Podcast: Martin Rees on the Prospects for Humanity: AI, Biotech, Climate Change, Overpopulation, Cryogenics, and More

How can humanity survive the next century of climate change, a growing population, and emerging technological threats? Where do we stand now, and what steps can we take to cooperate and address our greatest existential risks?

In this special podcast episode, Ariel speaks with Martin Rees about his new book, On the Future: Prospects for Humanity, which discusses humanity’s existential risks and the role that technology plays in determining our collective future. Martin is a cosmologist and space scientist based in the University of Cambridge. He is director of The Institute of Astronomy and Master of Trinity College, and he was president of The Royal Society, which is the UK’s Academy of Science, from 2005 to 2010. In 2005 he was also appointed to the UK’s House of Lords.

Topics discussed in this episode include:

  • Why Martin remains a technical optimist even as he focuses on existential risks
  • The economics and ethics of climate change
  • How AI and automation will make it harder for Africa and the Middle East to economically develop
  • How high expectations for health care and quality of life also put society at risk
  • Why growing inequality could be our most underappreciated global risk
  • Martin’s view that biotechnology poses greater risk than AI
  • Earth’s carrying capacity and the dangers of overpopulation
  • Space travel and why Martin is skeptical of Elon Musk’s plan to colonize Mars
  • The ethics of artificial meat, life extension, and cryogenics
  • How intelligent life could expand into the galaxy
  • Why humans might be unable to answer fundamental questions about the universe

Books and resources discussed in this episode include

You can listen to the podcast above and read the full transcript below. Check out our previous podcast episodes on SoundCloudiTunesGooglePlay, and Stitcher.

Ariel: Hello, I am Ariel Conn with The Future of Life Institute. Now, our podcasts lately have dealt with artificial intelligence in some way or another, and with a few focusing on nuclear weapons, but FLI is really an organization about existential risks, and especially x-risks that are the result of human action. These cover a much broader field than just artificial intelligence.

I’m excited to be hosting a special segment of the FLI podcast with Martin Rees, who has just come out with a book that looks at the ways technology and science could impact our future both for good and bad. Martin is a cosmologist and space scientist. His research interests include galaxy formation, active galactic nuclei, black holes, gamma ray bursts, and more speculative aspects of cosmology. He’s based in Cambridge where he has been director of The Institute of Astronomy, and Master of Trinity College. He was president of The Royal Society, which is the UK’s Academy of Science, from 2005 to 2010. In 2005 he was also appointed to the UK’s House of Lords. He holds the honorary title of Astronomer Royal. He has received many international awards for his research and belongs to numerous academies, including The National Academy of Sciences, the Russian Academy, the Japan Academy, and the Pontifical Academy.

He’s on the board of The Princeton Institute for Advanced Study, and has served on many bodies connected with international collaboration and science, especially threats stemming from humanity’s ever heavier footprint on the planet and the runaway consequences of ever more powerful technologies. He’s written seven books for the general public, and his most recent book is about these threats. It’s the reason that I’ve asked him to join us today. First, Martin thank you so much for talking with me today.

Martin: Good to be in touch.

Ariel: Your new book is called On the Future: Prospects for Humanity. In his endorsement of the book Neil deGrasse Tyson says, “From climate change, to biotech, to artificial intelligence, science sits at the center of nearly all decisions that civilization confronts to assure its own survival.”

I really liked this quote, because I felt like it sums up what your book is about. Basically science and the future are too intertwined to really look at one without the other. And whether the future turns out well, or whether it turns out to be the destruction of humanity, science and technology will likely have had some role to play. First, do you agree with that sentiment? Am I accurate in that description?

Martin: No, I certainly agree, and that’s truer of this century than ever before because of greater scientific knowledge we have, and the greater power to use it for good or ill, because the technologies allow tremendously advanced technologies which could be misused by a small number of people.

Ariel: You’ve written in the past about how you think we have essentially a 50/50 chance of some sort of existential risk. One of the things that I noticed about this most recent book is you talk a lot about the threats, but to me it felt still like an optimistic book. I was wondering if you could talk a little bit about, this might be jumping ahead a bit, but maybe what the overall message you’re hoping that people take away is?

Martin: Well, I describe myself as a technical optimist, but political pessimist because it is clear that we couldn’t be living such good lives today with seven and a half billion people on the planet if we didn’t have the technology which has been developed in the last 100 years, and clearly there’s a tremendous prospect of better technology in the future. But on the other hand what is depressing is the very big gap between the way the world could be, and the way the world actually is. In particular, even though we have the power to give everyone a decent life, the lot of the bottom billion people in the world is pretty miserable and could be alleviated a lot simply by the money owned by the 1,000 richest people in the world.

We have a very unjust society, and the politics is not optimizing the way technology is used for human benefit. My view is that it’s the politics which is an impediment to the best use of technology, and the reason this is important is that as time goes on we’re going to have a growing population which is ever more demanding of energy and resources, putting more pressure on the planet and its environment and its climate, but we are also going to have to deal with this if we are to allow people to survive and avoid some serious tipping points being crossed.

That’s the problem of the collective effect of us on the planet, but there’s another effect, which is that these new technologies, especially bio, cyber, and AI allow small groups of even individuals to have an effect by error or by design, which could cascade very broadly, even globally. This, I think, makes our society very brittle. We’re very interdependent, and on the other hand it’s easy for there to be a breakdown. That’s what depresses me, the gap between the way things could be, and the downsides if we collectively overreach ourselves, or if individuals cause disruption.

Ariel: You mentioned actually quite a few things that I’m hoping to touch on as we continue to talk. I’m almost inclined, before we get too far into some of the specific topics, to bring up an issue that I personally have. It’s connected to a comment that you make in the book. I think you were talking about climate change at the time, and you say that if we heard that there was 10% chance that an asteroid would strike in 2100 people would do something about it.

We wouldn’t say, “Oh, technology will be better in the future so let’s not worry about it now.” Apparently I’m very cynical, because I think that’s exactly what we would do. And I’m curious, what makes you feel more hopeful that even with something really specific like that, we would actually do something and not just constantly postpone the problem to some future generation?

Martin: Well, I agree. We might not even in that case, but the reason I gave that as a contrast to our response to climate change is that there you could imagine a really sudden catastrophe happening if the asteroid does hit, whereas the problem with climate change is really that it’s first of all, the effect is mainly going to be several decades in the future. It’s started to happen, but the really severe consequences are decades away. But also there’s an uncertainty, and it’s not a sort of sudden event we can easily visualize. It’s not at all clear therefore, how we are actually going to do something about it.

In the case of the asteroid, it would be clear what the strategy would be to try and deal with it, whereas in the case of climate there are lots of ways, and the problem is that the consequences are decades away, and they’re global. Most of the political focus obviously is on short-term worry, short-term problems, and on national or more local problems. Anything we do about climate change will have an effect which is mainly for the benefit of people in quite different parts of the world 50 years from now, and it’s hard to keep those issues up the agenda when there are so many urgent things to worry about.

I think you’re maybe right that even if there was a threat of an asteroid, there may be the same sort of torpor, and we’d fail to deal with it, but I thought that’s an example of something where it would be easier to appreciate that it would really be a disaster. In the case of the climate it’s not so obviously going to be a catastrophe that people are motivated now to start thinking about it.

Ariel: I’ve heard it go both ways that either climate change is yes, obviously going to be bad but it’s not an existential risk so therefore those of us who are worried about existential risk don’t need to worry about it, but then I’ve also heard people say, “No, this could absolutely be an existential risk if we don’t prevent runaway climate change.” I was wondering if you could talk a bit about what worries you most regarding climate.

Martin: First of all, I don’t think it is an existential risk, but it’s something we should worry about. One point I make in my book is that I think the debate, which makes it hard to have an agreed policy on climate change, stems not so much from differences about the science — although of course there are some complete deniers — but differences about ethics and economics. There’s some people of course who completely deny the science, but most people accept that CO2 is warming the planet, and most people accept there’s quite a big uncertainty, matter of fact a true uncertainty about how much warmer you get for a given increase in CO2.

But even among those who accept the IPCC projections of climate change, and the uncertainties therein, I think there’s a big debate, and the debate is really between people who apply a standard economic discount rate where you discount the future to a rate of, say 5%, and those who think we shouldn’t do it in this context. If you apply a 5% discount rate as you would if you were deciding whether it’s worth putting up an office building or something like that, then of course you don’t give any weight to what happens after about, say 2050.

As Bjorn Lomborg, the well-known environmentalist argues, we should therefore give a lower priority to dealing with climate change than to helping the world’s poor in other more immediate ways. He is consistent given his assumptions about the discount rate. But many of us would say that in this context we should not discount the future so heavily. We should care about the life chances of a baby born today as much as we should care about the life chances of those of us who are now middle aged and won’t be alive at the end of the century. We should also be prepared to pay an insurance premium now in order to remove or reduce the risk of the worst case climate scenarios.

I think the debates about what to do about climate change is essentially ethics. Do we want to discriminate on grounds of date of birth and not care about the life chances of those who are now babies, or are we prepared to make some sacrifices now in order to reduce a risk which they might encounter in later life?

Ariel: Do you think the risks are only going to be showing up that much later? We are already seeing these really heavy storms striking. We’ve got Florence in North Carolina right now. There’s a super typhoon hit southern China and the Philippines. We had Maria, and I’m losing track of all the hurricanes that we’ve had. We’ve had these huge hurricanes over the last couple of years. We saw California and much of the west coast of the US just on flames this year. Do you think we really need to wait that long?

Martin: I think it’s generally agreed that extreme weather is now happening more often as a consequence of climate change and the warming of the ocean, and that this will become a more serious trend, but by the end of the century of course it could be very serious indeed. And the main threat is of course to people in the disadvantaged parts of the world. If you take these recent events, it’s been far worse in the Philippines than in the United States because they’re not prepared for it. Their houses are more fragile, etc.

Ariel: I don’t suppose you have any thoughts on how we get people to care more about others? Because it does seem to be in general that sort of worrying about myself versus worrying about other people. The richer countries are the ones who are causing more of the climate change, and it’s the poorer countries who seem to be suffering more. Then of course there’s the issue of the people who are alive now versus the people in the future.

Martin: That’s right, yes. Well, I think most people do care about their children and grandchildren, and so to that extent they do care about what things will be like at the end of the century, but as you say, the extra-political problem is that the cause of the CO2 emissions is mainly what’s happened in the advanced countries, and the downside is going to be more seriously felt by those in remote parts of the world. It’s easy to overlook them, and hard to persuade people that we ought to make a sacrifice which will be mainly for their benefit.

I think incidentally that’s one of the other things that we have to ensure happens, is a narrowing of the gap between the lifestyles and the economic advantages in the advanced and the less advanced parts of the world. I think that’s going to be in everyone’s interest because if there continues to be great inequality, not only will the poorer people be more subject to threats like climate change, but I think there’s going to be massive and well-justified discontent, because unlike in the earlier generations, they’re aware of what they’re missing. They all have mobile phones, they all know what it’s like, and I think there’s going to be embitterment leading to conflict if we don’t narrow this gap, and this requires I think a sacrifice on the part of the wealthy nations to subsidize developments in these poorer countries, especially in Africa.

Ariel: That sort of ties into another question that I had for you, and that is, what do you think is the most underappreciated threat that maybe isn’t quite as obvious? You mentioned the fact that we have these people in poorer countries who are able to more easily see what they’re missing out on. Inequality is a problem in and of itself, but also just that people are more aware of the inequality seems like a threat that we might not be as aware of. Are there others that you think are underappreciated?

Martin: Yes. Just to go back, that threat is of course very serious because by the end of the century there might be 10 times as many people in Africa as in Europe, and of course they would then have every justification in migrating towards Europe with the result of huge disruption. We do have to care about those sorts of issues. I think there are all kinds of reasons apart from straight ethics why we should ensure that the less developed countries, especially in Africa, do have a chance to close the gap.

Incidentally, one thing which is a handicap for them is that they won’t have the route to prosperity followed by the so called “Asian tigers,” which were able to have high economic growth by undercutting the labor cost in the west. Now what’s happening is that with robotics it’s possible to, as it were, re-shore lots of manufacturing industry back to wealthy countries, and so Africa and the Middle East won’t have the same opportunity the far eastern countries did to catch up by undercutting the cost of production in the west.

This is another reason why it’s going to be a big challenge. That’s something which I think we don’t worry about enough, and need to worry about, because if the inequalities persist when everyone is able to move easily and knows exactly what they’re missing, then that’s a recipe for a very dangerous and disruptive world. I would say that is an underappreciated threat.

Another thing I would count as important is that we are as a society very brittle, and very unstable because of high expectations. I’d like to give you another example. Suppose there were to be a pandemic, not necessarily a genetically engineered terrorist one, but a natural one. Then in contrast to what happened in the 14th century when the Bubonic Plague, the Black Death, occurred and killed nearly half the people in certain towns and the rest went on fatalistically. If we had some sort of plague which affected even 1% of the population of the United States, there’d be complete social breakdown, because that would overwhelm the capacity of hospitals, and people, unless they are wealthy, would feel they weren’t getting their entitlement of healthcare. And if that was a matter of life and death, that’s a recipe for social breakdown. I think given the high expectations of people in the developed world, then we are far more vulnerable to the consequences of these breakdowns, and pandemics, and the failures of electricity grids, et cetera, than in the past when people were more robust and more fatalistic.

Ariel: That’s really interesting. Is it essentially because we expect to be leading these better lifestyles, just that expectation could be our downfall if something goes wrong?

Martin: That’s right. And of course, if we know that there are cures available to some disease and there’s not the hospital capacity to offer it to all the people who are afflicted with the disease, then naturally that’s a matter of life and death, and that is going to promote social breakdown. This is a new threat which is of course a downside of the fact that we can at least cure some people.

Ariel: There’s two directions that I want to go with this. I’m going to start with just transitioning now to biotechnology. I want to come back to issues of overpopulation and improving healthcare in a little bit, but first I want to touch on biotech threats.

One of the things that’s been a little bit interesting for me is that when I first started at FLI three years ago we were very concerned about biotechnology. CRISPR was really big. It had just sort of exploded onto the scene. Now, three years later I’m not hearing quite as much about the biotech threats, and I’m not sure if that’s because something has actually changed, or if it’s just because at FLI I’ve become more focused on AI and therefore stuff is happening but I’m not keeping up with it. I was wondering if you could talk a bit about what some of the risks you see today are with respect to biotech?

Martin: Well, let me say I think we should worry far more about bio threats than about AI in my opinion. I think as far as the bio threats are concerned, then there are these new techniques. CRISPR, of course, is a very benign technique if it’s used to remove a single damaging gene that gives you a particular disease, and also it’s less objectionable than traditional GM because it doesn’t cross the species barrier in the same way, but it does allow things like a gene drive where you make a species extinct by making it sterile.

That’s good if you’re wiping out a mosquito that carries a deadly virus, but there’s a risk of some effect which distorts the ecology and has a cascading consequence. There are risks of that kind, but more important I think there is a risk of the misuse of these techniques, and not just CRISPR, but for instance the the gain of function techniques that we used in 2011 in Wisconsin and in Holland to make influenza virus both more virulent and more transmissible, things like that which can be done in a more advanced way now I’m sure.

These are clearly potentially dangerous, even if experimenters have a good motive, then the viruses might escape, and of course they are the kinds of things which could be misused. There have, of course, been lots of meetings, you have been at some, to discuss among scientists what the guidelines should be. How can we ensure responsible innovation in these technologies? These are modeled on the famous Conference in Asilomar in the 1970s when recombinant DNA was first being discussed, and the academics who worked in that area, they agreed on a sort of cautious stance, and a moratorium on some kinds of experiments.

But now they’re trying to do the same thing, and there’s a big difference. One is that these scientists are now more global. It’s not just a few people in North America and Europe. They’re global, and there is strong commercial pressures, and they’re far more widely understood. Bio-hacking is almost a student recreation. This means, in my view, that there’s a big danger, because even if we have regulations about certain things that can’t be done because they’re dangerous, enforcing those regulations globally is going to be as hopeless as it is now to enforce the drug laws, or to enforce the tax laws globally. Something which can be done will be done by someone somewhere, whatever the regulations say, and I think this is very scary. Consequences could cascade globally.

Ariel: Do you think that the threat is more likely to come from something happening accidentally, or intentionally?

Martin: I don’t know. I think it could be either. Certainly it could be something accidental from gene drive, or releasing some dangerous virus, but I think if we can imagine it happening intentionally, then we’ve got to ask what sort of people might do it? Governments don’t use biological weapons because you can’t predict how they will spread and who they’d actually kill, and that would be an inhibiting factor for any terrorist group that had well-defined aims.

But my worst nightmare is some person, and there are some, who think that there are too many human beings on the planet, and if they combine that view with the mindset of extreme animal rights people, etc, they might think it would be a good thing for Gaia, for Mother Earth, to get rid of a lot of human beings. They’re the kind of people who, with access to this technology, might have no compunction in releasing a dangerous pathogen. This is the kind of thing that worries me.

Ariel: I find that interesting because it ties into the other question that I wanted to ask you about, and that is the idea of overpopulation. I’ve read it both ways, that overpopulation is in and of itself something of an existential risk, or a catastrophic risk, because we just don’t have enough resources on the planet. You actually made an interesting point, I thought, in your book where you point out that we’ve been thinking that there aren’t enough resources for a long time, and yet we keep getting more people and we still have plenty of resources. I thought that was sort of interesting and reassuring.

But I do think at some point that does become an issue. At then at the same time we’re seeing this huge push, understandably, for improved healthcare, and expanding life spans, and trying to save as many lives as possible, and making those lives last as long as possible. How do you resolve those two sides of the issue?

Martin: It’s true, of course, as you imply, that the population has risen double in the last 50 years, and there were doomsters who in the 1960s and ’70s thought that mass starvation by now, and there hasn’t been because food production has more than kept pace. If there are famines today, as of course there are, it’s not because of overall food shortages. It’s because of wars, or mal-distribution of money to buy the food. Up until now things have gone fairly well, but clearly there are limits to the food that can be produced on the earth.

All I would say is that we can’t really say what the carrying capacity of the earth is, because it depends so much on the lifestyle of people. As I say in the book, the world couldn’t sustainably have 2 billion people if they all lived like present day Americans, using as much energy, and burning as much fossil fuels, and eating as much beef. On the other hand you could imagine lifestyles which are very sort of austere, where the earth could carry 10, or even 20 billion people. We can’t set an upper limit, but all we can say is that given that it’s fairly clear that the population is going to rise to about 9 billion by 2050, and it may go on rising still more after that, we’ve got to ensure that the way in which the average person lives is less profligate in terms of energy and resources, otherwise there will be problems.

I think we also do what we can to ensure that after 2050 the population turns around and goes down. The base scenario is when it goes on rising as it may if people choose to have large families even when they have the choice. That could happen, and of course as you say, life extension is going to have an affect on society generally, but obviously on the overall population too. I think it would be more benign if the population of 9 billion in 2050 was a peak and it started going down after that.

And it’s not hopeless, because the actual number of births per year has already started going down. The reason the population is still going up is because more babies survive, and most of the people in the developing world are still young, and if they live as long as people in advanced countries do, then of course that’s going to increase the population even for a steady birth rate. That’s why, unless there’s a real disaster, we can’t avoid the population rising to about 9 billion.

But I think policies can have an affect on what happens after that. I think we do have to try to make people realize that having large numbers of children has negative externalities, as it were in economic jargon, and it is going to be something to put extra pressure on the world, and affects our environment in a detrimental way.

Ariel: As I was reading this, especially as I was reading your section about space travel, I want to ask you about your take on whether we can just start sending people to Mars or something like that to address issues of overpopulation. As I was reading your section on that, news came out that Elon Musk and SpaceX had their first passenger for a trip around the moon, which is now scheduled for 2023, and the timing was just entertaining to me, because like I said you have a section in your book about why you don’t actually agree with Elon Musk’s plan for some of this stuff.

Martin: That’s right.

Ariel: I was hoping you could talk a little bit about why you’re not as big a plan of space tourism, and what you think of humanity expanding into the rest of the solar system and universe?

Martin: Well, let me say that I think it’s a dangerous delusion to think we can solve the earth’s problems by escaping to Mars or elsewhere. Mass emigration is not feasible. There’s nowhere in the solar system which is as comfortable to live in as the top of Everest or the South Pole. I think the idea which was promulgated by Elon Musk and Stephen Hawking of mass emigration is, I think, a dangerous delusion. The world’s problems have to be solved here, dealing with climate change is a dawdle compared to terraforming Mars. SoI don’t think that’s true.

Now, two other things about space. The first is that the practical need for sending people into space is getting less as robots get more advanced. Everyone has seen pictures of the Curiosity Probe trundling across the surface of Mars, and maybe missing things that a geologist would notice, but future robots will be able to do much of what a human will do, and to manufacture large structures in space, et cetera, so the practical need to send people to space is going down.

On the other hand, some people may want to go simply as an adventure. It’s not really tourism, because tourism implies it’s safe and routine. It’ll be an adventure like Steve Fossett or the guy who fell supersonically from an altitude balloon. It’d be crazy people like that, and maybe this Japanese tourist is in the same style, who want to have a thrill, and I think we should cheer them on.

I think it would be good to imagine that there are a few people living on Mars, but it’s never going to be as comfortable as our Earth, and we should just cheer on people like this.

And I personally think it should be left to private money. If I was an American, I would not support the NASA space program. It’s very expensive, and it could be undercut by private companies which can afford to take higher risks than NASA could inflict on publicly funded civilians. I don’t think NASA should be doing manned space flight at all. Of course, some people would say, “Well, it’s a national aspiration, a national goal to show superpower pre-eminence by a massive space project.” That was, of course, what drove the Apollo program, and the Apollo program cost about 4% of The US federal budget. Now NASA has .6% or thereabouts. I’m old enough to remember the Apollo moon landings, and of course if you would have asked me back then, I would have expected that there might have been people on Mars within 10 or 15 years at that time.

There would have been, had the program been funded, but of course there was no motive, because the Apollo program was driven by superpower rivalry. And having beaten the Russians, it wasn’t pursued with the same intensity. It could be that the Chinese will, for prestige reasons, want to have a big national space program, and leapfrog what the Americans did by going to Mars. That could happen. Otherwise I think the only manned space flight will, and indeed should, be privately funded by adventurers prepared to go on cut price and very risky missions.

But we should cheer them on. The reason we should cheer them on is that if in fact a few of them do provide some sort of settlement on Mars, then they will be important for life’s long-term future, because whereas we are, as humans, fairly well adapted to the earth, they will be in a place, Mars, or an asteroid, or somewhere, for which they are badly adapted. Therefore they would have every incentive to use all the techniques of genetic modification, and cyber technology to adapt to this hostile environment.

A new species, perhaps quite different from humans, may emerge as progeny of those pioneers within two or three centuries. I think this is quite possible. They, of course, may download themselves to be electronic. We don’t know how it’ll happen. We all know about the possibilities of advanced intelligence in electronic form. But I think this’ll happen on Mars, or in space, and of course if we think about going further and exploring beyond our solar system, then of course that’s not really a human enterprise because of human life times being limited, but it is a goal that would be feasible if you were a near immortal electronic entity. That’s a way in which our remote descendants will perhaps penetrate beyond our solar system.

Ariel: As you’re looking towards these longer term futures, what are you hopeful that we’ll be able to achieve?

Martin: You say we, I think we humans will mainly want to stay on the earth, but I think intelligent life, even if it’s not out there already in space, could spread through the galaxy as a consequence of what happens when a few people who go into space and are away from the regulators adapt themselves to that environment. Of course, one thing which is very important is to be aware of different time scales.

Sometimes you hear people talk about humans watching the death of the sun in five billion years. That’s nonsense, because the timescale for biological evolution by Darwinian selection is about a million years, thousands of times shorter than the lifetime of the sun, but more importantly the time scale for this new kind of intelligent design, when we can redesign humans and make new species, that time scale is a technological time scale. It could be only a century.

It would only take one, or two, or three centuries before we have entities which are very different from human beings if they are created by genetic modification, or downloading to electronic entities. They won’t be normal humans. I think this will happen, and this of course will be a very important stage in the evolution of complexity in our universe, because we will go from the kind of complexity which has emerged by Darwinian selection, to something quite new. This century is very special, which is a century where we might be triggering or jump starting a new kind of technological evolution which could spread from our solar system far beyond, on the timescale very short compared to the time scale for Darwinian evolution and the time scale for astronomical evolution.

Ariel: All right. In the book you spend a lot of time also talking about current physics theories and how those could evolve. You spend a little bit of time talking about multiverses. I was hoping you could talk a little bit about why you think understanding that is important for ensuring this hopefully better future?

Martin: Well, it’s only peripherally linked to it. I put that in the book because I was thinking about, what are the challenges, not just challenges of a practical kind, but intellectual challenges? One point I make is that there are some scientific challenges which we are now confronting which may be beyond human capacity to solve, because there’s no particular reason to think that the capacity of our brains is matched to understanding all aspects of reality any more than a monkey can understand quantum theory.

It’s possible that there be some fundamental aspects of nature that humans will never understand, and they will be a challenge for post-humans. I think those challenges are perhaps more likely to be in the realm of complexity, understanding the brain for instance, than in the context of cosmology, although there are challenges in cosmology which is to understand the very early universe where we may need a new theory like string theory with extra dimensions, et cetera, and we need a theory like that in order to decide whether our big bang was the only one, or whether there were other big bangs and a kind of multiverse.

It’s possible that in 50 years from now we will have such a theory, we’ll know the answers to those questions. But it could be that there is such a theory and it’s just too hard for anyone to actually understand and make predictions from. I think these issues are relevant to the intellectual constraints on humans.

Ariel: Is that something that you think, or hope, that things like more advanced artificial intelligence or however we evolve in the future, that that evolution will allow “us” to understand some of these more complex ideas?

Martin: Well, I think it’s certainly possible that machines could actually, in a sense, create entities based on physics which we can’t understand. This is perfectly possible, because obviously we know they can vastly out-compute us at the moment, so it could very well be, for instance, that there is a variant of string theory which is correct, and it’s just too difficult for any human mathematician to work out. But it could be that computers could work it out, so we get some answers.

But of course, you then come up against a more philosophical question about whether competence implies comprehension, whether a computer with superhuman capabilities is necessarily going to be self-aware and conscious, or whether it is going to be just a zombie. That’s a separate question which may not affect what it can actually do, but I think it does affect how we react to the possibility that the far future will be dominated by such things.

I remember when I wrote an article in a newspaper about these possibilities, the reaction was bimodal. Some people thought, “Isn’t it great there’ll be these even deeper intellects than human beings out there,” but others who thought these might just be zombies thought it was very sad if there was no entity which could actually appreciate the beauties and wonders of nature in the way we can. It does matter, in a sense, to our perception of this far future, if we think that these entities which may be electronic rather than organic, will be conscious and will have the kind of awareness that we have and which makes us wonder at the beauty of the environment in which we’ve emerged. I think that’s a very important question.

Ariel: I want to pull things back to a little bit more shorter term I guess, but still considering this idea of how technology will evolve. You mentioned that you don’t think it’s a good idea to count on going to Mars as a solution to our problems on Earth because all of our problems on Earth are still going to be easier to solve here than it is to populate Mars. I think in general we have this tendency to say, “Oh, well in the future we’ll have technology that can fix whatever issue we’re dealing with now, so we don’t need to worry about it.”

I was wondering if you could sort of comment on that approach. To what extent can we say, “Well, most likely technology will have improved and can help us solve these problems,” and to what extent is that a dangerous approach to take?

Martin: Well, clearly technology has allowed us to live much better, more complex lives than we could in the past, and on the whole the net benefits outweigh the downsides, but of course there are downsides, and they stem from the fact that we have some people who are disruptive, and some people who can’t be trusted. If we had a world where everyone could trust everyone else, we could get rid of about a third of the economy I would guess, but I think the main point is that we are very vulnerable.

We have huge advances, clearly, in networking via the Internet, and computers, et cetera, and we may have the Internet of Things within a decade, but of course people worry that this opens up a new kind of even more catastrophic potential for cyber terrorism. That’s just one example, and ditto for biotech which may allow the development of pathogens which kill people of particular races, or have other effects.

There are these technologies which are developing fast, and they can be used to great benefit, but they can be misused in ways that will provide new kinds of horrors that were not available in the past. It’s by no means obvious which way things will go. Will there be a continued net benefit of technology, as I think we’ve said there as been up ’til now despite nuclear weapons, et cetera, or will at some stage the downside run ahead of the benefits.

I do worry about the latter being a possibility, particularly because of this amplification factor, the fact that it only takes a few people in order to cause disruption that could cascade globally. The world is so interconnected that we can’t really have a disaster in one region without its affecting the whole world. Jared Diamond has this book called Collapse where he discusses five collapses of particular civilizations, whereas other parts of the world were unaffected.

I think if we really had some catastrophe, it would affect the whole world. It wouldn’t just affect parts. That’s something which is a new downside. The stakes are getting higher as technology advances, and my book is really aimed to say that these developments are very exciting, but they pose new challenges, and I think particularly they pose challenges because a few dissidents can cause more trouble, and I think it’ll make the world harder to govern. It’ll make cities and countries harder to govern, and a stronger tension between three things we want to achieve, which is security, privacy, and liberty. I think that’s going to be a challenge for all future governments.

Ariel: Reading your book I very much got the impression that it was essentially a call to action to address these issues that you just mentioned. I was curious: what do you hope that people will do after reading the book, or learning more about these issues in general?

Martin: Well, first of all I hope that people can be persuaded to think long term. I mentioned that religious groups, for instance, tend to think long term, and the papal encyclical in 2015 I think had a very important effect on the opinion in Latin America, Africa, and East Asia in the lead up to the Paris Climate Conference, for instance. That’s an example where someone from outside traditional politics would have an effect.

What’s very important is that politicians will only respond to an issue if it’s prominent in the press, and prominent in their inbox, and so we’ve got to ensure that people are concerned about this. Of course, I ended the book saying, “What are the special responsibilities of scientists,” because scientists clearly have a special responsibility to ensure that their work is safe, and that the public and politicians are made aware of the implications of any discovery they make.

I think that’s important, even though they should be mindful that their expertise doesn’t extend beyond their special area. That’s a reason why scientific understanding, in a general sense, is something which really has to be universal. This is important for education, because if we want to have a proper democracy where debate about these issues rises above the level of tabloid slogans, then given that the important issues that we have to discuss involve health, energy, the environment, climate, et cetera, which have scientific aspects, then everyone has to have enough feel for those aspects to participate in a debate, and also enough feel for probabilities and statistics to be not easily bamboozled by political arguments.

I think an educated population is essential for proper democracy. Obviously that’s a platitude. But the education needs to include, to a greater extent, an understanding of the scope and limits of science and technology. I make this point at the end and hope that it will lead to a greater awareness of these issues, and of course for people in universities, we have a responsibility because we can influence the younger generation. It’s certainly the case that students and people under 30 may be alive towards the end of the century are more mindful of these concerns than the middle aged and old.

It’s very important that these activities like the Effective Altruism movement, 80,000 Hours, and these other movements among students should be encouraged, because they are going to be important in spreading an awareness of long-term concerns. Public opinion can be changed. We can see the change in attitudes to drunk driving and things like that, which have happened over a few decades, and I think perhaps we can have a more environmental sensitivity so to become regarded as sort of rather naff or tacky to waste energy, and to be extravagant in consumption.

I’m hopeful that attitudes will change in a positive way, but I’m concerned simply because the politics is getting very difficult, because with social media, panic and rumor can spread at the speed of light, and small groups can have a global effect. This makes it very, very hard to ensure that we can keep things stable given that only a few people are needed to cause massive disruption. That’s something which is new, and I think is becoming more and more serious.

Ariel: We’ve been talking a lot about things that we should be worrying about. Do you think there are things that we are currently worrying about that we probably can just let go of, that aren’t as big of risks?

Martin: Well, I think we need to ensure responsible innovation in all new technologies. We’ve talked a lot about bio, and we are very concerned about the misuse of cyber technology. As regards AI, of course there are a whole lot of concerns to be had. I personally think that the takeover AI would be rather slower than many of the evangelists suspect, but of course we do have to ensure that humans are not victimized by some algorithm which they can’t have explained to them.

I think there is an awareness to this, and I think that what’s being done by your colleagues at MIT has been very important in raising awareness of the need for responsible innovation and ethical application of AI, and also what your group has recognized is that the order in which things happen is very important. If some computer is developed and goes rogue, that’s bad news, whereas if we have a powerful computer which is under our control, then it may help us to deal with these other problems, the problems of the misuse of biotech, et cetera.

The order in which things happen is going to be very important, but I must say I don’t completely share these concerns about machines running away and taking over, ’cause I think there’s a difference in that, for biological evolution there’s been a drive toward intelligence being favored, but so is aggression. In the case of computers, they may drive towards greater intelligence, but it’s not obvious that that is going to be combined with aggression, because they are going to be evolving by intelligent design, not the struggle of the fittest, which is the way that we evolved.

Ariel: What about concerns regarding AI just in terms of being mis-programmed, and AI just being extremely competent? Poor design on our part, poor intelligent design?

Martin: Well, I think in the short term obviously there are concerns about AI making decisions that affect people, and I think most of us would say that we shouldn’t be deprived of our credit rating, or put in prison on the basis of some AI algorithm which can’t be explained to us. We are entitled to have an explanation if something is done to us against our will. That is why it is worrying if too much is going to be delegated to AI.

I also think that constraint on the development of self-driving cars, and things of that kind, is going to be constrained by the fact that these become vulnerable to hacking of various kinds. I think it’ll be a long time before we will accept a driverless car on an ordinary road. Controlled environments, yes. In particular lanes on highways, yes. In an ordinary road in a traditional city, it’s not clear that we will ever accept a driverless car. I think I’m frankly less bullish than maybe some of your colleagues about the speed at which the machines will really take over and be accepted, that we can trust ourselves to them.

Ariel: As I mentioned at the start, and as you mentioned at the start, you are a techno optimist, for as much as the book is about things that could go wrong it did feel to me like it was also sort of an optimistic look at the future. What are you most optimistic about? What are you most hopeful for looking at both short term and long term, however you feel like answering that?

Martin: I’m hopeful that biotech will have huge benefits for health, will perhaps extend human life spans a bit, but that’s something about which we should feel a bit ambivalent. So, I think health, and also food. If you asked me, what is one of the most benign technologies, it’s to make artificial meat, for instance. It’s clear that we can more easily feed a population of 9 billion on a vegetarian diet than on a traditional diet like Americans consume today.

To take one benign technology, I would say artificial meat is one, and more intensive farming so that we can feed people without encroaching too much on the natural part of the world. I’m optimistic about that. If we think about very long term trends then life extension is something which obviously if it happens too quickly is going to be hugely disruptive, multi-generation families, et cetera.

Also, even though we will have the capability within a century to change human beings, I think we should constrain that on earth and just let that be done by the few crazy pioneers who go away into space. But if this does happen, then as I say in the introduction to my book, it will be a real game changer in a sense. I make the point that one thing that hasn’t changed over most of human history is human character. Evidence for this is that we can read the literature written by the Greeks and Romans more than 2,000 years ago and resonate with the people, and their characters, and their attitudes and emotions.

It’s not at all clear that on some scenarios, people 200 years from now will resonate in anything other than an algorithmic sense with the attitudes we have as humans today. That will be a fundamental, and very fast change in the nature of humanity. The question is, can we do something to at least constrain the rate at which that happens, or at least constrain the way in which it happens? But it is going to be almost certainly possible to completely change human mentality, and maybe even human physique over that time scale. One has only to listen to listen to people like George Church to realize that it’s not crazy to imagine this happening.

Ariel: You mentioned in the book that there’s lots of people who are interested in cryogenics, but you also talked briefly about how there are some negative effects of cryogenics, and the burden that it puts on the future. I was wondering if you could talk really quickly about that?

Martin: There are some people, I know some, who have a medallion around their neck which is an injunction of, if they drop dead they should be immediately frozen, and their blood drained and replaced by liquid nitrogen, and that they should then be stored — there’s a company called Alcor in Arizona that does this — and allegedly revived at some stage when technology advanced. I find it hard to take these seriously, but they say that, well the chance may be small, but if they don’t invest this way then the chance is zero that they have a resurrection.

But I actually think that even if it worked, even if the company didn’t go bust, and sincerely maintained them for centuries and they could then be revived, I still think that what they’re doing is selfish, because they’d be revived into a world that was very different. They’d be refugees from the past, and they’d therefore be imposing an obligation on the future.

We obviously feel an obligation to look after some asylum seeker or refugee, and we might feel the same if someone had been driven out of their home in the Amazonian forest for instance, and had to find a new home, but these refugees from the past, as it were, they’re imposing a burden on future generations. I’m not sure that what they’re doing is ethical. I think it’s rather selfish.

Ariel: I hadn’t thought of that aspect of it. I’m a little bit skeptical of our ability to come back.

Martin: I agree. I think the chances are almost zero, even if they were stored and et cetera, one would like to see this technology tried on some animal first to see if they could freeze animals at liquid nitrogen temperatures and then revive it. I think it’s pretty crazy. Then of course, the number of people doing it is fairly small, and some of the companies doing it, there’s one in Russia, which are real ripoffs I think, and won’t survive. But as I say, even if these companies did keep going for a couple of centuries, or however long is necessary, then it’s not clear to me that it’s doing good. I also quoted this nice statement about, “What happens if we clone, and create a neanderthal? Do we put him in a zoo or send him to Harvard,” said the professor from Stanford.

Ariel: Those are ethical considerations that I don’t see very often. We’re so focused on what we can do that sometimes we forget. “Okay, once we’ve done this, what happens next?”

I appreciate you being here today. Those were my questions. Was there anything else that you wanted to mention that we didn’t get into?

Martin: One thing we didn’t discuss, which was a serious issue, is the limits of medical treatment, because you can make extraordinary efforts to keep people alive long before they’d have died naturally, and to keep alive babies that will never live a normal life, et cetera. Well, I certainly feel that that’s gone too far at both ends of life.

One should not devote so much effort to extreme premature babies, and allow people to die more naturally. Actually, if you asked me about predictions I’d make about the next 30 or 40 years, first more vegetarianism, secondly more euthanasia.

Ariel: I support both, vegetarianism, and I think euthanasia should be allowed. I think it’s a little bit barbaric that it’s not.

Martin: Yes.

I think we’ve covered quite a lot, haven’t we?

Ariel: I tried to.

Martin: I’d just like to mention that my book does touch a lot of bases in a fairly short book. I hope it will be read not just by scientists. It’s not really a science book, although it emphasizes how scientific ideas are what’s going to determine how our civilization evolves. I’d also like to say that for those in universities, we know it’s only interim for students, but we have universities like MIT, and my University of Cambridge, we have convening power to gather people together to address these questions.

I think the value of the centers which we have in Cambridge, and you have in MIT, are that they are groups which are trying to address these very, very big issues, these threats and opportunities. The stakes are so high that if our efforts can really reduce the risk of a disaster by one part in 10,000, we’ve more than earned our keep. I’m very supportive of our Centre for Existential Risk in Cambridge, and also the Future of Life Institute which you have at MIT.

Given the huge numbers of people who are thinking about small risks like which foods are carcinogenic, and the threats of low radiation doses, et cetera, it’s not at all inappropriate that there should be some groups who are focusing on the more extreme, albeit perhaps rather improbable threats which could affect the whole future of humanity. I think it’s very important that these groups should be encouraged and fostered, and I’m privileged to be part of them.

Ariel: All right. Again, the book is On the Future: Prospects for Humanity by Martin Rees. I do want to add, I agree with what you just said. I think this is a really nice introduction to a lot of the risks that we face. I started taking notes about the different topics that you covered, and I don’t think I got all of them, but there’s climate change, nuclear war, nuclear winter, biodiversity loss, overpopulation, synthetic biology, genome editing, bioterrorism, biological errors, artificial intelligence, cyber technology, cryogenics, and the various topics in physics, and as you mentioned the role that scientists need to play in ensuring a safe future.

I highly recommend the book as a really great introduction to the potential risks, and the hopefully much greater potential benefits that science and technology can pose for the future. Martin, thank you again for joining me today.

Martin: Thank you, Ariel, for talking to me.

[end of recorded material]