomorrow, they have a different threshold. Then if you have a healthy baby who has every expectation of being healthy, and you're going to give it something that might extend its life by ten years eight years from now, that barrier is very hard to get permission to do that study. And rightly so, there's a line of thought that says, using the FDA as one pole of you describe pretty admirable cost benefit weighing, and then on the other side, looking at China, where at least in one instance, there seems to have been a lot of risk taking, it says, look, it's all well and good to say that the FDA should be cautious, but out there in the world, outside the reach of more cautious and maybe typically democratic although that doesn't have to follow governments, we're going to get lots of innovation, lots of risk taking, and that we'll fall behind if we listen to the FDA or we allow the FDA to be cautious, or alternatively, that says we ought to be very aggressively going out and trying to export our own limitations by pressuring foreign governments to make sure that they crack down to a greater extent. How do you think about that phenomenon? The science is increasingly globalized, as prices come down, as scientists train across borders. There's no in principal reason that a country with a reasonable infrastructure scientific infrastructure can't make all kinds of innovations under a very different regular environment. Well, and first of all, I would say that the Chinese FDA to c FDA is very similar. I think they have a similar risk profile. I think their government whether you want to call it capital hyper capitalistic, or less than perfectly democratic, it doesn't really matter. The point is they are capable and often do crackdown better than we do. So in the United States it's very hard to interfere with the industry, not because they run Congress or anything, but because there's an obsession with freedom. And in the Chinese government they can crack down on things that are unsafe in ways that are very difficult to do otherwise. So for example, cleaning up Pijing for the Olympics, and that's something that I don't think we could have done. And they're very technocratic too. I meant I'd heard that a huge faction, maybe eighty percent of their top politicians have degrees in science for engineering, and not because they're failed scientists or engineers, but because as politicians they went back and got those degrees because I felt it was important for understanding the issues. That is so far from where we are. We're more in a culture of if I need some facts, I'll give them to you, right. You know, this is a question that I've been very curious to ask you today. To be a cutting edge researcher as you are, it's also perhaps expect it is too strong a word, but not surprising when one also holds lots of patents and starts companies that develop those patents or that try to develop those patterns. So the entrepreneurial side seems to coexist very naturally, easily, almost normatively with being a leading scientist. How do you think about that relationship, because that's been an important part of your practice. So if you have an invention where you've gone from the vast cloud of scientific discovery to something that you think might be useful, it's not sufficient to write up academic ivory tower paper and hope for the best because nobody's going to use it. You know, sometimes there's some concern that capitalism in that form will infect us, you know, that affects us by manipulating us, or as market scientists entrepreneur cannot be as pure as the scientist who's received grants and does basic research. I think a lot of people fear that. Yeah, and they fear that even after the sign us steps away and goes back the Ivory Tower, a company that has been created as a life of its own, and it will be motivated to use marketing and advertising and operative lobbying to give us our opinion, run listen to us. And I think these are all valid things. On the other hand, if you're talking about companies going rogue, there's certain challenges to that, meaning that the CEO reports the bord of directors and to the stockholders in general, and so to do something blatantly irresponsible result in a huge public relations problem, your stots value and rash and so forth. While an academic with tenure who has acted responsible, they can go off and do whatever they want. They don't really report to anybody. So there's a danger in other words, also in having the pure scientists who can do whatever he wants or she wants, and it's not responsible at all. So let's take this then back to the question of the future of Crisper or other similar editing technologies in terms of the state of the science. Now, how credible is it that existing editing technologies could, if they were allowed to by regulators, actually make interventions at population level in a significant way. Are there actually traits enough traits that matter that are addressable by editing one or two genes or a handful of genes to actually make it make an impact. How much of this is science fiction fantasy and how much of this is within the reach of reality. So there's an interesting phenomenon. I would say a huge fraction of my colleagues would reassure you that human genetics is so complex that we are so far away we don't need to worry about it. I think that is false to some of them. I think that is false reassurance. So I think that we have sort of the textbook cases of complex genetics or things like height. Right, thousands of genes involved, possibly all the genes are involved, and they each have a very tiny effect along with hundreds of environmental effects. You know, why is it over the generations we've gotten taller and tallerance better nutritions. So this seems hard to predict, much less to manipulate. And when your colleagues just to flush out the argument for listeners when they say nothing to worry about, they often do use height and they say, well, look, if we know right now that there are four or five hundred genes involved in just producing a time percentage of the variants right in thousands to get you to the whole variety, we're never, or at least not in the foreseeable future, are going to be able to edit all of those different locations. So therefore, you know, don't worry your pretty little head about it, or that's what they say to me. Sore. There's three things wrong with that. One is, there are single genes that are so impactful that even though in the natural population they hardly add anything to it, in the medical situation, they have a huge impact, in particular for height. There's about seven different medical situations where physicians and patients want to do something about height. It isn't necessarily because they're genetically defective, and it's some other reason, And the answer is one gene. One gene product was somatotropin or sometimes called growth hormone. To me, that just really nails the counter arguments. So the counter argument is the geneticists who don't want you to worry, say there's so many different genes they're involved, and you naturally right, And you say, well, that might be true naturally, but that doesn't mean there isn't an intervention we could make exactly on a particular gene that would actually have the effect of overshadowing overshadowing the natural variations. So, in other words, and it's a mistake to reason from natural variation to the capacities of technological intervention exactly. And this is not an argument, and this is not hypothetical, because there are seven different so called diseases or medical treatments that are approved and are in routine use to involve this one genes a metatropa. Second, they'll say, oh, we'll have unintended consequences. Well, the fact is in medicine we always have unintended consequences. We almost always deal with them. We just say the benefits outweigh the risks. So that's an ethical argument that they make. That The first argument is a don't worry because it's not going to be practical to engage in these kinds of systematic edits. This argument set is the is the ethical concern that says, we don't know there might be off target effects, and you're saying, there always are, and we were accustomed to that, and we we balanced the costs in the benefit. The distinction is blurred between practical and ethical very often because you'll say it's impractical because there's these off targets. The thing is it is totally practical because almost everything has off targets and you maybe take a second drug to deal with it, or you stratify the population, or there's all sorts of things. There's the tricks you can use, so don't they might be target effects. When you hear those two arguments, these are the grains of salt, you should take it with, okay, because they're pretty big grains. And the third one is the technology is so far from standing still. I mean, it is growing exponentially that that even if you say we can't do it today, it could be you know, somewhere between minutes in years, not centuries. They often use the word centuries. It's incredible, there's no centuries anymore. So. So for example, my blab's personal record, a pretty much world record for editing genes free Crisper or around the time of Christophers two, we could addit two genes, and we were very proud of ourselves. We edited more than one. Just just literally two years later, we had edited sixty two genes simultaneously to make a pig that didn't have any viruses and retroviruses in its you know. Just a year or two after that, we now have edited thirteen thousand genes simultaneously in a mammal in human human cells. Okay, not a human important to think. So the point is this is moving very quickly. What's great is we're having these conversations well in advance. But what used to be well in advance. We started talking about designer babies in the early days of our comment at DNA and in vitro fertilizations sort of in the early seventies. That was well in advance. But the conversations we're having now are not forty years in advance. They're more like a decade. When I look at the overall picture that you describe, you know, the extraordinary change, the speed of change. I wonder whether the institutions that we've talked about before the FDA or the Chinese FDA are really strong enough and capable enough to resist the temptations they're going to come with the tremendous technological capacities that are being developed. I mean, if we look at our history, we don't have a great history of using institutions, regulatory institutions to block the cutting edge. The cutting edge usually tends to emerge. And that's partly, I think because of the power of capital, you know, which you've mentioned here. It's also partly that once something is possible, there will be people advocating for it. If you have a rare disease, you will say, look, it's not that it would be unethical to try to cure this disease. It would be unethical not to try to cure this disease. So can we hold back against the potential risks here? You know, we say, now, well, of course we would never do this to give people purely esthetic advantages, but in reality maybe we would. It is true that capitalism can distort the FDA, and it's ilk around the world, but I think there is a there is still feedback mechanism that's bigger than all of this, which is that if it really doesn't work or isn't cost effective or desirable, we will stop using it. In fact, if it doesn't work. My worry is more, what if by work, I mean doesn't work for society that the long term benefits society. And they are examples. So first, what do FDAs around the world not do? So they do safety and efficacy, but they don't necessarily do equitable distribution or very long term risk. They have long term risks in mind, but the way they've set up the structure is if you can convince that on a short term risk basis that you've done all the experiments, you get to test in phase four, which is basically selling it, and then once capitalism kicks in, then it's very hard to regulate. So perfect examples of this are nutritional supplements and stem cells. Those managed to get into the marketplace without full FDA approval and now they're bigger than the rest of the pharmaceutical industry and so they're hard to regulate. But bad example will be the opioid crisis. I mean, it was just about gerty. It's a good one. So this is something where if you create something that's sufficiently addictive, people will try it and then they become lobbyists. They vote, They vote with their wallet, which is in many ways more powerful than voting with ballots. That said, there will be a feedback system. If any country manages to get a business model or a country policy model that prevents this, they will win because the nations that can't control their opiated crisis become impoverished, mismanaged, etc. They become the new cash. Can I push back on that? I mean, so everything we've talked about until now, I'm the purest lay person, But now when we talk about feedback and the governments, I have something to say. And here's what I would want to say. We've got lots of examples where societies inflict huge costs on themselves, but the costs are not so great that they turn them into, as you would say, third world countries, because they have other tremendous advantages. So tobacco would be a really good example. Yeah, you know, I remember my grandfather, who was trained in pharmacy school in the nineteen twenties, describing a film that they were all shown of the horrible effects on the human lung of tobacco smoking. And then the tobacco industry literally came bought up all the copies of those films, destroyed them and block that from, you know, from being publicly. No, it wasn't the same as a cause of a relationship to cancer necessarily, but they knew the health effects were absolutely terrible incomparable to what you got by working in a mind. Yeah, so you know, there's an example of total failure of regulatory institution. Absolutely, and the handful of countries that we're able to regulate smoking to a slightly greater degree did not enjoy any great In fact, there's an even an argument, it's a perverse argument, but the tobacco companies didn't make it at one point that it was good for the economy that people smoked because you didn't spend a lot of money on their latent life healthcare because they died of lung cancer. So it's not always the case that really bad social effects right impoverish a country. That's correct. So I was I was just saying that there is a feedback mechanism. Yeah, do we need other things? We probably do, But do we think that we can be a voluntary moratoria do something more powerful in capitalism that I think is a little naive. We need to think of all the ways we can do surveillance, for example, is one of the things I've advocated. Rather than being falsely reassured, let's have strong surveillance. You know, in the topic today of these Cristopher babies, there were plenty of people in the world, both in China and the United States who knew about it, but they weren't reporting it. We need to encourage whistleblowers, and we need to optimize surveillance. I mean this one form of surveillance. The other form of surveillance is computer surveillance. So we should have computers looking through all the orders that are relevant to synthetic biology broadly, and maybe other technologies as well, to see if people are doing things for which they do not have a license. So the privacy concerns interview are just outweighed by the tremendous dangers associate with lack of So I think that everybody who practices synthetic biology and it's close relatives has given up some of their privacy rights in the same sense that when you get a driver's license, you give up certain rights. There are radars that are looking to see what you're doing. You can get pulled over if you're weaving around and you can get an alcohol test, and you can get they can check your age and so forth. You have to have surveillance, and we do have more resistance typically those forms of surveillance in the US than they do, for example, in Europe. Here we think, you know, it would be a terrible violation of our civil liberties if when we got onto the highway they check the time, and when we got off the highway they check the time and saw whether we were speeding in between. But in you know, if you did it on the mass Pike, you know there would be a riot here in Massachusetts. But in Europe there are speed cameras everywhere. The fines are enormous. You know, you are meant to feel surveilled when you drive, right that you do feel surveiled when you drive there, And that's pro social, that's that's that's desirable, and that's why I think when we create boogeymen outside the United States, were completely ignoring how we have a huge fraction of the world's billionaires, many of whom are very pro libertarian. I can do whatever I want, or I think it's much more likely that the rogues are going to be us. George, thank you so much for your fear time, and thanks for sharing your ideas. I'm not sure I emerged less worried than when I started, but you've told me that worries healthy, so I guess I've got plenty of it. Thank you very much, thank you, thank you. I asked George Church to come on the show because I wanted to get a better sense than I had about just how worried we should be about the potential advent of designer babies. When George first started talking, I started to feel better. I began to think that maybe the FDA was a reasonable institution for regulating the possible bad effects of designer babies. And George was also reassuring about the thought that the Chinese government also is interested in blocking progress from going too quickly or getting out of hand. But as our conversation proceeded, I started to get more and more were nervous. In particular, George was pretty convincing in suggesting that those scientists who say we don't have to worry about gene editing leading to designer babies soon are actually completely wrong about that. Then, as he began to describe the potential effects of companies and capitalism on the development of science. I began to think that maybe they were going to be economic pressures driving us in the direction of greater and greater innovation and greater and greater risk taking. And finally, when we circled all the way back to those same regulatory agencies, George seemed to think that there was some possibility that in the long run we would make mistakes, but that we would be self correcting. And there he and I parted paths because I'm deeply afraid that we're not that good at fixing our mistakes, that there are too many examples of our regulatory regimes breaking down. And we ended with George saying that he thinks that if anyone's going to go rogue, it's going to be us right here in the United States, and that, my friends, leaves me more nervous than anything else. Deep Background is brought to you by Pushkin Industries. Our producer is Lydia gene Coott, with engineering by Jason Gambrel and Jason Roskowski. Our showrunner is Sophie mckibbon. Our theme music is composed by Luis GERA special thanks to the Pushkin Brass Malcolm Gladwell, Jacob Weisberg, and Mia Lobel. I'm Noah Feldman. You can follow me on Twitter at Noah R. Feldman. This is deep background