Martin Rees: Black Holes, Alien Life, Dark Matter, and the Big Bang

Martin Rees (03:00.800)

our galaxy where some sort of biosphere has evolved. And settling where the truth lies

Martin Rees (03:07.680)

between those two extremes is a challenge for the coming decades.

Lex Fridman (03:12.720)

So certainly we're either lucky to be here or very, very, very lucky to be here.

Martin Rees (03:18.880)

I guess that's the difference. Where do you fall? Your own estimate,

Lex Fridman (03:25.280)

your own guess on this question. Are we alone in the universe, do you think?

Martin Rees (03:29.200)

I think we're foolish to give any firm estimate because we just don't know. And that's just an

Martin Rees (03:35.360)

example of how we are depending on greater observations. And also, incidentally,

Martin Rees (03:40.880)

in the case of life, we've got to take account of the fact that, as I always say to my scientific

Martin Rees (03:47.120)

colleagues, biology is a much harder subject than physics. And most of the universe that we know

Martin Rees (03:54.080)

about could be understood by physics. But we've got to remember that even the smallest living

Martin Rees (03:59.680)

organism, an insect, is far more complicated with layer on layer complexity than the most

Martin Rees (04:08.400)

complicated star or galaxy. That's the funny thing about physics

Lex Fridman (04:13.280)

and biology. The dream of physicists in the 20th century and maybe this century is to discover

Martin Rees (04:19.920)

the theory of everything. And there's a sense that once you discover that theory,

Martin Rees (04:27.760)

you will understand everything. If we unlock the mysteries of how the universe works,

Lex Fridman (04:32.160)

would you be able to understand how life emerges from that fabric of the universe that we understand?

Martin Rees (04:39.680)

I think the phrase, theory of everything, is very misleading because it's used to describe

Martin Rees (04:45.840)

a theory which unifies the three laws of microphysics, electricity,

Martin Rees (04:52.080)

magnetism, and weak interaction with gravity. So it's an important step forward for particle

Martin Rees (04:57.920)

physicists. But the lack of such a theory doesn't hold up any other scientists. Anyone doing

Martin Rees (05:04.400)

biology or most of physics is not held up at all through not understanding subnuclear physics.

Martin Rees (05:10.240)

They're held up because they're dealing with things that are very complicated.

Lex Fridman (05:12.880)

And that's especially true of anything biological. So what's holding up biologists is not a lack of

Martin Rees (05:18.720)

the so called theory of everything. It's the inability to understand things which are very

Martin Rees (05:23.920)

complicated. What do you think we'll understand first? How the universe works or how the human

Martin Rees (05:30.000)

body works deeply, like from a fundamental deep level? Well, I think, perhaps we can come back to

Martin Rees (05:36.800)

it later, that there are only limited prospects of ever being able to understand with unaided human

Martin Rees (05:44.800)

brains the most fundamental theories linking together all the forces of nature. I think that

Martin Rees (05:51.120)

may be a limitation of the human brains. But I also think that we can, perhaps aided by computer

Martin Rees (05:58.960)

simulations, understand a bit more of the complexity of nature. But even understanding a

Martin Rees (06:06.640)

simple organism from the atom up is very, very difficult. And I think extreme reductionists have

Martin Rees (06:15.040)

a very misleading perception. They tend to think that, in a sense, we are all solutions to

Martin Rees (06:20.960)

Schrodinger's equation, etc. But that isn't the way we'll ever understand anything. It may be true

Martin Rees (06:28.000)

that we are reductionists in the sense that we believe that that's the case. We don't believe in

Martin Rees (06:34.960)

any special life force in living things. But nonetheless, no one thinks that we can understand

Martin Rees (06:40.240)

a living thing by solving Schrodinger's equation. To take an example which isn't as complicated,

Martin Rees (06:46.320)

lots of people study the flow of fluids like water, why waves break, why flows go turbulent,

Martin Rees (06:52.720)

things like that. This is a serious branch of applied mathematics and engineering. And in doing

Martin Rees (06:59.120)

this, you have concepts of viscosity, turbulence, and things like that. Now, you can understand

Martin Rees (07:06.000)

quite a lot about how water behaves and how waves break in terms of those concepts. But the fact

Martin Rees (07:13.280)

that any breaking wave is a solution of Schrodinger's equation for 10 to the 30 particles,

Martin Rees (07:20.320)

even if you could solve that, which you clearly can't, would not give you any insight.

Lex Fridman (07:26.080)

So the important thing is that every science has its own irreducible concepts

Martin Rees (07:32.080)

in which you get the best explanation. So it may be in chemistry, it's things like valence.

Martin Rees (07:38.720)

In biology, there are concepts in cell biology. And in ecology, there are concepts like

Martin Rees (07:47.280)

imprinting, etc. And in psychology, there are other concepts. So in a sense, the sciences are

Martin Rees (07:55.520)

like a tall building where you have basic physics, most fundamental, then the rest of

Martin Rees (08:01.920)

physics, then chemistry, then cell biology, etc., all the way up to the, I guess, economist in the

Martin Rees (08:08.160)

penthouse and all that. And we have that. And that's true in a sense. But it's not true that

Martin Rees (08:15.920)

it's like a building in that it's made unstable by an unstable base. Because if you're a chemist,

Lex Fridman (08:23.600)

biologist, or an economist, you're facing challenging problems.

Lex Fridman (08:27.760)

But they're not made any worse by uncertainty about subnuclear physics.

Lex Fridman (08:33.120)

TK And at every level, just because you understand the rules of the game,

Martin Rees (08:37.840)

or have some understanding of the rules of the game, doesn't

Lex Fridman (08:40.720)

TK mean you know what kind of beautiful things that game creates.

Martin Rees (08:46.160)

TK Right. So if you're interested in birds and how they fly, then things like imprinting

Martin Rees (08:56.240)

the baby on the mother and all that, and things like that, are what you need to understand.

Martin Rees (09:02.800)

You couldn't even in principle solve this fairly good equation, how an albatross

Martin Rees (09:08.160)

wanders for thousands of miles in the Southern Ocean and comes back and then coughs up

Martin Rees (09:12.400)

food for its young. That's something we can understand, in a sense,

Lex Fridman (09:16.480)

and predict the behavior. But it's not because we can solve it on the atomic scale.

Martin Rees (09:21.360)

TK You mentioned that there might be some fundamental limitation to the human brain

Martin Rees (09:26.720)

that limits our ability to understand some aspect of how the universe works. That's really

Lex Fridman (09:32.080)

interesting. That's sad, actually. To the degree it's true, it's sad. So what do you mean by that?

Martin Rees (09:39.600)

TK I would simply say that just as a monkey can't understand quantum theory or even Newtonian

Martin Rees (09:46.000)

physics, there's no particular reason why the human brain should have evolved to be well matched

Martin Rees (09:52.400)

to understanding the deepest aspects of reality. And I suspect that there may be aspects that we

Martin Rees (09:59.120)

are not even aware of and couldn't really fully comprehend.

Lex Fridman (10:03.280)

But as an intermediate step towards that, one thing which I think is a very interesting

Martin Rees (10:07.200)

possibility is the extent to which AI can help us. I mean, I think if you take the example of

Lex Fridman (10:14.640)

so called theories of everything, one of which is string theory.

Martin Rees (10:18.480)

String theory involves very complicated geometry and structures in 10 dimensions.

Lex Fridman (10:24.880)

And it's certainly, in my view on the cards, that the physics of 10 dimensions,

Martin Rees (10:36.480)

very complicated geometry, may be too hard for a human being to work through,

Lex Fridman (10:42.400)

but could be worked through by an AI with the advantage of the huge processing power,

Martin Rees (10:50.640)

which enables them to learn world championship chess within a few hours just by watching games.

Lex Fridman (10:55.600)

So there's every reason to expect that these machines could help us to solve these problems.

Lex Fridman (11:02.800)

And of course, if that's the way we came to understand whether string theory was right,

Martin Rees (11:08.000)

it should be in a sense frustrating because you wouldn't get the sort of aha insight,

Martin Rees (11:13.120)

which is the greatest satisfaction from doing science. But on the other hand, if a machine

Martin Rees (11:19.040)

churns away at 10 dimensional geometry, figuring out all the possible origamis wound up in extra

Martin Rees (11:26.560)

dimensions, if it comes out at the end, spews out the correct mass of the electron, the fact that

Martin Rees (11:32.960)

there are three kinds of neutrinos, something like that, you would know that there was some truth in

Martin Rees (11:36.960)

the theory. And so we may have a theory which we come to trust because it does predict things that

Martin Rees (11:43.200)

we can observe and check, but we may never really understand the full workings of it to the extent

Martin Rees (11:49.760)

that we do more or less understand how most phenomena can be explained in a fundamental way.

Lex Fridman (11:57.840)

Of course, in the case of quantum theory, many people would say,

Martin Rees (12:01.120)

understand if there's still some mystery, if you don't quite understand why it works.

Lex Fridman (12:04.640)

But there could be deeper mysteries when we get to these unified theories, where there's a big gap

Martin Rees (12:09.200)

between what a computer can print out for us at the end and what we can actually grasp and think

Martin Rees (12:16.560)

through in our heads. Yeah, it's interesting that the idea that there could be things a computer

Martin Rees (12:21.280)

could tell us that is true. And maybe it can even help us understand why it's true a little bit,

Lex Fridman (12:29.040)

but ultimately it's still a long journey to really deeply understand the whys of it.

Martin Rees (12:34.400)

Yes, and that's the limitation of our brain. We can try to sneak up to it in different ways,

Martin Rees (12:40.800)

given the limitations of our brain. I've gotten a chance to spend the day at DeepMind,

Martin Rees (12:45.680)

talk to Demis Hassabis. His big dream is to apply AI to the questions of science,

Lex Fridman (12:52.320)

certainly to the questions of physics. Have you gotten a chance to interact with him?

Martin Rees (12:55.840)

Well, I know him quite well. He's one of my heroes, certainly.

Lex Fridman (12:59.520)

I'm sure he would say the same.

Lex Fridman (13:02.080)

And I remember the first time I met him, he said that he was like me, he wants to understand the

Martin Rees (13:07.280)

universe, but he thought the best thing to do was to try and develop AI. And then with the help of

Martin Rees (13:12.880)

AI, he'd stand more chance of understanding the universe. And I think he's right about that.

Lex Fridman (13:18.080)

And of course, although we're familiar with the way his computers play go and chess,

Martin Rees (13:26.640)

he's already made contributions to science through understanding protein folding better

Martin Rees (13:33.040)

than the best human chemists. And so already he's on the path to showing ways in which computers

Martin Rees (13:40.240)

have the power to learn and do things by having ability to analyze enormous samples in a short

Martin Rees (13:45.840)

time to do better than humans. And so I think he would resonate for what I just said, that it may

Martin Rees (13:52.240)

be that in these other fundamental questions, the computers will play a crucial role.

Martin Rees (13:58.080)

Yeah. And they're also doing a quantum mechanical simulation of electrons. They're doing

Martin Rees (14:03.600)

control of high temperature plasmas, fusion reactors.

Martin Rees (14:07.760)

Yes, that's a new thing, which is very interesting. They can suppress the instabilities

Martin Rees (14:11.520)

in these tokamaks better than any other way. Yeah.

Lex Fridman (14:14.880)

And it's just the march of progress by AIs in science is making big strides. Do you think

Lex Fridman (14:24.720)

an AI system will win a Nobel Prize in the century? What do you think?

Lex Fridman (14:31.520)

And does that make you sad?

Martin Rees (14:33.520)

If I can digress and put in a plug for my next book, it has a chapter saying,

Lex Fridman (14:38.240)

why Nobel Prizes do more harm than good. So on a quite separate subject, I think

Martin Rees (14:43.760)

the Nobel Prizes do a great deal of damage to the perception of the way science is done.

Martin Rees (14:49.120)

Of course, if you ask who or what deserves the credit for any scientific discovery,

Martin Rees (14:53.760)

it may be often someone who has an idea, a team of people who work a big experiment, etc.

Lex Fridman (14:59.600)

And of course, it's the quality of the equipment, which is crucial. And certainly in the subjects

Martin Rees (15:07.360)

I do in astronomy, the huge advances we've had come not from us being more intelligent than

Martin Rees (15:15.120)

Aristotle was, but through us having far, far better data from powerful telescopes on the

Martin Rees (15:22.240)

ground and in space. And also, incidentally, we've benefited hugely in astronomy from computer

Martin Rees (15:29.040)

simulations. Because if you are a subatomic physicist, then of course, you crash together

Martin Rees (15:36.960)

the particles in the big accelerator like the one at CERN and see what happens. But I can't

Martin Rees (15:43.280)

crash together two galaxies or two stars and see what happens. But in the virtual world of a

Martin Rees (15:48.880)

computer, one can do simulations like that. And the power of computers is such that these

Martin Rees (15:56.160)

simulations can yield phenomena and insights which we wouldn't have guessed beforehand. And the way

Martin Rees (16:05.840)

we can feel we're making progress and trying to understand some of these phenomena, why galaxies

Martin Rees (16:11.280)

have the size and shape they do and all that, is because we can do simulations and tweaking

Martin Rees (16:16.880)

different initial conditions and seeing which gives the best fit to what we actually observe.

Lex Fridman (16:22.560)

And so that's a way in which we've made progress in using computers. And incidentally, we also now

Martin Rees (16:30.720)

need to analyze data because one thinks of astronomy as being traditionally rather data

Martin Rees (16:34.880)

poor subjects. But the European satellite called Gaia has just put online the speeds and colors

Lex Fridman (16:43.840)

and properties of nearly two billion stars in the Milky Way, which we can do fantastic analyses of.

Lex Fridman (16:50.560)

And that, of course, could not be done at all without just the number crunching capacity

Martin Rees (16:54.320)

of computers. And the new methods of machine learning actually love raw data, the kind that

Martin Rees (17:01.280)

astronomy provides, organized, structured, raw data. Well, indeed, because the reason they really

Martin Rees (17:06.640)

have a benefit over us is that they can learn and think so much faster. That's how they can

Martin Rees (17:12.720)

learn to play chess and go. That's how they can learn to diagnose lung cancer better than a

Martin Rees (17:17.920)

radiologist because they can look at 100,000 scans in a few days, whereas no human radiologist sees

Martin Rees (17:26.400)

that many a lifetime. Well, there's still magic to the human intelligence, to the intuition,

Lex Fridman (17:33.040)

to the common sense reasoning. Well, we hope so. For now. What is the new book that you mentioned?

Martin Rees (17:40.480)

The book I mentioned is called If Science is to Save Us. It's coming out in September.

Lex Fridman (17:44.880)

And it's on the big challenges of science, climate, dealing with biosafety and dealing

Martin Rees (17:56.560)

with cyber safety. And also, it's got chapters on the way science is organized, universities

Lex Fridman (18:03.840)

and academies, et cetera, and the ethics of science and education. And the limits, yes.

Martin Rees (18:12.560)

Yes. Well, let me actually just stroll around the beautiful and the strange of the universe.

Martin Rees (18:18.960)

Over 20 years ago, you hypothesized that we would solve the mystery of dark matter by now.

Lex Fridman (18:25.760)

So unfortunately, we didn't quite yet. First, what is dark matter and why has it been so tough

Martin Rees (18:34.320)

to figure out? Well, I mean, we learned that galaxies and other large scale structures,

Martin Rees (18:39.520)

which are moving around but preventive of flying apart by gravity, would be flying apart if they

Martin Rees (18:50.720)

only contain the stuff we see, if everything in them is shining. And to understand how galaxies

Martin Rees (18:57.520)

formed and why they do remain confined the same size, one has to infer that there's about five

Martin Rees (19:06.000)

times as much stuff producing gravitational forces than the total amount of stuff in the

Martin Rees (19:12.720)

gas and stars that we see. And that stuff is called dark matter. That's the leading name,

Martin Rees (19:18.960)

it's not dark, it's just transparent, et cetera. And the most likely interpretation is that it's

Martin Rees (19:26.560)

a swarm of microscopic particles which have no electric charge. And the very small cross sections

Martin Rees (19:34.480)

were hitting each other and hitting anything else. So they swarm around and we can detect their

Martin Rees (19:39.680)

collective effects. And when we do computer simulations of how galaxies form and evolve

Lex Fridman (19:45.680)

and how they emerge from the Big Bang, then we get a nice consistent picture if we put in

Martin Rees (19:53.680)

five times as much mass in the form of these mysterious dark particles. And for instance,

Martin Rees (1:00:08.560)

you know, people talk about climate change and that we need to respond to climate change. And

Martin Rees (1:00:13.040)

that's a long term investment we need to make, but it's not really long term. It's a span of decades.

Martin Rees (1:00:19.520)

I think what Elon is doing is a really long term investment.

Martin Rees (1:00:22.800)

We should be working on multi planetary colonization now if we were to have it ready

Martin Rees (1:00:29.280)

five centuries from now. And so taking those early steps. And then also there's something

Martin Rees (1:00:35.440)

happens when you go into the unknown and do this really difficult thing. You discover

Martin Rees (1:00:42.480)

something very new. You discover something about robotics. You discover something about

Martin Rees (1:00:47.520)

materials engineering or nutrition or neuroscience or human relations or political systems that

Martin Rees (1:00:53.680)

actually work well with humans. You discover all those things. And so it's a worthy effort to go

Martin Rees (1:00:59.600)

out there and try to become cyborgs. Yeah, no, I agree with that. I think the only different point

Martin Rees (1:01:08.000)

I'd make is that this is going to be very, very, very, very, very, very, very, very, very, very, very,

Martin Rees (1:01:17.360)

very expensive if it's done in a risk averse way. And that's why I think we should be grateful to

Martin Rees (1:01:24.480)

the billionaires if they're going to sort of foster these opportunities for thrill seeking

Martin Rees (1:01:33.520)

risk takers who we can all admire. Yeah. By the way, I should push back on the billionaires

Martin Rees (1:01:38.160)

because there's sometimes a negative connotation to the word billionaire. It's not a billionaire.

Martin Rees (1:01:41.760)

It's a company versus government because governments are billionaires and trillionaires.

Martin Rees (1:01:46.640)

Yeah. It's not the wealth. It's the capitalist imperative, which I think

Martin Rees (1:01:58.320)

deserves a lot more praise than people are giving it. I'm troubled by the sort of criticism like

Martin Rees (1:02:04.240)

it's billionaires playing with toys for their own pleasure. I think what some of these companies

Martin Rees (1:02:11.760)

like SpaceX and Blue Origin are doing is some of the most inspiring engineering and even scientific

Martin Rees (1:02:19.840)

work ever done in human history. No, I agree. I think the people who've made the greatest wealth

Lex Fridman (1:02:26.480)

are people who've really been mega benefactors. I mean, I think, you know.

Martin Rees (1:02:31.120)

Some of them, some of them. Yeah, yes, some of them. But those who've founded Google and all

Martin Rees (1:02:37.040)

that and even Amazon, they're beneficiaries. They're in a quite different category, in my view,

Martin Rees (1:02:44.560)

from those who just shuffle around money or crypto coins and things like that.

Lex Fridman (1:02:52.080)

Now you're really talking trash.

Martin Rees (1:02:54.880)

Yes. But I think if they use their money in these ways, that's fine. But I think it's true that far

Martin Rees (1:03:02.800)

more money is owned by us collectively as taxpayers. But I think the fact is that in a democracy,

Martin Rees (1:03:09.920)

there'd be big resistance to exposing human beings to very high risks if in a sense we

Martin Rees (1:03:16.640)

share responsibility for it. And that's the reason I think it would be done much more cheaply by

Martin Rees (1:03:23.280)

these private funders.

Martin Rees (1:03:24.560)

That's an interesting hypothesis, but I have to push back. I don't know if it's obvious

Lex Fridman (1:03:29.600)

why NASA spends so much money and takes such a long time to develop the things it was doing.

Lex Fridman (1:03:38.240)

So before Elon Musk came along, because I would love to live in a world where government

Martin Rees (1:03:44.880)

actually uses taxpayer money to get some of the best engineers and scientists in the world

Lex Fridman (1:03:49.200)

and actually work across governments, Russia, China, United States, European Union together

Martin Rees (1:03:55.040)

to do some of these big projects. It's strange that Elon is able to do this much cheaper,

Lex Fridman (1:04:00.960)

much faster. It could have to do with risk aversion, you're right.

Lex Fridman (1:04:06.160)

But I think it's that he had all the whole assembly within this one building as it were,

Martin Rees (1:04:15.600)

rather than depending on a supply chain. But I think it's also that he had a Silicon Valley

Martin Rees (1:04:22.000)

culture and had younger people, whereas the big aerospace companies, Boeing and Lockheed Martin,

Martin Rees (1:04:28.720)

they had people who were left over from the Apollo program in some cases. And so they weren't

Martin Rees (1:04:33.840)

quite so lively. And indeed, quite apart from the controversial issues of the future of human

Martin Rees (1:04:40.000)

spaceflight, in terms of the next generation of big rockets, then the one that Musk is going to

Martin Rees (1:04:48.000)

launch for the first time this year, the huge one, is going to be far, far cheaper than the

Martin Rees (1:04:55.200)

one that NASA has been working on at the same time. And that's because it will have a reusable

Martin Rees (1:05:01.680)

first stage. And it's going to be great. It can launch over 100 tons into Earth orbit. And

Martin Rees (1:05:09.680)

incidentally, that's going to make it feasible to do things that I used to think were crazy,

Martin Rees (1:05:14.320)

like having solar energy from space. That's no longer so crazy if you can do that. And also,

Martin Rees (1:05:21.200)

for science, because its nose cone could contain within it something as big as the entire unfurled

Martin Rees (1:05:31.520)

James Webb telescope mirror. And therefore, you can have a big telescope much more cheaply if you

Martin Rees (1:05:37.440)

can launch it all in one piece. And so it's going to be hugely beneficial to science and to any

Martin Rees (1:05:43.440)

practical use of space to have these cheaper rockets that are far more completely reusable

Martin Rees (1:05:49.280)

than anyways NASA had. So I think Musk's done a tremendous service to space exploration and the

Martin Rees (1:05:56.320)

whole space technology through these rockets, certainly. Plus, it's some big sexy rocket.

Martin Rees (1:06:03.120)

It's just great engineering. Of course, yeah. It's like looking at a beautiful big bridge

Martin Rees (1:06:08.160)

that humans are capable. Us descendants of apes are capable to do something so majestic.

Lex Fridman (1:06:12.880)

Yes. And also the way they land coming down on this bar. That's amazing.

Martin Rees (1:06:16.800)

It's both controls engineering. It's increasing sort of intelligence in these rockets, but also

Martin Rees (1:06:24.240)

great propulsion engineering materials, entrepreneurship. And it just inspires so

Martin Rees (1:06:31.440)

many people. No, I'm entirely with you on that. So would it be exciting to you to see a human

Martin Rees (1:06:38.400)

being step foot on Mars in your lifetime? Yes, I think it's unlikely in my lifetime

Martin Rees (1:06:43.280)

since I'm so ancient. But I think this century, it's going to happen. And I think that will

Martin Rees (1:06:50.080)

indeed be exciting. And I hope there will be a small community by the end of a century.

Lex Fridman (1:06:55.040)

But as I say, I think they may go with one way tickets or accepting the risk of no return.

Lex Fridman (1:07:03.600)

And so they've got to be people like that. And I still think it's going to be hard to persuade

Martin Rees (1:07:10.240)

the public to send people when you say straight out that they may never come back. But of course,

Martin Rees (1:07:18.160)

the Apollo astronauts, they took a high risk. And in fact, in my previous book, I quote the

Martin Rees (1:07:25.600)

speech that has been written for Nixon to be read out if Neil Armstrong got stuck on the moon.

Lex Fridman (1:07:31.200)

And it was written by one of his advisors and very eloquent speech about how they have come

Martin Rees (1:07:42.480)

to a noble end, et cetera. But of course, there was a genuine risk at that time. But that may have

Martin Rees (1:07:51.600)

been accepted. But clearly, the crashes of the shuttle were not acceptable to the American

Martin Rees (1:07:59.200)

public, even when they were told that this was only a 2% risk given how often they launched it.

Lex Fridman (1:08:05.280)

And so that's what leads me to think that it's got to be left to the kind of people who are

Martin Rees (1:08:13.760)

prepared to take these risks. And I think of American Avengers, there was a guy called Steve

Martin Rees (1:08:19.520)

Fossett who was an aviator who did all kinds of crazy things. And then a guy who fell

Martin Rees (1:08:25.600)

supersonically with a parachute from very high altitude. All these people, we all share them on.

Martin Rees (1:08:32.640)

They extend the bounds of humanity. But I don't think the public will be so happy to fund them.

Martin Rees (1:08:39.360)

I mean, I disagree with that. I think if we change the narrative, we should change the story.

Lex Fridman (1:08:44.640)

You think so?

Martin Rees (1:08:45.200)

I think there's a lot of people... Because the public is happy to fund folks in other domains

Lex Fridman (1:08:53.760)

that take bold, giant risks. First of all, military, for example.

Martin Rees (1:08:57.600)

Oh, in the military, obviously, yes.

Martin Rees (1:08:59.600)

I think this is, in the name of science, especially if it's sold correctly, I sure as hell would go

Martin Rees (1:09:05.200)

up there with a risk... I would take a 40% chance risk of death for something that's...

Lex Fridman (1:09:12.480)

I would. I might want to be even older than I am now. But then I would go.

Martin Rees (1:09:17.600)

I guess what I'm trying to communicate is there's a lot of people on Earth. That's the nice feature

Lex Fridman (1:09:22.400)

and I'm sure there's going to be a significant percentage or some percentage of people that are...

Martin Rees (1:09:27.360)

They take on the risk for the adventure. So, and I particularly love that that risk of adventure

Martin Rees (1:09:36.240)

when taking on inspires people and just the ripple effect it has across the generation,

Martin Rees (1:09:40.960)

especially among the young minds, is perhaps immeasurable. But you're thinking that sending

Martin Rees (1:09:50.880)

humans should be something we do less and less, sending humans to space. That it should be

Martin Rees (1:09:56.880)

primarily an effort. The work of space exploration should be done primarily by robots.

Martin Rees (1:10:03.680)

Well, I think it can be done much more cheaply, obviously, on Mars. And no one's thinking of

Martin Rees (1:10:09.600)

sending humans to Enceladus or Europa, the outer planets. And the point is we will have much better

Martin Rees (1:10:18.000)

robots. Because let's take an example. You've seen the pictures of the moons of Saturn and the

Martin Rees (1:10:27.920)

picture of Pluto and the comet taken by probes and Cassini spent 13 years going around Saturn and

Martin Rees (1:10:37.520)

its moons after seven year voyage. And those are all based on 1990s technology. And if you think

Martin Rees (1:10:43.040)

of how smartphones have advanced in the 20 years since then, just think how much better one could

Martin Rees (1:10:47.600)

do instrumenting some very small, sophisticated probe. You could send dozens of them to explore

Martin Rees (1:10:54.160)

the outer planets. And that's the way to do that. Because no one thinks you could send humans that

Martin Rees (1:10:58.960)

far. But I would apply the same argument to Mars. And if you want to assemble big structures like,

Martin Rees (1:11:07.760)

for instance, radio astronomers would like to have a big radio telescope on the far side of the moon

Lex Fridman (1:11:12.240)

so it's away from the Earth's background artificial radio waves. And that could be

Martin Rees (1:11:20.640)

done by assembling using robots without people. So on the moon and on Mars, I think

Martin Rees (1:11:29.520)

everything that's useful can be done by machines much more cheaper than by humans.

Lex Fridman (1:11:35.840)

Do you know the movie 2001 A Space Odyssey?

Martin Rees (1:11:39.120)

Of course, yes. You must be too young to have seen that when it came out, obviously.

Martin Rees (1:11:44.720)

I remember seeing it when it came out.

Lex Fridman (1:11:46.800)

You saw it when it came out.

Martin Rees (1:11:48.800)

Yeah, yeah, 50 years ago.

Lex Fridman (1:11:50.640)

60, when was it? 60, in the 60s.

Martin Rees (1:11:54.960)

Yeah, that's right. Still a classic.

Lex Fridman (1:11:58.240)

It's still probably, for me, the greatest AI movie ever made.

Martin Rees (1:12:03.280)

Yes, yes, I agree.

Lex Fridman (1:12:04.720)

One of the great space movies ever made.

Martin Rees (1:12:06.720)

Yes. Well, let me ask you a philosophical question. Since we're talking about robots

Martin Rees (1:12:12.240)

exploring space, do you think HAL 9000 is good or bad? So for people who haven't watched,

Martin Rees (1:12:20.800)

this computer system makes a decision to basically prioritize the mission that

Lex Fridman (1:12:28.720)

the ship is on over the humans that are part of the mission. Do you think HAL is good or evil?

Martin Rees (1:12:38.160)

If you ask me, probably in that context, it was probably good. But I think you're raising

Lex Fridman (1:12:43.040)

what is, of course, very much an active issue in everyday life about the extent to which we should

Martin Rees (1:12:50.640)

entrust any important decision to a machine. And there again, I'm very worried, because I think

Martin Rees (1:12:58.560)

if you are recommended for an operation or not given parole from prison or even denied credit

Martin Rees (1:13:06.960)

by your bank, you feel you should be entitled to an explanation. It's not enough to be told that

Martin Rees (1:13:13.920)

the machine has a more reliable record on the whole than humans have of making these decisions.

Martin Rees (1:13:20.720)

You think you should be given reasons you could understand. And that's why I think the present

Martin Rees (1:13:26.560)

societal trend to take away the humans and leave us in the hands of decisions that we can't contest

Martin Rees (1:13:37.360)

is a very dangerous one. I think we've got to be very careful of the extent to which AI,

Martin Rees (1:13:44.480)

which can handle lots of information, actually makes the decision without oversight. And I think

Martin Rees (1:13:50.000)

we can use them as a supplement. But to take the case of radiology and cancer,

Martin Rees (1:13:57.840)

I mean, it's true that the radiologist hasn't seen as many x rays of cancer lungs as the machine. So

Martin Rees (1:14:08.160)

the machine could certainly help, but you want the human to make the final decision. And I think

Martin Rees (1:14:12.400)

that's true in most of these instances. But if we turn a bit to the short term concerns with

Martin Rees (1:14:19.360)

robotics, I think the big worry, of course, is the effect it has on people's self respect and

Martin Rees (1:14:27.120)

their labor market. And I think my solution would be that we should arrange to tax more heavily

Martin Rees (1:14:36.080)

the big international conglomerates, which use the robots and use that tax to fund

Martin Rees (1:14:48.080)

decently paid, dignified posts of the kind where being a human being is important.

Martin Rees (1:14:54.000)

Above all, carers for old people, teachers assistants for young, gardens in public parks,

Lex Fridman (1:14:59.520)

and things like that. And if the people who are now working in mind numbing jobs in Amazon

Martin Rees (1:15:05.920)

warehouses, or in telephone call centers, are automated, but those same people are given

Martin Rees (1:15:15.680)

jobs where being a human is an asset, then that's a plus plus situation. And so that's

Martin Rees (1:15:22.320)

the way I think that we should benefit from these technologies, take over the mind numbing jobs,

Lex Fridman (1:15:31.360)

and use machines to make them more efficient, but enable the people so displaced to do jobs

Martin Rees (1:15:40.800)

where we do want a human being. I mean, most people when they're old, they're rich people

Martin Rees (1:15:46.720)

if they have the choice. They want human carers and all that, don't they? They may want robots

Martin Rees (1:15:51.440)

to help with some things, empty the bed pans and things like that, but they want real people. And

Martin Rees (1:15:59.760)

certainly in this country, I think even worse in America, the care of old people is completely

Martin Rees (1:16:06.400)

inadequate. And it needs just more human beings to help them cope with everyday life and look after

Martin Rees (1:16:11.360)

them when they're sick. And so that seems to be the way in which the money raised in tax from

Martin Rees (1:16:19.360)

these big companies should be deployed. So that's in the short term, but if you actually just look,

Martin Rees (1:16:24.240)

the fact is where we are today to longterm future in a hundred years, it does seem that

Martin Rees (1:16:31.520)

there is some significant chance that the human species is coming to an end in its

Lex Fridman (1:16:39.200)

pure biological form. There's going to be greater and greater integration

Martin Rees (1:16:43.680)

through genetic modification and cyborg type of creatures. And so you have to think, all right,

Martin Rees (1:16:50.800)

well, we're going to have to get from here to there and that process is going to be painful.

Lex Fridman (1:16:56.080)

And there's so many different trajectories that take us from one place to another.

Martin Rees (1:17:02.400)

It does seem that we need to deeply respect humanness and humanity, basic human rights,

Martin Rees (1:17:09.840)

human welfare, like happiness and all that kind of stuff.

Martin Rees (1:17:14.880)

No, absolutely. And that's why I think we ought to try and slow down the application of these

Martin Rees (1:17:20.240)

human enhancement techniques and cyborg techniques for humans for just that reason. I mean,

Martin Rees (1:17:26.560)

that's why I want to lead into the people on Mars, let them do it, but for just that reason.

Lex Fridman (1:17:32.320)

But there are people too. Okay. People on Mars are people too. I tend to, you know.

Lex Fridman (1:17:37.360)

But they are very poorly adapted to where they are. That's why they need this modification,

Martin Rees (1:17:42.800)

whereas we're adapted to the Earth quite well. So we don't need these modifications. We're

Martin Rees (1:17:49.200)

happy to be humans living in the environment where our ancestors lived. So we don't have the same

Martin Rees (1:17:55.520)

motive. So I think there's a difference. But I agree, we don't want drastic changes probably in

Martin Rees (1:18:01.040)

our lifestyle. And that indeed is a worry because some things are changing so fast.

Lex Fridman (1:18:06.480)

But I think I'd like to inject a note of caution. If you think of the way progress in one technology

Martin Rees (1:18:13.760)

goes, it goes in a sort of spurt. It goes up very fast and then it levels off. Let me give you

Martin Rees (1:18:22.000)

two examples. One we've had already, a human space flight. At the time of the Apollo program,

Martin Rees (1:18:28.400)

which was only 12 years after Sputnik 1, I was alive then. And I thought it would only be 10

Martin Rees (1:18:35.920)

or 20 years further before there were footprints on Mars. But as we know, for reasons we could all

Martin Rees (1:18:41.600)

understand, that still remains the high point of human space exploration. That's because it was

Martin Rees (1:18:50.560)

funded for reasons of superpower rivalry at huge public expense. But let me give you another case,

Martin Rees (1:18:57.600)

civil aviation. If you think of the change between 1919, when that was Alcock and Brown's first

Martin Rees (1:19:05.200)

transatlantic fight, to 1979, the first flight of the jumbo jet. It was a big change. But it's more

Martin Rees (1:19:12.400)

than 50 years since 1969. And we still have jumbo jets more or less the same. So that's an example

Martin Rees (1:19:17.280)

of something which developed fast. And to take another analogy, we've had huge developments in

Martin Rees (1:19:24.880)

mobile phones. But I suspect the iPhone 24 may not be too different from the iPhone 13.

Martin Rees (1:19:32.880)

They develop, but then they saturate, and then maybe some new innovation takes over

Martin Rees (1:19:39.680)

in stimulating economic growth. Yeah, so it's that we have to be cautious about being too optimistic,

Lex Fridman (1:19:47.680)

and we have to be cautious about being too cynical. I think that is the

Lex Fridman (1:19:52.160)

optimistic is begging the question. I mean, do we want this rapid change?

Martin Rees (1:19:56.240)

Right. So first of all, there's some degree to which technological advancement is something,

Martin Rees (1:20:01.760)

is a force that can't be stopped. And so the question is about directing it versus stopping it

Martin Rees (1:20:07.680)

or slowing it. Well, it can be sort of sloped or slow. We'll take human spaceflight. There could

Martin Rees (1:20:11.680)

have been footprints on Mars if America had gone on spending 4% of the federal budget on the project

Martin Rees (1:20:19.040)

after Apollo. There were very good reasons, and we could have had supersonic flight, but Concorde

Martin Rees (1:20:29.600)

came and went during the 50 years. But the reason it didn't progress is not because we realize it's

Martin Rees (1:20:37.440)

not good for human society. The reason it didn't progress is because it couldn't make,

Martin Rees (1:20:41.760)

sort of from a capitalist perspective, it couldn't make, there was no short term or long term way for

Martin Rees (1:20:47.760)

it to make money. So for me, it isn't, but that's the same as saying it's not good for society.

Martin Rees (1:20:54.080)

I don't think everything that makes money is good for society and everything that doesn't make money

Martin Rees (1:20:59.760)

is bad for society, right? That's a difficult thing we're always contending with.

Martin Rees (1:21:06.640)

That's a difficult thing we're always contending with when we look at social networks. It's not

Martin Rees (1:21:14.800)

obvious, even though they make a tremendous amount of money, that they're good for society,

Martin Rees (1:21:19.040)

especially how they're currently implemented with advertisement and engagement maximization.

Lex Fridman (1:21:24.160)

So that's the constant struggle.

Lex Fridman (1:21:26.240)

But I would have thought that supersonic flight was something that would benefit only a tiny

Martin Rees (1:21:38.720)

elite at a huge expense and environmental damage. That was obviously something which we're very

Lex Fridman (1:21:44.480)

glad not to have, in my opinion.

Martin Rees (1:21:46.720)

Yeah, but perhaps there was a way to do it where it could benefit the general populace.

Martin Rees (1:21:51.840)

If you were to think about airplanes, wouldn't you think that in the early days, airplanes would

Martin Rees (1:21:56.880)

have been seen as something that can surely only benefit 1% at most of the population,

Martin Rees (1:22:03.600)

as opposed to a much larger percentage? There's another aspect of capitalist system that's able

Martin Rees (1:22:09.520)

to drive down costs once you get the thing kind of going. So we get together maybe with taxpayer

Martin Rees (1:22:15.760)

money and get the thing going at first. And once it gets going, companies step up and drive down

Martin Rees (1:22:21.200)

the cost and actually make it so that blue collar folks can actually start using the stuff.

Lex Fridman (1:22:26.400)

Yeah, sometimes that does happen. That's good.

Martin Rees (1:22:29.360)

Yeah. So that's, again, the double edged sword of human civilization, that some technology

Martin Rees (1:22:37.280)

hurts us, some benefits us, and we don't know ahead of time. We can just do our best.

Martin Rees (1:22:41.840)

Yes. There's a gap between what could be done and what we collectively decide to do.

Lex Fridman (1:22:49.120)

Yes.

Lex Fridman (1:22:49.440)

And it could push forward some developments faster than we do.

Lex Fridman (1:22:55.120)

Let me ask you, in your book on the future prospects for humanity,

Martin Rees (1:23:01.280)

you imagine a time machine that allows you to send a tweet length message to scientists in

Martin Rees (1:23:05.440)

the past, like to Newton. What tweet would you send? It's an interesting thought experiment.

Lex Fridman (1:23:12.160)

What message would you send to Newton about what we know today?

Martin Rees (1:23:15.680)

Well, I think he'd love to know that there were planets around other stars. He'd like to know

Martin Rees (1:23:22.000)

that everything was made of atoms. He'd like to know that if he looked a bit more carefully

Martin Rees (1:23:29.440)

through his prisms and looked at light, not just from the sun, but from some flames,

Martin Rees (1:23:38.400)

he might get the idea that different substances emitted light of different colors, and he might

Martin Rees (1:23:44.320)

have been twigged to discover some things that had to wait 200 or 300 years. Could have given him

Martin Rees (1:23:51.920)

those clues, I think. It's fascinating to think, to look back at how little

Lex Fridman (1:23:59.360)

he understood, people at that time understood about our world.

Martin Rees (1:24:04.880)

Yes. And certainly about the cosmos.

Lex Fridman (1:24:07.280)

About the cosmos, yes.

Martin Rees (1:24:08.880)

Well, if you think about astronomy, then until about 1850, astronomy was a matter of

Martin Rees (1:24:19.280)

the positions of how the stars and the planets moved around. Of course, that goes back a long

Martin Rees (1:24:25.440)

way, but Newton understood why the planets moved around in ellipses. But he didn't understand

Lex Fridman (1:24:31.760)

why the solar system was all in a plane, what we call the ecliptic, and he didn't understand it.

Martin Rees (1:24:39.760)

No one did until the mid 19th century what the stars are made of. I mean, they were thought to

Martin Rees (1:24:44.400)

be made of some fifth essence, not earth, air, fire, and water like everything else,

Lex Fridman (1:24:49.040)

and it was only after 1850 when people did use prisms more precisely to get spectra that they

Martin Rees (1:24:56.640)

realized that the sun was made of the same stuff as the earth, and indeed the stars were. It wasn't

Martin Rees (1:25:04.960)

until 1930 that people knew about nuclear energy and knew what kept the sun shining for so long.

Lex Fridman (1:25:12.480)

So it was quite late that some of these key ideas came in, which would have completely

Martin Rees (1:25:17.600)

transformed Newton's views, and of course, the entire scale of the galaxy and the rest of the

Martin Rees (1:25:25.360)

universe. Just imagine what he would have thought about the Big Bang, or even just general

Martin Rees (1:25:30.320)

relativity, just gravity, just him and Einstein talking for a couple weeks. Would he be able to

Martin Rees (1:25:39.200)

make sense of space time and the curvature of space time? Well, I think given a quick course,

Martin Rees (1:25:46.240)

I mean, he was sort of, if one looks back, he was really a unique intellect in a way,

Martin Rees (1:25:51.680)

you know, and he said that he thought better than everyone else by thinking on things continually,

Lex Fridman (1:26:00.640)

and thinking very deep thoughts, and so he was an utterly remarkable intellect, obviously.

Lex Fridman (1:26:06.400)

But of course, scientists aren't all like that. I think one thing that's interesting to me,

Martin Rees (1:26:10.320)

having spent a life among scientists, is what a variety of mindsets and mental styles they have.

Lex Fridman (1:26:16.560)

And just to contrast Newton and Darwin, Darwin said, and he's probably correct, that he thought

Martin Rees (1:26:30.560)

he just had as much common sense and reasoning power as the average lawyer. And that's probably

Martin Rees (1:26:38.720)

true because his ability was to sort of collect data and think through things deeply. That's a

Martin Rees (1:26:45.200)

quite different kind of thinking from what was involved in Newton or someone doing abstract

Martin Rees (1:26:50.560)

mathematics. I think in the 20th century, the coolest, well, there's the theory, but

Martin Rees (1:26:59.200)

from an astronomy perspective, black holes is one of the most fascinating entities to have been,

Martin Rees (1:27:08.960)

through theory and through experiment, to have emerged from. Obviously, I agree. It's an amazing

Martin Rees (1:27:13.680)

story. Well, of course, what's interesting is Einstein's reaction because, as you know,

Martin Rees (1:27:22.080)

we now accept this is one of the most remarkable predictions of Einstein's theory. He never took

Martin Rees (1:27:27.520)

it seriously, even believed it, although it was a consequence of a solution of his equations,

Martin Rees (1:27:34.480)

which someone discovered just a year after his theory, Schwarzschild. But he never took it

Martin Rees (1:27:40.640)

seriously, and others did. But then, of course, well, this is something that I've been involved

Martin Rees (1:27:47.600)

in actually finding evidence for black holes, and that's come in the last 50 years. So now,

Martin Rees (1:27:53.040)

there's pretty compelling evidence that they exist as the remnants of stars or big ones in

Martin Rees (1:27:58.800)

the center of galaxies. And we understand what's going on. We have ideas, vaguely on how they form.

Martin Rees (1:28:07.120)

And, of course, gravitational waves have been detected, and that's an amazing piece of technology.

Lex Fridman (1:28:14.160)

Ligo is one of the most incredible engineering efforts of all time.

Martin Rees (1:28:18.240)

That's an example where the engineers deserve most of the credit because the precision is,

Martin Rees (1:28:23.680)

as I said, it's like measuring the thickness of a hair at the distance of Alpha Centauri.

Martin Rees (1:28:28.960)

Yeah, it's incredible.

Lex Fridman (1:28:30.320)

Tens to minus 21.

Lex Fridman (1:28:31.440)

So maybe, actually, if we step back, what are black holes? What do we humans understand about

Lex Fridman (1:28:36.240)

black holes and what's still unknown?

Martin Rees (1:28:40.160)

Einstein's theory, extended by people like Roger Penrose, tells us that black holes are,

Martin Rees (1:28:47.120)

in a sense, rather simple things, basically, because they are solutions of Einstein's equations.

Lex Fridman (1:28:54.880)

And the thing that was shown in the 1960s by Roger Penrose in particular, and by a few other

Martin Rees (1:29:01.280)

people, was that a black hole, when it forms and settles down, is defined just by two quantities,

Martin Rees (1:29:09.040)

its mass and its spin. So they're actually very standardized objects. It's amazing that objects

Martin Rees (1:29:14.640)

as standardized as that can be so big and can lurk in the rest of the solar system.

Lex Fridman (1:29:20.560)

And so that's the situation for a ready formed black hole. But the way they form, obviously,

Martin Rees (1:29:27.120)

is very messy and complicated. And one of the things that I've worked on a lot is what the

Martin Rees (1:29:36.080)

phenomena are, which are best attributed to black holes, and what may lead to them, and all that.

Lex Fridman (1:29:43.040)

Which, can you explain to that? So what are the different phenomena

Martin Rees (1:29:46.320)

that lead to a black hole? Let's talk about it. This is so cool.

Lex Fridman (1:29:49.920)

Oh, okay. Well, I think one thing that only became understood, really,

Martin Rees (1:29:56.880)

in the 1950s, I suppose, and beyond was how stars evolve differently depending on how

Martin Rees (1:30:05.280)

heavy they are. The sun burns hydrogen to helium, and then when it's run out of that,

Martin Rees (1:30:11.360)

it contracts to be a white dwarf. And we know how long that will take. It'll take about 10 billion

Martin Rees (1:30:16.160)

years altogether for its lifetime. But big stars burn up their fuel more quickly, and more

Martin Rees (1:30:23.600)

interestingly, because when they've turned hydrogen to helium, they then get even hotter,

Lex Fridman (1:30:27.680)

so they can fuse helium into carbon and go up the periodic table. And then they eventually

Martin Rees (1:30:33.440)

explode when they have an energy crisis, and they blow out that process material,

Martin Rees (1:30:38.400)

which, as a digression, is crucially important because all the atoms inside our bodies

Martin Rees (1:30:44.880)

were synthesized inside a star, a star that lived and died more than five billion years ago

Martin Rees (1:30:50.640)

before our solar system formed. And so we each have inside us atoms made in thousands of different

Martin Rees (1:30:55.920)

stars all over the Milky Way, and that's an amazing idea. My predecessor, Fred Hoyle,

Martin Rees (1:31:00.800)

in 1946 was the first person to suggest that idea, and that's been borne out. That's a wonderful

Martin Rees (1:31:05.600)

idea. So that's how massive stars explode. And they leave behind something which is very exotic

Martin Rees (1:31:15.200)

of two kinds. One possibility is a neutron star, and these were first discovered in 1967,

Martin Rees (1:31:21.760)

68. These are stars a bit heavier than the sun, which are compressed to an amazing density,

Lex Fridman (1:31:29.040)

so the whole mass of more than the sun's mass is in something about 10 miles across.

Lex Fridman (1:31:34.720)

So they're extraordinarily dense, they're exotic physics, and they've been studied in immense

Martin Rees (1:31:46.880)

detail. And they've been real laboratories because the good thing about astronomy, apart from

Martin Rees (1:31:51.920)

exploring what's out there, is to use the fact that the cosmos has provided us with a lab

Martin Rees (1:31:57.040)

with far more extreme conditions than we could ever simulate. And so we learn lots of basic

Martin Rees (1:32:01.040)

physics from looking at these objects, and that's been true of neutron stars. But for black holes,

Martin Rees (1:32:06.320)

that's even more true because the bigger stars, when they collapse, they leave something behind

Martin Rees (1:32:14.160)

in the center which is too big to be a stable white 2 or 4 neutron star and becomes a black hole.

Lex Fridman (1:32:19.600)

And we know that there are lots of black holes weighing about 10 or up to 50 times as much as the

Martin Rees (1:32:25.520)

sun, which are the remnants of stars. They were detected first 50 years ago, when a black hole

Martin Rees (1:32:33.840)

was orbiting around another star and grabbing material from the other star which swirled into

Martin Rees (1:32:40.240)

it and gave us X rays. So the X rays astronomers found these objects orbiting around an ordinary

Martin Rees (1:32:48.960)

star and emitting X ray radiation very intensely, varying on a very short time scale. So something

Martin Rees (1:32:56.240)

very small and dense was giving that radiation. That was the first evidence for black holes.

Lex Fridman (1:33:00.800)

But then the other thing that happened was realizing that there was a different class

Martin Rees (1:33:05.680)

of monster black holes in the centers of galaxies. And these are responsible for what's called

Martin Rees (1:33:12.480)

quasars, which is when something in the center of a galaxy is grabbing some fuel and outshines all

Martin Rees (1:33:19.680)

the hundred billion stars or so in the rest of the galaxy. A giant beam of light. And in many

Martin Rees (1:33:26.640)

cases, it's a beam of... That's got to be the most epic thing the universe produces is quasars.

Martin Rees (1:33:35.200)

Well, it's a debate about what's the most epic, but quasars maybe or maybe gamma ray bursts or

Martin Rees (1:33:40.560)

something, but they are remarkable and they were a mystery for a long time. And they're

Martin Rees (1:33:45.840)

one of the things I worked on in my younger days. So even though they're so bright,

Martin Rees (1:33:50.160)

they're still a mystery and you can only see them. I think they're less of a mystery now.

Lex Fridman (1:33:54.960)

I think we do understand basically what's going on. How were quasars discovered?

Martin Rees (1:34:00.400)

Well, they were discovered when astronomers found things that looked like stars and that they were

Martin Rees (1:34:05.760)

small enough to be a point like and not resolved by a telescope, but outshone an entire galaxy.

Lex Fridman (1:34:13.440)

Yeah. That's suspicious.

Martin Rees (1:34:15.840)

Yes. But then they realized that what they were, they were objects which you now know are black

Martin Rees (1:34:25.280)

holes and black holes were capturing gas and that gas was getting very hot, but it was producing

Martin Rees (1:34:33.840)

far more energy than all the stars added together. And it was the energy of the black hole that was

Martin Rees (1:34:43.920)

lighting up all the gas in the galaxy. So you've got a spectrum of it there. So this was something

Martin Rees (1:34:50.000)

which was realized from the 1970s onwards. And as you say, the other thing we've learned is that

Martin Rees (1:34:56.160)

they often do produce these jets squirting out, which could be detected in all wave bands. So

Martin Rees (1:35:03.840)

there's now a standard picture.

Lex Fridman (1:35:05.040)

So there's a giant black hole generating jets of light at the center of most galaxies.

Martin Rees (1:35:10.480)

Yes. That's right.

Lex Fridman (1:35:11.200)

Do we know, do we have a sense if every galaxy has one of these big boys, big black holes?

Martin Rees (1:35:18.480)

Most galaxies have big black holes. They vary in size. The one in our galactic center.

Lex Fridman (1:35:23.840)

Do we know much about ours?

Martin Rees (1:35:25.120)

We do. Yes. We know it weighs about as much as 4 million suns, which is less than some,

Martin Rees (1:35:34.560)

it's several billion other galaxies. But we know this one in our galactic center

Martin Rees (1:35:40.560)

isn't very bright or conspicuous. And that's because not much is falling into it at the

Martin Rees (1:35:46.000)

moment. If a black hole is isolated, then of course it doesn't radiate. All that radiates

Martin Rees (1:35:52.560)

is gas swirling into it, which is very hot or has magnetic fields.

Lex Fridman (1:35:57.440)

It's only radiating the thing it's murdering or consuming, however you put it.

Martin Rees (1:36:01.840)

Yeah, that's right. And so it's thought that our galaxy may have been bright at some time

Martin Rees (1:36:07.440)

in the past, but now that's when the black hole formed or grew. But now it's not capturing

Martin Rees (1:36:16.320)

very much gas. And so it's rather faint and detected indirectly and by fairly weak radio

Martin Rees (1:36:24.320)

emission. And so I think the answer to your question is that we suspect that most galaxies

Martin Rees (1:36:31.520)

have a black hole in them. So that means at some stage in their lives, or maybe one or

Martin Rees (1:36:36.960)

more stages, they went through a phase of being like a quasar where that black hole

Martin Rees (1:36:41.680)

captured gas and became very, very bright. But for the rest of their lives, the black

Lex Fridman (1:36:47.680)

holes are fairly quiescent because there's not much gas falling into them.

Lex Fridman (1:36:53.120)

And so this universe of ours is sprinkled with a bunch of galaxies and giant black holes with

Martin Rees (1:37:01.840)

very large number of stars orbiting these black holes and then planets orbiting.

Martin Rees (1:37:08.400)

Likely, it seems like planets orbiting almost every one of those stars and just this beautiful

Martin Rees (1:37:17.680)

universe of ours. Well, what happens when galaxies collide, when these two big black holes

Martin Rees (1:37:24.560)

collide? Well, what would happen is that... Well, and I should say that this is going to happen

Martin Rees (1:37:32.800)

near us one day, but not for 4 billion years because the Andromeda galaxy, which is the

Martin Rees (1:37:37.840)

biggest galaxy near to us, which is about nearly 3 million light years away, which is a big disk

Martin Rees (1:37:44.640)

galaxy with a black hole in its hub, rather like our Milky Way. And that's falling towards us

Martin Rees (1:37:53.120)

because they're both in a common gravitational potential well. And that will collide with our

Martin Rees (1:37:59.200)

galaxy in about 4 billion years. But maybe it'll be less a collision and more of a dance because

Martin Rees (1:38:05.440)

it'll be like a swirling situation. But eventually, there'll be a merger. They'll

Martin Rees (1:38:10.960)

go through each other and then merge. In fact, there are nice movies to be made of this,

Martin Rees (1:38:16.880)

computer simulations, and it'll go through. And then there's a black hole in the center of

Martin Rees (1:38:26.000)

Andromeda and our galaxy. And the black holes will settle towards the center. Then they will

Martin Rees (1:38:34.640)

orbit around each other very fast, and then they will eventually merge. And that'll produce a big

Martin Rees (1:38:39.840)

burst of gravitational waves, a very big burst. That an alien civilization with a LIGO like

Martin Rees (1:38:45.760)

detector will be able to detect. Yes. Well, in fact, we can detect these

Martin Rees (1:38:50.240)

with their lower frequencies than the ways that will be detected by LIGO. So there's a

Martin Rees (1:38:57.440)

spatial interferometer which can detect these. It's about one cycle per hour rather than about

Martin Rees (1:39:04.480)

100 cycles per second. It's the ones that detect it. But that will happen. But thinking back to

Lex Fridman (1:39:11.680)

what will happen in 4 billion years to any of our descendants, they'll be okay because the two disk

Martin Rees (1:39:19.600)

galaxies will merge. It'll end up as a sort of amorphous elliptical galaxy. But the stars won't

Martin Rees (1:39:26.320)

be much closer together than they are now. It'll still be just twice as many stars in a structure

Martin Rees (1:39:32.800)

almost as big. And so the chance of another star colliding with our sun would still be very small.

Martin Rees (1:39:41.280)

Because there's actually a lot of space between stars and planets.

Martin Rees (1:39:45.120)

Yes, the chance of a star getting close enough to affect our solar system's orbit

Martin Rees (1:39:49.200)

is small. And it won't change that very much. So you could be reassured.

Lex Fridman (1:39:53.440)

A heck of a starry sky though. What would that look like?

Martin Rees (1:39:57.680)

Well, it won't make much difference even to that actually. It'll just be...

Martin Rees (1:40:01.600)

Wouldn't that look kind of beautiful when you're swirling? Oh, because it's swirling so slowly.

Martin Rees (1:40:05.600)

Yeah, but they're far away. So it'll be twice as many stars in the sky.

Lex Fridman (1:40:10.320)

Yeah, but the pattern changes in interesting ways.

Martin Rees (1:40:12.880)

Yeah, the pattern will change a bit. And there won't be the Milky Way because the Milky Way

Martin Rees (1:40:17.200)

across the sky is because we are looking in the disk of our galaxy. And you lose that.

Lex Fridman (1:40:22.400)

And because the disk will be so disrupted. And it'll be a more sort of spherical distribution.

Lex Fridman (1:40:28.800)

And of course, many galaxies are like that. And that's probably because they have been through

Martin Rees (1:40:33.680)

mergers of this kind. If we survive four billion years,

Lex Fridman (1:40:36.720)

we would likely be able to survive beyond that. Oh, yeah.

Martin Rees (1:40:39.520)

What's the other thing on the horizon for humans in terms of the sun burning out,

Lex Fridman (1:40:45.440)

all those kinds of interesting cosmological threats to our civilization?

Martin Rees (1:40:51.360)

Well, I think on the cosmological time scale, because it won't be humans, because even if

Martin Rees (1:40:57.920)

evolution is no faster than Darwinian, and I would argue it will be faster than Darwinian

Martin Rees (1:41:03.120)

in the future, then we're thinking about six billion years before the sun dies. So any

Martin Rees (1:41:09.760)

entities watching the death of the sun, if they're still around, they'll be as different

Martin Rees (1:41:14.320)

from much as we are from slime mold or something. And far more different still if they become

Martin Rees (1:41:20.640)

electronic. So on that time scale, we just can't predict anything. But I think going back to

Martin Rees (1:41:25.600)

the human time scale, then we've talked about whether there'll be people on Mars by the end

Martin Rees (1:41:32.880)

of a century. And even in these long perspectives, then indeed, this century is very special,

Martin Rees (1:41:39.760)

because it may see the transition between purely flesh and blood entities to those which are sort

Lex Fridman (1:41:45.760)

of cyborgs. And that'll be an important transition in the future.

Lex Fridman (1:41:51.040)

But of course, the other importance, and this has been the theme of a couple of my older books,

Martin Rees (1:41:55.760)

is that this is the first century when one species, namely our species, has the future of

Martin Rees (1:42:02.400)

the planet in its hands. And that's because of two types of concerns. One is that there are more

Martin Rees (1:42:10.800)

of us where more of us are living in the past, and the other is that there are more of us who are

Martin Rees (1:42:16.720)

living in the future. So those are two concerns. One is that there are more of us where more

Martin Rees (1:42:23.840)

demanding of energy and resources, and therefore we are for the first time changing the whole

Martin Rees (1:42:30.640)

planet through climate change, loss of biodiversity, and all those issues. This has never

Martin Rees (1:42:37.360)

happened in the past, because there haven't been enough humans. So this is an effect that's

Martin Rees (1:42:43.760)

now, and rightly so, because we've got to ensure that we leave a heritage that isn't eroded or

Lex Fridman (1:42:51.040)

damaged to future generations. And so that's one class of threats.

Lex Fridman (1:42:55.760)

But there's another thing that worries me, perhaps more than many people seem to worry,

Lex Fridman (1:43:02.800)

and that's the threat of misuse of technology. And so this is particularly because technologies

Martin Rees (1:43:11.520)

empower even small groups of malevolent people, or indeed even careless people, to create some

Martin Rees (1:43:21.920)

effect which could cascade globally. And to take an example, a dangerous pathogen or pandemic,

Martin Rees (1:43:30.400)

I mean, my worst nightmare is that there could be some small group that can engineer a virus

Martin Rees (1:43:42.160)

to make it more variant or more transmissible than a natural virus. This is a gain of function

Martin Rees (1:43:47.600)

experiments which were done on the flu virus 10 years ago and can be done for others.

Lex Fridman (1:43:51.840)

And of course, we now know from COVID 19 that our world is so interconnected that a disaster in one

Martin Rees (1:44:02.720)

part of the world can't be confined to that part and will spread globally. So it's possible for

Martin Rees (1:44:08.240)

a few dissidents with expertise in biotech could create a global catastrophe of that kind.

Lex Fridman (1:44:14.320)

And also, I think we need to worry about very large scale disruption by cyber attacks and

Martin Rees (1:44:21.600)

in fact, I quote in one of my books, a 2012 report from the American Pentagon about the possibility

Martin Rees (1:44:31.680)

of a state level cyber attack on the electricity grid in Eastern United States, which is it could

Martin Rees (1:44:38.720)

happen. And it says at the end of this chapter that this would merit a nuclear response. This

Martin Rees (1:44:45.680)

is a pretty scary possibility. And that was 10 years ago. And I think now what would have needed

Martin Rees (1:44:51.600)

a state actor then could be done perhaps by a small group empowered by AI. And so there's

Martin Rees (1:44:57.520)

obviously been an arms race between the cyber criminals and the cyber security people. Not

Martin Rees (1:45:06.080)

clear which side is winning. But the main point is that as we become more dependent on more

Martin Rees (1:45:11.600)

integrated systems, then we get more vulnerable. And so we have the knowledge, then the misuse of

Martin Rees (1:45:21.680)

that knowledge becomes more and more of a threat. And I'd say bio and cyber are the two biggest

Martin Rees (1:45:28.960)

concerns. And if we depend too much on AI and complex systems, then just breakdowns, it may be

Martin Rees (1:45:37.680)

that they break down. And even if it's an innocent breakdown, then it may be pretty hard to mend it.

Lex Fridman (1:45:43.600)

And just think how much worse the pandemic would have been if we'd lost the internet in the middle

Martin Rees (1:45:48.720)

of it. We'd be dependent more than ever for communication and everything else on the internet

Lex Fridman (1:45:55.440)

and Zooms and all that. And if that had broken down, that would have made things far worse. And

Martin Rees (1:46:01.360)

those are the kinds of threats that we, I think, need to be more energized and politicians need to

Martin Rees (1:46:07.520)

be more energized to minimize. And one of the things I've been doing in the last year through

Martin Rees (1:46:13.680)

being a member of our part of our parliament is I have to instigate a committee to think more on

Martin Rees (1:46:20.720)

better preparedness for extreme technological risks and things like that. So they're a big

Martin Rees (1:46:26.400)

concern in my mind that we've got to make sure that we can benefit from these advances but safely

Martin Rees (1:46:36.800)

because the stakes are getting higher and the benefits are getting great, as we know, huge

Martin Rees (1:46:42.720)

benefits from computers, but also huge downsides as well. And one of the things this war in Ukraine

Martin Rees (1:46:50.400)

has shown, one of the most terrifying things outside of the humanitarian crisis, is that at

Martin Rees (1:46:58.080)

least for me, I realized that the human capacity to initiate nuclear war is greater than I thought.

Martin Rees (1:47:11.120)

I thought the lessons of the past have been learned. It seems that we hang on the brink of

Martin Rees (1:47:17.120)

nuclear war with this conflict like every single day with just one mistake or bad actor or the

Martin Rees (1:47:28.560)

actual leaders of the particular nations launching a nuclear strike and all hell breaks loose. So

Martin Rees (1:47:35.760)

then add into that picture, cyber attacks and so on, that can lead to confusion and chaos. And then

Martin Rees (1:47:43.200)

out of that confusion, calculations are made such that a nuclear weapon is launched and then

Martin Rees (1:47:56.000)

you're talking about, I mean, it directs probably 60, 70% of humans on Earth are dead instantly.

Lex Fridman (1:48:05.520)

And then the rest, I mean, it's basically 99% of the human population is wiped out in the period

Martin Rees (1:48:14.320)

of five years. Well, it may not be that bad, but it will be a devastation for civilization,

Martin Rees (1:48:18.640)

of course. And of course, you were quite right that this could happen very quickly because of

Martin Rees (1:48:26.240)

information coming in and there's hardly enough time for human collected and careful thought.

Lex Fridman (1:48:34.560)

And there have been recorded cases of false alarms. There's several where there have been

Martin Rees (1:48:41.520)

suspected attacks from the other side. And fortunately, they've been really false alarm

Martin Rees (1:48:48.080)

soon enough, but this could happen. And there's a new class of threats actually, which in our

Martin Rees (1:48:53.440)

center in Cambridge, people are thinking about, which is that the commander control system

Martin Rees (1:49:00.400)

of the nuclear weapons and the submarine fleet and all that is now more automated and could be

Martin Rees (1:49:08.880)

subject to cyber attacks. And that's a new threat which didn't exist 30 years ago. And so I think

Martin Rees (1:49:16.800)

indeed, we're in a sort of scary world, I think. And it's because things happen faster and human

Martin Rees (1:49:26.800)

beings aren't in such direct and immediate control because so much is delegated to machines. And

Martin Rees (1:49:32.960)

also because the world is so much more interconnected, then some local event can cascade

Lex Fridman (1:49:40.720)

globally in a way it never could in the past and much faster.

Martin Rees (1:49:44.800)

Yeah, it's a double edged sword because the interconnectedness brings a higher quality of

Lex Fridman (1:49:54.000)

life across a lot of metrics.

Martin Rees (1:49:56.800)

Yeah, it can do. But of course, there again, I mean, if you think of supply chains where we

Martin Rees (1:50:02.000)

get stuff from around the world, then one lesson we've learned is that there's a trade off between

Martin Rees (1:50:06.960)

resilience and efficiency and it's resilient to have an inventory in stock and to depend on local

Martin Rees (1:50:16.240)

supplies, whereas they're more efficient to have long supply chains. But the risk there is that a

Martin Rees (1:50:24.160)

break in one link in one chain can screw up car production. This has already happened in the

Martin Rees (1:50:30.080)

pandemic. So there's a trade off. And there are other examples. I mean, for instance, the other

Martin Rees (1:50:35.120)

thing we learned was that it may be efficient to have 95% of your hospital intensive care beds

Martin Rees (1:50:42.000)

occupied all the time, which has been the UK situation, whereas to do what the Germans do and

Martin Rees (1:50:47.120)

always keep 20% of them free for an emergency is really a sensible precaution. And so I think

Martin Rees (1:50:53.200)

we've probably learned a lot of lessons from COVID 19. And they would include rebalancing the

Martin Rees (1:51:01.200)

trade off between resilience and efficiency.

Martin Rees (1:51:04.800)

Boy, the fact that COVID 19, a pandemic that could have been a lot, a lot worse,

Martin Rees (1:51:12.800)

brought the world to its knees anyway.

Martin Rees (1:51:15.360)

It could be far worse in terms of its fatality rate or something like that, of course.

Lex Fridman (1:51:20.800)

So the fact that that, I mean, it revealed so many flaws in our human institutions.

Lex Fridman (1:51:26.000)

And I think I'm rather pessimistic because I do worry about the baractol, a small group

Martin Rees (1:51:38.800)

who can produce catastrophe. And if you imagine someone with access to the kind of equipment that's

Martin Rees (1:51:46.720)

available in university labs or industrial labs, and they could create some dangerous pathogen,

Martin Rees (1:51:52.880)

then even one such person is too many. And how can we stop that? Because it's true that you can

Martin Rees (1:52:00.880)

have regulations. I mean, academies are having meetings, et cetera, about how to regulate these

Martin Rees (1:52:07.040)

new biological experiments, et cetera, make them safe. But even if you have all these regulations,

Martin Rees (1:52:13.280)

then enforcing regulations is pretty hopeless. We can't enforce the tax laws globally. We can't

Martin Rees (1:52:20.560)

enforce the drug laws globally. And so similarly, we can't readily enforce the laws against people

Martin Rees (1:52:28.720)

doing these dangerous experiments, even if all the governments say they should be prohibited.

Lex Fridman (1:52:33.680)

And so my line on this is that all nations are going to face a big trade off between

Martin Rees (1:52:40.480)

three things we value, freedom, security, and privacy. And I think different nations will

Martin Rees (1:52:52.080)

make that choice differently. The Chinese would give up privacy and have certainly more security,

Martin Rees (1:52:59.920)

if not more liberty. But I think in our countries, I think we're going to have to give up more

Martin Rees (1:53:08.800)

privacy in the same way. That's a really interesting trade off. But there's also something

Martin Rees (1:53:15.840)

about human nature here, where I personally believe that all humans are capable of good and

Martin Rees (1:53:21.600)

evil. And there's some aspect to which we can fight this by encouraging people, incentivizing people

Martin Rees (1:53:32.160)

towards the better angels of their nature. So in order for a small group of people to create,

Martin Rees (1:53:38.960)

to engineer deadly pathogens, you have to have people that, for whatever trajectory took them

Martin Rees (1:53:47.600)

in life, wanting to do that kind of thing. And if we can aggressively work on a world

Martin Rees (1:53:55.840)

that sort of sees the beauty in everybody and encourages the flourishing of everybody in terms

Martin Rees (1:54:03.360)

of mental health, in terms of meaning, in terms of all those kinds of things, that's one way

Martin Rees (1:54:08.320)

to fight the development of weapons that can lead to atrocities.

Martin Rees (1:54:14.320)

RL Yes, and I completely agree with that and to reduce the reason why people feel embittered.

Lex Fridman (1:54:19.360)

But of course, we've got a long way to go to do that. Because if you look at the present world,

Martin Rees (1:54:26.000)

nearly everyone in Africa has reason to feel embittered because their economic development

Martin Rees (1:54:34.720)

is lagging behind most of the rest of the world. And the prospects of getting out of

Martin Rees (1:54:39.760)

the poverty trap is rather bleak, especially as the population grows. Because for instance,

Martin Rees (1:54:46.000)

they can't develop like the Eastern Tigers by cheap manufacturing, because robots have taken

Martin Rees (1:54:50.160)

that over. So they naturally feel embittered by the inequality. And of course, what we need to have

Martin Rees (1:54:59.120)

is some sort of mega version of the Marshall Plan that helped Europe in the post World War II era

Martin Rees (1:55:06.400)

to enable Africa to develop. That would be not just an altruistic thing for Europe to do,

Lex Fridman (1:55:11.840)

but in our interest because otherwise, those in Africa will feel massively disaffected.

Lex Fridman (1:55:20.240)

And indeed, it's a manifestation of the excessive inequalities, the fact that the

Martin Rees (1:55:24.960)

2000 richest people in the world have enough money to double the income of the bottom billion.

Lex Fridman (1:55:30.640)

And that's an indictment of the ethics of the world. And this is where my friend Stephen Pinker

Lex Fridman (1:55:39.040)

and I have had some contact. We wrote joint articles on bio threats and all that. But he

Martin Rees (1:55:46.400)

writes these books, being very optimistic about quoting figures about how life expectancy has gone

Martin Rees (1:55:54.560)

up, infant mortality has gone down, literacy has gone up, and all those things. And he's quite

Martin Rees (1:55:58.480)

right about that. And so he says the world's getting better. Do you disagree with your

Martin Rees (1:56:04.320)

friends, Stephen Pinker? Well, I mean, I agree with those facts. But I think he misses out part

Martin Rees (1:56:11.680)

of the picture. Because there's a new class of threats, which hang over us now, which didn't

Martin Rees (1:56:19.120)

hang over us in the past. And I would also question whether we have collectively improved

Martin Rees (1:56:24.880)

our ethics at all. Because let's think back to the Middle Ages. It's true that, as Pinker says,

Martin Rees (1:56:30.640)

the average person was in a more miserable state than they are today, on average. For all the

Martin Rees (1:56:37.680)

reasons he quantifies, that's fine. But in the Middle Ages, there wasn't very much that could

Martin Rees (1:56:44.720)

have been done to improve people's lot in life because of lack of knowledge and lack of science,

Martin Rees (1:56:51.120)

etc. So the gap between the way the world was, which was pretty miserable, and the way the world

Martin Rees (1:56:57.200)

could have been, which wasn't all that much better, was fairly narrow. Whereas now, the gap

Martin Rees (1:57:03.520)

between the way the world is and the way the world could be is far, far wider. And therefore, I think

Martin Rees (1:57:09.520)

we are ethically more at fault in allowing this gap to get wider than it was in medieval times.

Lex Fridman (1:57:18.720)

And so I would very much question and dispute the idea that we are ethically in advance of

Martin Rees (1:57:26.160)

our predecessors collectively. That's a lot of interesting hypotheses in there. It's a fascinating

Martin Rees (1:57:33.200)

question of how much is the size of that gap between the way the world is and the way the world

Martin Rees (1:57:39.040)

could be is a reflection of our ethics. Or maybe sometimes it's just a reflection of a very large

Martin Rees (1:57:44.640)

number of people. Maybe it's a technical challenge too. It's not just... Well, of our political

Martin Rees (1:57:51.280)

systems. And we're trying to figure this thing out. There's 20th century, tried this thing that

Martin Rees (1:57:58.960)

sounded really good on paper of collective, the communism type of things. And it's like,

Martin Rees (1:58:05.600)

turned out at least the way that was done there, that leads to atrocities and the suffering and

Martin Rees (1:58:10.720)

the murder of tens of millions of people. Okay, so that didn't work. Let's try democracy.

Lex Fridman (1:58:16.400)

And that seems to have a lot of flaws, but it seems to be the best thing we got so far.

Lex Fridman (1:58:20.880)

So we're trying to figure this out as our technologies become more and more powerful,

Martin Rees (1:58:25.360)

have the capacity to do a lot of good to the world, but also unfortunately have the capacity

Martin Rees (1:58:30.160)

to destroy the entirety of the human civilization. And I think it's social media generally,

Martin Rees (1:58:35.840)

which makes it harder to get a sort of moderate consensus because in the old days when people got

Martin Rees (1:58:43.840)

their news filtered through responsible journalists in this country, the BBC and the

Martin Rees (1:58:49.680)

main newspapers, et cetera, they would muffle the crazy extremes. Whereas now, of course,

Martin Rees (1:58:56.480)

they're on the internet. And if you click on them, you get something still more extreme.

Lex Fridman (1:59:00.800)

And so I think we are seeing a sort of dangerous polarization, which I think is going to make all

Lex Fridman (1:59:06.240)

countries harder to govern. And that's something which I'm pessimistic about.

Martin Rees (1:59:09.920)

RL So to push back, it is true that brilliant people like you highlighting the limitations

Martin Rees (1:59:15.360)

of social media is making them realize the stakes and the failings of social media companies,

Lex Fridman (1:59:22.480)

but at the same time, they're revealing the division. It's not like they're creating it,

Martin Rees (1:59:26.560)

they're revealing it in part. And so that puts a lot of, that puts the responsibility

Martin Rees (1:59:34.160)

into the hands of social media and the opportunity in the hands of social media

Martin Rees (1:59:38.480)

to alleviate some of that division. So it could in the long arc of human history result. So bringing

Martin Rees (1:59:45.120)

some of those divisions and the anger and the hatred to the surface so that we can talk about

Martin Rees (1:59:52.000)

it. And as opposed to disproportionately promoting it, actually just surfacing it so we can get over

Martin Rees (1:59:57.680)

it. Well, you're assuming that the fat cats are more public spirited than the politicians.

Martin Rees (20:01.520)

it works better if we think they're not interacting particles than if you think they're a gas which

Martin Rees (20:05.840)

would have shockwaves and things. So we know something about the properties of these, but we

Martin Rees (20:09.760)

don't know what they are. And the disappointment compared to my guess 20 years ago is that

Martin Rees (20:17.120)

particles answering this description have not yet been found. It was thought that the

Martin Rees (20:21.440)

big accelerator, the Large Hadron Collider at CERN, which is the world's biggest,

Martin Rees (20:26.000)

might have found a new class of particles which would have been the obvious candidates.

Lex Fridman (20:30.560)

And it hasn't. And some people say, well, dark matter can't be there, etc. But what I would argue

Martin Rees (20:38.160)

is that there's a huge amount of parameter space that hasn't been explored. There are other kinds

Martin Rees (20:44.800)

of particles called axions which behave slightly differently, which are good candidates. And

Martin Rees (20:50.640)

there's a factor of 10 powers of 10 between the heaviest particles that could be created by

Martin Rees (20:58.000)

the Large Hadron Collider and the heaviest particles which on theoretical grounds could exist

Martin Rees (21:04.560)

without turning into black holes. So there's a huge amount of possible particles which could be

Martin Rees (21:09.920)

out there as remnants of the Big Bang, but which we wouldn't be able to detect so easily. So the

Martin Rees (21:16.160)

fact that we've got new constraints on what the dark matter could be doesn't diminish my belief

Martin Rees (21:22.320)

that it's there in the form of particles because we've only explored a small fraction of parameter

Martin Rees (21:27.120)

space. So there's this search, you're literally, upon an unintended, are searching in the dark

Martin Rees (21:36.480)

here in this giant parameter space of possible particles. You're searching for, I mean,

Martin Rees (21:41.360)

there could be all kinds of particles. There could be, and there's some which may be very,

Martin Rees (21:46.240)

very hard to detect. But I think we can hope for some new theoretical ideas because one point which

Martin Rees (21:53.920)

perhaps we'd like to discuss more is about the very early stages of the Big Bang.

Lex Fridman (21:58.800)

And the situation now is that we have an outline picture for how the universe has evolved from

Martin Rees (22:06.160)

the time when it was expanding in just a nanosecond up to the present. And we could do

Martin Rees (22:12.640)

that because after nanosecond, the physics of the material is in the same range that we can test in

Martin Rees (22:20.880)

the lab. After a nanosecond, the particles move around like those in the Large Hadron Collider.

Martin Rees (22:27.920)

If you wait for one second, they're rather like in the centers of the hottest stars. And nuclear

Martin Rees (22:34.000)

reactions produce hydrogen, helium, et cetera, which fit the data. So we can with confidence

Martin Rees (22:38.720)

extrapolate back to when the universe was a nanosecond old. Indeed, I think we can do it with

Martin Rees (22:43.200)

as much confidence as anything a geologist tells you about the early history of the Earth.

Lex Fridman (22:47.760)

And that's huge progress in the last 50 years. But any progress puts in sharper focus new mysteries.

Lex Fridman (22:56.080)

And of course, the new mysteries in this context are, why is the universe expanding the way it is?

Lex Fridman (23:02.880)

Why does it contain this mixture of atoms and dark matter and radiation? And why does it have

Lex Fridman (23:10.480)

the properties which allow galaxies to form, being fairly smooth but not completely smooth?

Lex Fridman (23:15.040)

And the answer to those questions, I generally believe, to lie in a much, much earlier stage

Martin Rees (23:21.440)

of the universe, when conditions were much more extreme, and therefore far beyond the stage where

Martin Rees (23:27.840)

we have the foothold in experiments. Very theoretical. And so we don't have a convincing

Martin Rees (23:33.440)

theory. We just have ideas. Until we have something like string theory or some other

Martin Rees (23:38.320)

clues to the ultra early universe, that's going to remain speculative. So there's a big gap.

Lex Fridman (23:44.400)

And to say how big the gap is, if we take the observable universe out of a bit more than 10

Martin Rees (23:50.880)

billion light years, then when the universe was a nanosecond old, that would have been squeezed down

Martin Rees (23:57.840)

to the size of our solar system, or compressed into that volume. But the times we're talking

Martin Rees (24:05.360)

about when the key properties of the universe were first imprinted were times when that entire

Martin Rees (24:11.600)

universe was squeezed down to the size of a tennis ball, or baseball if you prefer, and it emerged

Martin Rees (24:18.000)

from something microscopic. So it's a huge extrapolation. And it's not surprising that,

Martin Rees (24:24.000)

since it's so far from our experimental range of detectability, we are still groping for ideas.

Martin Rees (24:30.400)

TG But you think first theory will reach into that place, and then experiment will perhaps

Martin Rees (24:38.400)

one day catch up, maybe simulation. JG Well, I think in a sense it's a combination. I think

Lex Fridman (24:43.360)

what we hope for is that there'll be a theory which applies to the early universe,

Martin Rees (24:51.760)

which also has consequences which we can test in our present day universe, like discovering

Lex Fridman (25:00.400)

why neutrinos exist or things like that. And that's the thing which, as I mentioned,

Martin Rees (25:05.200)

we may perhaps need a bit of AI to help us to calculate. But I think the hope would be

Martin Rees (25:11.760)

that we will have a theory which applies under the very, very extreme early stages of the universe,

Lex Fridman (25:18.000)

but which gains credibility and gains confidence, because it also manages to account for otherwise

Martin Rees (25:23.440)

unexplained features of the low energy world, what people call the standard model of particle

Martin Rees (25:29.040)

physics, where there are lots of undetermined numbers. So it may help with that. TG So we're

Martin Rees (25:35.680)

dancing between physics and philosophy a little bit, but what do you think happened before the

Martin Rees (25:42.640)

Big Bang? So this feels like something that's out of the reach of science. JG It's out of the reach

Martin Rees (25:49.200)

of present science, because science develops and as the front is advanced, then new problems come

Martin Rees (25:55.920)

into focus that couldn't even have been postulated before. I mean, if I think of my own career,

Martin Rees (26:00.240)

when I was a student, the evidence for the Big Bang was pretty weak, whereas now it's extremely

Martin Rees (26:07.120)

strong. But we are now thinking about the reason why the universe is the way it is and all that.

Lex Fridman (26:14.400)

So I would put all these things we've just mentioned in the category of speculative science.

Lex Fridman (26:21.120)

And I don't see a bifurcation between that and philosophy. But of course, to answer your question,

Martin Rees (26:29.200)

if we do want to understand the very early universe, then we've got to realize that

Martin Rees (26:34.080)

it may involve even more counterintuitive concepts than quantum theory does, because it's a

Martin Rees (26:40.320)

condition even further away from everyday world than quantum theory is. And remember, our lives,

Martin Rees (26:47.120)

our brains evolved and haven't changed much since our ancestors roamed the African savanna and

Martin Rees (26:55.200)

looked at the everyday world. And it's rather amazing that we've been able to make some sense

Martin Rees (27:00.160)

of the quantum micro world and of the cosmos. But there may be some things which are beyond us. And

Martin Rees (27:06.000)

certainly, as you implied, there are things that we don't yet understand at all. And of course,

Martin Rees (27:11.520)

one concept we might have to jettison is the idea of three dimensions of space and time just ticking

Martin Rees (27:18.320)

away. There are lots of ideas. I mean, I think Stephen Hawking had an idea. They're talking about

Lex Fridman (27:26.880)

what happened before the Big Bang. It's like asking what happens if you go north from the

Martin Rees (27:30.960)

North Pole. It somehow closes off. That's just one idea. I don't like that idea, but that's a

Martin Rees (27:37.040)

possible one. And so we just don't know what happened at the very beginning of the Big Bang,

Martin Rees (27:45.440)

were there many Big Bangs rather than one, etc. And those are issues which we may be able to

Martin Rees (27:53.840)

get some foothold on from some new theory. But even then, we won't be able to directly

Martin Rees (28:04.080)

test the theories. But I think it's a heresy to think you have to be able to test every prediction

Martin Rees (28:10.880)

of a theory. Let me give you another example. We take seriously what Einstein's theory says

Martin Rees (28:16.960)

about the inside of black holes, even though we can't observe them, because that theory has been

Martin Rees (28:23.120)

vindicated in many other places in cosmology and black holes, gravitational waves, and all those

Martin Rees (28:28.880)

things. Likewise, if we had a theory which explains some things about the early history

Martin Rees (28:36.160)

of our Big Bang and the present universe, then we would take seriously the inference if it

Martin Rees (28:43.440)

predicted many Big Bangs, not one, even though we can't predict the other ones. So the example is

Martin Rees (28:49.040)

we can take seriously a prediction if it's the consequence of a theory that we believe

Martin Rees (28:54.880)

on other grounds. We don't need to be able to detect another Big Bang in order to take it

Martin Rees (29:00.880)

seriously. It may not be a proof, but it's a good indication that this is the direction where the

Lex Fridman (29:08.560)

truth lies. Yeah, if the theory is getting confidence in other ways. Where do you sense?

Lex Fridman (29:14.400)

Do you think there's other universes besides our own? There are sort of well defined theories which

Martin Rees (29:20.800)

make assumptions about the physics at the relevant time, and this time, incidentally,

Martin Rees (29:26.800)

is 10 to the power minus 36 seconds or earlier than that, so this tiny sliver of time.

Martin Rees (29:34.720)

There are some theories, a famous one due to Andrei Linde, the Russian cosmologist now at

Martin Rees (29:41.360)

Stanford, called eternal inflation, which did predict an eternal production of new Big Bangs,

Martin Rees (29:48.640)

as it were, and that's based on specific assumptions about the physics. But those

Martin Rees (29:54.800)

assumptions, of course, are just hypotheses which aren't vindicated. But there are other

Martin Rees (29:59.840)

theories which only predict one Big Bang. So I think we should be open minded and not dogmatic

Lex Fridman (2:00:02.560)

And I'm not sure about that.

Martin Rees (2:00:03.680)

RL I think there's a lot of money to be made in being publicly spirited. I think there's a lot of

Martin Rees (2:00:10.800)

money to be made in increasing the amount of love in the world, despite the sort of public perception

Martin Rees (2:00:17.040)

that all the social media companies heads are interested in doing is making money.

Martin Rees (2:00:23.680)

I think that may be true, but I just personally believe people being happy

Martin Rees (2:00:31.840)

is a hell of a good business model. And so making as many people happy, helping them flourish in a

Martin Rees (2:00:39.040)

long term way, that's a good way to make money.

Martin Rees (2:00:42.240)

RL Well, I think on the other hand, guilt and shame are good motives to make you behave better

Martin Rees (2:00:47.120)

in the future. That's my experience.

Martin Rees (2:00:49.760)

RL From maybe in the political perspective, certainly it's the case. But it does make sense

Martin Rees (2:00:56.480)

now that we can destroy ourselves with nuclear weapons, with engineered pandemics and so on,

Martin Rees (2:01:02.480)

that the aliens would show up. If I had a leadership position, maybe as a scientist

Martin Rees (2:01:14.560)

or otherwise in an alien civilization, and I would come upon Earth, I would try to watch from

Martin Rees (2:01:22.320)

a distance, do not interfere. And I would start interfering when these life forms start

Martin Rees (2:01:31.680)

becoming quite, have the capacity to be destructive. And so, I mean, it is an interesting

Martin Rees (2:01:38.720)

question when people talk about UFO sightings and all those kinds of things that at least...

Martin Rees (2:01:43.920)

RL These are benign aliens you're thinking of.

Martin Rees (2:01:45.920)

RL Benign, yes. I mean, they benign, almost curious, almost, partially, as with all curiosity,

Martin Rees (2:01:55.920)

partially selfish to try to observe, is there something interesting about this particular

Lex Fridman (2:02:00.720)

evolutionary system? Because I'm sure even to aliens, Earth is a curiosity.

Martin Rees (2:02:06.560)

RL Yeah. Well, it's in its very special stage.

Lex Fridman (2:02:09.760)

RL It's very special. Perhaps it's very short.

Martin Rees (2:02:12.560)

RL This century is very special among the 45 million centuries the Earth experienced already.

Lex Fridman (2:02:17.200)

So it is a very special time where they should be specially interested. But I think going back

Martin Rees (2:02:22.960)

to the politics, the other problem is getting people who have short term concerns to care

Martin Rees (2:02:31.600)

about the long term. By the long term, I now mean just looking 30 years or so ahead. I

Martin Rees (2:02:38.320)

know people who've been scientific advisors to governments and things, and they may make

Martin Rees (2:02:43.680)

these points, but of course, they don't have much traction because as we know very well,

Martin Rees (2:02:48.080)

any politician has an urgent agenda of very worrying things to deal with. And so they

Martin Rees (2:02:54.080)

aren't going to prioritize these issues, which are longer term and less immediate, and don't

Martin Rees (2:03:01.280)

just concern their constituents, they concern distant parts of the world. And so I think

Lex Fridman (2:03:08.080)

what we have to do is to enlist charismatic individuals to convert the public, because

Martin Rees (2:03:17.200)

if the politicians know the public care about something, climate change as an example, then

Martin Rees (2:03:25.840)

they will make decisions which take cognizance of that. And I think for that to happen, then

Martin Rees (2:03:34.400)

we do need some public individuals who are respected by everyone, and to have a high

Martin Rees (2:03:42.000)

profile. And in the climate context, I would say that I've mentioned four very disparate

Martin Rees (2:03:48.800)

people who've had such a big effect in the last few years. One is Pope Francis, the other is

Martin Rees (2:03:54.080)

David Attenborough, the other is Bill Gates, and the other is Greta Thornberg. And those

Martin Rees (2:03:58.800)

four people have certainly had a big shift in public opinion, and even changed the rhetoric

Martin Rees (2:04:07.040)

of business, although how deep that is, I don't know. But politicians can't let these

Martin Rees (2:04:14.960)

issues drop down off the agenda if there's a public clamor, and it needs people like that

Lex Fridman (2:04:21.600)

to keep the public clamor going.

Martin Rees (2:04:23.040)

To push back a little bit, so those four are very interesting, and I have deep respect

Martin Rees (2:04:27.440)

for them. They have, except David Attenborough, David Attenborough is really, I mean, everybody

Martin Rees (2:04:33.520)

loves him. I can't say anything. But Bill Gates and Greta, that also has created a lot

Martin Rees (2:04:40.480)

of division. And this is a big problem, so it's not just charismatic. I put that responsibility

Martin Rees (2:04:46.800)

actually on the scientific community and the politicians. So we need the charismatic leaders,

Lex Fridman (2:04:58.080)

and they're rare. When you look at human history, those are the ones that make a difference.

Martin Rees (2:05:03.200)

Those are the ones that, not deride, they inspire the populace to think long term. The

Martin Rees (2:05:13.520)

JFK will go to the moon in this decade, not because it's easy, but because it is hard.

Martin Rees (2:05:20.320)

There's no discussion about short term political gains or any of that kind of stuff in the

Martin Rees (2:05:30.320)

vision of going to the moon, or going to Mars, or taking on gigantic projects, or taking

Martin Rees (2:05:37.200)

on world hunger, or taking on climate change, or the education system, all those things

Lex Fridman (2:05:42.400)

that require long term significant investment. That requires...

Lex Fridman (2:05:47.360)

But it's hard to find those people. And incidentally, I think another problem, which is a downside

Martin Rees (2:05:53.440)

of social media, is that of younger people I know, the number who would contemplate a

Martin Rees (2:06:01.120)

political career has gone down because of the pressures on them and their family from

Martin Rees (2:06:07.440)

social media. It's a hell of a job now. And so I think we are all losers because the quality

Martin Rees (2:06:14.560)

of people who choose that path is really dropping. And as we see by the quality of those who

Lex Fridman (2:06:24.240)

are in these compositions.

Martin Rees (2:06:25.360)

That said, I think the silver lining there is the quality of the competition actually

Martin Rees (2:06:31.840)

is inspiring because it shows to you that there's a dire need of leaders, which I think

Martin Rees (2:06:39.200)

would be inspiring to young people to step into the fold. I mean, great leaders are not

Martin Rees (2:06:43.600)

afraid of a little bit of fire on social media. So if you have a 20 year old kid now, 25 year

Martin Rees (2:06:51.200)

old kid is seeing how the world has responded to the pandemic, seeing the geopolitical division

Martin Rees (2:06:57.680)

over the war in Ukraine, seeing the brewing war between the West and China. We need great

Martin Rees (2:07:03.520)

leaders and there's a hunger for them and the time will come when they step up. I believe

Martin Rees (2:07:11.520)

that. But also to add to your list of four, he doesn't get enough credit. I've been defending

Martin Rees (2:07:16.480)

him in this conversation, Elon Musk, in terms of the fight in climate change. But he also

Lex Fridman (2:07:22.880)

has led to a lot of division, but we need more David Edinburgh.

Martin Rees (2:07:27.120)

Yeah, no, no. I mean, I'm a fan. Definitely. I mean, I've heard him described as a 21st

Martin Rees (2:07:33.040)

century Brunel for his innovation and that's true. But whether he's an ethical inspiration,

Martin Rees (2:07:40.320)

I don't know.

Martin Rees (2:07:41.680)

Yeah, he has a lot of fun on Twitter. Well, let me ask you to put on your wise sage hat.

Lex Fridman (2:07:50.000)

What advice would you give to young people today? Maybe they're teenagers in high school,

Martin Rees (2:07:56.240)

maybe early college. What advice would you give to a career or have a life they can be

Lex Fridman (2:08:04.560)

proud of?

Martin Rees (2:08:05.360)

Yes. Well, I'd be very diffident, really, about offering any wisdom. But I think they

Martin Rees (2:08:14.480)

should realise that the choices they make at that time are important. And from the experience

Martin Rees (2:08:27.440)

I've had with many friends, many people don't realise that opportunities open until it's

Martin Rees (2:08:33.200)

too late. They somehow think that some opportunities are only open to a few privileged people and

Martin Rees (2:08:38.560)

they don't even try and that they could succeed. But if I focus on people working in some profession

Martin Rees (2:08:48.720)

I know about, like science, I would say pick an area to work in where new things are happening,

Martin Rees (2:08:56.000)

where you can do something that the old guys never had a chance to think about. Don't go

Martin Rees (2:09:02.640)

into a field that's fairly stagnant because then there's nothing much to do or you'll

Martin Rees (2:09:07.680)

be trying to tackle the problems that the old guys got stuck on. And so I think in science,

Martin Rees (2:09:13.120)

I can give people good advice that they should pick a subject where there are exciting new

Martin Rees (2:09:19.200)

developments. And also, of course, something which suits their style because even within

Martin Rees (2:09:24.560)

science, which is just one profession, there's a big range of style between the sort of solitary

Martin Rees (2:09:29.520)

thinker, the person who does field work, the person who works in a big team, et cetera,

Lex Fridman (2:09:34.080)

and whether you like computing or mathematical thought, et cetera. So pick some subject that

Martin Rees (2:09:40.560)

suits your style and where things are happening fast. And be prepared to be flexible. That's what

Martin Rees (2:09:46.960)

I'd say, really. Keep your eyes open for the opportunity throughout, like you said. Go to

Martin Rees (2:09:51.600)

a new field. Go to a field where new cool stuff is happening. Just keep your eyes open.

Martin Rees (2:09:56.240)

Yes, that's patitudinous. But I think most of us, and I include myself in this, didn't realize

Lex Fridman (2:10:01.840)

this sort of thing isn't too late.

Lex Fridman (2:10:05.840)

Yeah, I think this applies way beyond science. What do you make of this finiteness of our life?

Lex Fridman (2:10:13.920)

Do you think about death? Do you think about mortality? Do you think about your mortality?

Lex Fridman (2:10:18.560)

And are you afraid of death?

Martin Rees (2:10:20.320)

Well, I mean, I'm not afraid because I think I'm lucky. I feel lucky to have lasted as long as I

Martin Rees (2:10:25.600)

have and to have been fairly lucky in my life in many respects compared to most people. So I feel

Martin Rees (2:10:35.040)

very fortunate. This reminds me of this current emphasis on living much longer, the so called

Martin Rees (2:10:44.960)

Altos Laboratories, which have been set up by billionaires. There's one in San Francisco,

Martin Rees (2:10:53.680)

one in La Jolla, I think, and one in Cambridge. And they're funded by these guys who when young

Martin Rees (2:11:03.280)

wanted to be rich, and now they're rich, they want to be young again. They won't find that quite so

Martin Rees (2:11:07.840)

easy. And do we want this? I don't know. If there was some elite that was able to live much longer

Martin Rees (2:11:14.560)

than others, that would be a really fundamental kind of inequality. And I think if it happened

Martin Rees (2:11:23.040)

to everyone, then that might be an improvement. It's not so obvious. But I think for my part,

Martin Rees (2:11:32.800)

I think to have lived as long as most people and had a fortunate life is all I can

Lex Fridman (2:11:39.280)

expect and a lot to be grateful for. Those are all past issues.

Martin Rees (2:11:44.720)

Well, I am incredibly honored that you sit down with me today. I thank you so much for life,

Martin Rees (2:11:51.680)

of exploring some of the deepest mysteries of our universe and of our humanity and thinking

Martin Rees (2:11:57.360)

about our future with existential risks that are before us. It's a huge honor,

Lex Fridman (2:12:02.960)

Martin, that you sit with me and I really enjoyed it.

Martin Rees (2:12:06.240)

Well, thank you, Lex. I thought we couldn't go on for as long as this,

Lex Fridman (2:12:09.440)

but we could have gone on much longer.

Martin Rees (2:12:11.600)

Exactly. Thank you so much. Thank you for listening to this conversation with Martin

Martin Rees (2:12:16.480)

Rees. To support this podcast, please check out our sponsors in the description. And now,

Martin Rees (2:12:21.520)

let me leave you with some words from Martin Rees himself.

Martin Rees (2:12:24.720)

I'd like to widen people's awareness of the tremendous time spent lying ahead for our planet

Lex Fridman (2:12:30.640)

and for life itself. Most educated people are aware that we're the outcome of nearly 4 billion

Martin Rees (2:12:37.440)

years of Darwinian selection, but many tend to think that humans are somehow the culmination.

Martin Rees (2:12:44.880)

Our sun, however, is less than halfway through its lifespan. It will not be humans who watch

Martin Rees (2:12:51.360)

the sun's demise 6 billion years from now. Any creatures that then exist will be as different

Martin Rees (2:12:58.400)

from us as we are from bacteria or amoeba. Thank you for listening, and hope to see you next time.

Martin Rees (30:06.480)

about these options until we do understand the relevant physics. But there are these different

Martin Rees (30:12.880)

scenarios of very different ideas about this. But I think all of them have the feature that

Martin Rees (30:19.920)

physical reality is a lot more extensive than what we can see through our telescope. I think

Martin Rees (30:24.800)

even most conservative astronomers would say that because we can see out with our telescopes to a

Martin Rees (30:31.280)

sort of horizon, which is about, depending on how you measure it, maybe 15 billion light years away

Martin Rees (30:38.800)

or something like that. But that horizon of observations is no more physical reality than

Martin Rees (30:45.840)

the horizon around you if you're in the ocean and looking out at your horizon. There's no reason to

Martin Rees (30:53.760)

think that the ocean ends just beyond your horizon. And likewise, there's no reason to think that

Martin Rees (30:59.360)

the aftermath of our Big Bang ends just at the boundary of what we can see. Indeed,

Martin Rees (31:05.600)

there are quite strong arguments that it probably goes on about 100 times further.

Martin Rees (31:10.880)

It may even go on so much further that all combinatorials are replicated. And there's

Martin Rees (31:17.600)

another set of people like us sitting in a room like this. That's not logically impossible.

Lex Fridman (31:27.520)

But I think many people would accept that it does go on and contain probably a million times

Martin Rees (31:35.040)

as much stuff as what we can see within a horizon. The reason for that, incidentally,

Martin Rees (31:40.240)

is that if we look as far as we can in one direction and in the opposite direction,

Martin Rees (31:44.960)

then the conditions don't differ by more than one part in 100,000. So that means that if we're part

Martin Rees (31:50.640)

of some finite structure, the gradient across the part we can see is very small. And so that

Martin Rees (31:56.400)

suggests that it probably does go on a lot further. And the best estimates say it must

Lex Fridman (32:01.440)

go on at least 20 times further. LR – Is that exciting or terrifying to you?

Martin Rees (32:07.520)

Just the spans of it all, the wide, everything that lies beyond the horizon.

Martin Rees (32:15.360)

That example doesn't even hold for Earth, so it goes way, way farther. And on top of that,

Martin Rees (32:20.080)

just to take your metaphor further on the ocean, while we're on top of this ocean,

Martin Rees (32:25.840)

not only can we not see beyond the horizon, we also don't know much about the depth of the ocean,

Martin Rees (32:31.760)

nor the actual mechanism of observation that's in our head.

Martin Rees (32:36.560)

RL – I don't think the perception of this utterly vast cosmos need have any deeper impact

Martin Rees (32:59.360)

on us than just realizing that we are very small on the scale of the external world.

Martin Rees (33:04.000)

LR – Yeah, it's humbling though. It's humbling, depending where your ego is, it's humbling.

Martin Rees (33:13.120)

RL – Well, if you start off very unhumble indeed, it may make a difference. But for most of us,

Martin Rees (33:17.680)

I don't think it makes much of a difference. Well, there's a more general question, of course,

Martin Rees (33:22.720)

about whether the human race as such is something which is very special, or if on the other hand,

Martin Rees (33:34.400)

it's just one of many such species elsewhere in the universe, or indeed existing at different

Martin Rees (33:42.080)

times in our universe. LR – To me, it feels almost obvious

Martin Rees (33:48.160)

that the universe should be full of alien life, perhaps dead alien civilizations, but

Martin Rees (33:54.560)

just the vastness of space. And it just feels wrong to think of Earth as somehow special. It

Martin Rees (34:04.160)

sure as heck doesn't look that special. The more we learn, the less special it seems.

Martin Rees (34:09.680)

RL – Well, I mean, I don't agree with that as far as life is concerned, because remember that

Martin Rees (34:16.560)

we don't understand how life began here on Earth. We don't understand, although we know there were

Martin Rees (34:22.240)

any evolution of simple life to complex life, we don't understand what caused the transition

Martin Rees (34:28.160)

between complex chemistry and the first replicating, metabolizing entity we call alive.

Martin Rees (34:35.360)

That's a mystery, and serious physicists and chemists are now thinking about it,

Lex Fridman (34:42.720)

but we don't know. So we therefore can't say, was it a rare fluke which would not have happened

Martin Rees (34:50.400)

anywhere else, or was it something which involves a process that would have happened in any other

Martin Rees (34:57.200)

planet where conditions were like they were on the young Earth? So we can't say that now. I think

Martin Rees (35:04.880)

many of us would indeed bet that probably some kind of life exists elsewhere. But even if you

Martin Rees (35:11.920)

accept that, then there are many contingencies going from simple life to present day life.

Martin Rees (35:18.960)

Some biologists like Stephen Jay Gould thought that if you reround evolution,

Martin Rees (35:24.720)

you'd end up with something quite different, and maybe not with an intelligent species. So

Martin Rees (35:28.800)

the contingencies in evolution may militate against the emergence of intelligence,

Martin Rees (35:35.520)

even if life gets started in lots of places. So I think these are still completely open questions,

Lex Fridman (35:40.320)

and that's why it's such an exciting time now that we are starting to be able to address these.

Martin Rees (35:45.120)

I mentioned the fact that the origin of life is a question that we may be able to understand,

Lex Fridman (35:52.720)

and serious people are working on it. It used to be put in a sort of too difficult box. Everyone

Martin Rees (35:56.720)

knew it was important, but they didn't know how to tackle it or what experiments to do.

Lex Fridman (36:00.240)

But it's not like that now. That's partly because of clever experiments, but I think most

Martin Rees (36:06.240)

importantly because we are aware that we can look for life in other places, other places in our

Martin Rees (36:13.120)

solar system, and of course, on the exoplanets around other stars. And within 10 or 20 years,

Martin Rees (36:20.800)

I think two things could happen, which would be really, really important. We might,

Martin Rees (36:25.840)

with the next big telescope, be able to image some of the Earth like planets around other stars.

Lex Fridman (36:32.960)

TG. Image, like get a picture?

Martin Rees (36:36.240)

MR. Well, actually, caveat that, it'd take 50 years to get a resolved image,

Lex Fridman (36:41.600)

but try to detect the light. Because now, of course, these exoplanets are detected by their

Martin Rees (36:47.280)

effects on the parent star. They either cause their parent star to dim slightly when they

Martin Rees (36:52.720)

transit across in front of it, and so we see the dips, or their gravitational pull makes the star

Martin Rees (36:58.880)

wobble a bit. So most of the 5,000 plus planets that have been found around other stars, they've

Martin Rees (37:05.440)

been found indirectly by their effect in one of those two ways on the parent star.

Martin Rees (37:09.440)

TG. You can still do a pretty good job estimating size, all those kinds of things.

Martin Rees (37:13.600)

MR. The size and the mass, you can estimate. But detecting the actual light from one of these

Martin Rees (37:22.160)

exoplanets hasn't really been done yet, except for one or two very bright, big planets.

Lex Fridman (37:27.600)

TG. So maybe like James Webb Telescope.

Martin Rees (37:29.360)

MR. Well, James Webb may do this, but even better will be the European ground based telescope

Martin Rees (37:36.000)

called Unimaginatively Extremely Large Telescope, which has a 39 meter diameter mirror. 39 meters,

Martin Rees (37:42.480)

a mosaic of eight hundred sheets of glass, and that will collect enough light from one of these

Martin Rees (37:49.920)

exoplanets around a nearby star to be able to separate out its light from that of the star,

Martin Rees (37:57.760)

which is millions of times brighter, and get the spectrum of the planet and see if it's got

Martin Rees (38:03.280)

oxygen or chlorophyll and things in it. So that will come. James Webb may make some steps there.

Lex Fridman (38:11.600)

But I think we can look forward to learning quite a bit in the next 20 years, because I like to say,

Martin Rees (38:18.240)

supposing that aliens were looking at the solar system. Then they'd see the sun as an ordinary

Martin Rees (38:24.320)

star. They'd see the Earth as, in Carl Sagan's nice phrase, a pale blue dot lying very close in

Martin Rees (38:31.120)

the sky to its star, our sun, and much, much, much fainter. But if they could observe that dot,

Martin Rees (38:38.960)

they could learn quite a bit. They could perhaps get the spectrum of the light and find the

Martin Rees (38:44.320)

atmosphere. They'd find the shade of blue was slightly different, depending on whether the

Martin Rees (38:49.120)

Pacific Ocean or the land mass of Asia was facing them, so they could infer the length of the day

Lex Fridman (38:54.720)

and the ocean and continents, and maybe something about the seasons and the climate.

Martin Rees (39:00.800)

That's the kind of calculation and inference we might be able to draw within the next 10 or 20

Martin Rees (39:08.160)

years about other exoplanets. Evidence of some sort of biosphere on one of them would, of course,

Martin Rees (39:16.560)

be crucial, and it would rule out the still logical possibility that life is unique.

Lex Fridman (39:22.080)

But there's another way in which this may happen in the next 20 years. People think there could be

Martin Rees (39:26.080)

something swimming under the ice of Europa and Enceladus, and probes are being sent to maybe not

Martin Rees (39:33.600)

quite go under the ice but detect the spray coming out to see if there's evidence for organics in

Martin Rees (39:39.360)

that. And if we found any evidence for an origin of life that had happened in either of those

Martin Rees (39:47.360)

places, that would immediately be important. Because if life has originated twice independently

Martin Rees (39:54.640)

in one planetary system, the solar system, that would tell us straight away it wasn't a rare

Martin Rees (39:59.280)

accident and must have happened billions of times in the galaxy. At the moment, we can't rule out

Martin Rees (40:05.920)

it being unique. And incidentally, if we found life on Mars, then that would still be ambiguous

Martin Rees (40:11.600)

because people have realized that this early life could have got from Mars to Earth or vice versa

Martin Rees (40:18.080)

on meteorites. So if you found life on Mars, then some skeptics could still say if it's a single

Martin Rees (40:24.480)

origin. But I think that's far enough away statistically. So that's why that would be

Martin Rees (40:31.360)

especially... It's always the skeptics that ruin a good party. But we need them, of course. We need

Martin Rees (40:38.560)

them at the party. We need some skeptics at the party. But boy, would that be so exciting to find

Martin Rees (40:44.800)

life on one of the moons. Because it means that life is everywhere. That'll just be any kind of

Martin Rees (40:52.160)

vegetation or life. The question of the aliens of science fiction is a different matter.

Martin Rees (40:58.160)

LW. Intelligent aliens. Yeah, but if you have a good indication that there's life elsewhere in

Martin Rees (41:05.200)

the solar system, that means life is everywhere. I don't know if that's terrifying or what that is.

Martin Rees (41:15.040)

Because if life is everywhere, why is intelligent life not everywhere? You've talked about that

Martin Rees (41:22.000)

most likely alien civilizations, if they are out there, they would likely be far ahead of us.

Martin Rees (41:29.760)

The ones that would actually communicate with us. And that, again, one of those things is both

Martin Rees (41:38.800)

exciting and terrifying. You've mentioned that they're likely not to be of biological nature.

Martin Rees (41:44.080)

MR. Well, I think that's important. Of course, again, it's speculation. But in speculating about

Martin Rees (41:51.840)

intelligent life, and I take the search seriously. In fact, I chair the committee that the Russian

Martin Rees (41:59.200)

American investor Yuri Milner supports looking for intelligent life. He's putting $10 million

Martin Rees (42:05.840)

a year into better equipment and getting time on telescopes to do this. And so I think it's

Martin Rees (42:11.120)

worthwhile, even though I don't hold my breath for success. It's very exciting. But that does

Martin Rees (42:18.000)

lead me to wonder what might be detected. And I think, well, we don't know. We've got to be

Martin Rees (42:24.480)

open minded about anything. We have no idea what it could be. And so any anomalous objects,

Martin Rees (42:28.800)

or even some strange shiny objects in the solar system, or anything we've got to keep our eyes

Martin Rees (42:34.000)

open for. But I think if we ask about a planet like the Earth, where evolution had taken more

Martin Rees (42:44.240)

of the same track, then, as you say, it wouldn't be synchronized. If it had lagged behind, then,

Martin Rees (42:51.120)

of course, it would not have got to advanced life. But it may have had a head start. It may

Martin Rees (42:57.840)

have formed on a planet around an older star. But then let's ask what we would see. It's taken

Martin Rees (43:04.960)

nearly four billion years from the first life to us. And we now got this technological civilization

Martin Rees (43:11.600)

which could make itself detectable to any aliens out there. But I think most people would say

Martin Rees (43:21.840)

that this civilization of flesh and blood creatures in a collective civilization may not last more

Martin Rees (43:29.280)

than a few hundred years more. I think that some people would say it will kill itself off. But I'm

Martin Rees (43:38.960)

more optimistic. And I would say that what we're going to have in future is no longer the slow

Martin Rees (43:47.200)

Darwinian selection. But we're going to have what I call secular intelligent design, which will be

Martin Rees (43:54.640)

humans designing their progeny to be better adapted to where they are.

Lex Fridman (44:02.640)

And if they go to Mars or somewhere, they're badly adapted and they want to adapt a lot.

Lex Fridman (44:09.280)

And so they will adapt. But there may be some limits to what could be done with flesh and blood.

Lex Fridman (44:15.760)

And so they may become largely electronic, download their brains and be electronic entities.

Lex Fridman (44:23.840)

And if they're electronic, then what's important is that they're near immortal.

Lex Fridman (44:30.640)

And also, they won't necessarily want to be on a planet with an atmosphere or gravity. They may

Martin Rees (44:36.240)

go off into the blue yonder. And if they're near immortal, they won't be daunted by interstellar

Martin Rees (44:41.200)

travel taking a long time. And so if we looked at what would happen on the Earth in the next

Martin Rees (44:51.200)

millions of years, then there may be these electronic entities, which have been sent out

Lex Fridman (44:58.320)

and are now far away from the Earth, but still sort of burping away in some fashion to be detected.

Lex Fridman (45:04.640)

And so this therefore leads me to think that if there was another planet which had evolved like

Martin Rees (45:13.920)

the Earth and was ahead of us, it wouldn't be synchronized, so we wouldn't see a flesh and

Martin Rees (45:19.680)

blood civilization, but we would see these electronic progeny, as it were. And then this

Martin Rees (45:26.240)

raises another question, because there's the famous argument against there being lots of

Martin Rees (45:33.600)

aliens out there, which is that they would come and invade us and eat us or something like that.

Lex Fridman (45:39.600)

That's a common idea, which Fermi is attributed to have been the first to say.

Lex Fridman (45:46.160)

And I think there's an escape clause to that, because these entities would be

Martin Rees (45:54.800)

evolved by second intelligent design, designed by their predecessors and then designed by us.

Martin Rees (45:59.920)

Whereas Darwinian selection requires two things. It requires aggression and intelligence. This

Martin Rees (46:11.120)

future intelligent design may favor intelligence, because that's what they were designed for,

Lex Fridman (46:16.960)

but it may not favor aggression. And so these future entities, they may be sitting deep thoughts,

Martin Rees (46:23.760)

they're thinking deep thoughts, and not being at all expansionist. So they could be out there.

Lex Fridman (46:32.240)

And we can't refute their existence in the way the Fermi paradox is supposed to refute their

Lex Fridman (46:38.000)

existence, because these would not be aggressive or expansionist.

Martin Rees (46:41.520)

Well, maybe evolution requires competition, not aggression. And I wonder if competition

Martin Rees (46:46.880)

can take forms that are non expansionary. So you can still have fun competing in the space

Martin Rees (46:53.040)

of ideas, which maybe primarily...

Lex Fridman (46:56.240)

The Darwin philosophers, perhaps, yeah.

Martin Rees (46:59.440)

In a way, right. It's an intellectual exercise versus a sort of violent exercise.

Lex Fridman (47:07.840)

So what does this civilization on Mars look like? So do you think we would more and more

Martin Rees (47:15.280)

maybe start with some genetic modification and then move to basically cyborgs,

Lex Fridman (47:20.560)

increasing integration of electronic systems, computational systems into our bodies and brains?

Martin Rees (47:25.920)

This is a theme of my other new book out this year, which is called The End of Astronauts.

Lex Fridman (47:31.920)

The End of Astronauts.

Martin Rees (47:33.040)

It's co written with my old friend and colleague from Berkeley, Don Goldsmith. And it's really

Martin Rees (47:40.320)

about the role of human spaceflight versus sort of robotic spaceflight. And just to summarize

Lex Fridman (47:48.400)

what it says, it argues that the practical case for sending humans into space is getting

Martin Rees (47:56.400)

weaker all the time as robots get better and more capable. Robots 50 years ago couldn't

Martin Rees (48:02.960)

do anything very much, but now they could assemble big structures on space or in space

Martin Rees (48:08.560)

or on the moon, and they could probably do exploration. But present ones on Mars can't

Martin Rees (48:17.360)

actually do the geology, but future AI will be able to do the geology and already they

Martin Rees (48:24.080)

can dig on Mars. And so if you want to do exploration of Mars, and of course, even more

Martin Rees (48:31.360)

of Enceladus or Europa where you could never send humans, we depend on robots. And they're

Martin Rees (48:37.840)

far, far cheaper because to send a human to Mars requires feeding them for 200 days on

Martin Rees (48:43.760)

the journey there and bringing them back. And neither of those are necessary for robots.

Lex Fridman (48:48.000)

So the practical case for humans is getting very, very weak. And if humans go, it's only

Martin Rees (48:54.000)

as an adventure, really. And so the line in our book is that human spaceflight should

Martin Rees (49:03.360)

not be pursued by NASA or public funding agencies because it has no practical purpose, but also

Martin Rees (49:12.800)

because it's especially expensive if they do it because they would have to be risk averse

Martin Rees (49:19.440)

in launching civilians into space. I can illustrate that by noting that the shuttle

Martin Rees (49:27.200)

was launched 135 times and it had two spectacular failures, which each killed the seven people in

Martin Rees (49:35.040)

the crew. And it had been mistakenly presented as safe for civilians. And there was a woman

Martin Rees (49:42.800)

schoolteacher killed in one of them. It was a big national trauma and they tried to make

Martin Rees (49:46.800)

it safer still. But if you launch into space, just the kind of people prepared to accept that

Martin Rees (49:54.800)

sort of risk, and of course, test pilots and people who go hang gliding and go to the South

Martin Rees (50:00.400)

Pole, et cetera, are prepared to accept a 2% risk at least for a big challenge, then of course,

Martin Rees (50:06.800)

you do it more cheaply. And that's why I think human spaceflight should be left to the billionaires

Lex Fridman (50:16.240)

and their sponsors because then the taxpayers aren't paying and they can launch simply those

Martin Rees (50:22.960)

people who are prepared to accept high risks. Space adventure, not space tourism. And we should

Martin Rees (50:31.440)

cheer them on. And as regards where they would go, then low Earth orbit, as I suspect, can be done

Martin Rees (50:40.480)

quite cheaply in the future. But going to Mars, which is very, very expensive and dangerous for

Martin Rees (50:46.640)

humans. The only people who would go would be these adventurers, maybe on one way trip,

Martin Rees (50:55.360)

like some of the early polar explorers and Magellan and people like that. And we would

Martin Rees (51:00.160)

cheer them on. And I expect and I've ever hoped that by the end of a century, there will be a

Martin Rees (51:07.920)

small community of such people on Mars living very uncomfortably, far less comfortably than

Martin Rees (51:16.000)

at the South Pole or the bottom of the ocean or the top of Everest. But they will be there

Lex Fridman (51:21.520)

and they won't have a return ticket, but they'll be there. Incidentally, I think it's a dangerous

Martin Rees (51:29.040)

illusion to think, as Elon Musk has said, that we can have mass emigration from the Earth to Mars

Martin Rees (51:38.640)

to escape the Earth's problems. It's a dangerous illusion because it's far easier to deal with

Martin Rees (51:44.960)

climate change on Earth than to terraform Mars to make it probably habitable to humans.

Lex Fridman (51:51.520)

And so there's no planet B for ordinary risk averse people. But for these crazy adventurers,

Martin Rees (51:56.240)

then you can imagine that they would be trying to live on Mars as great pioneers.

Lex Fridman (52:03.040)

And by the end of a century, then there will be huge advances compared to the present in two

Martin Rees (52:09.040)

things. First, in understanding genetics, so as to genetically redesign one's offspring. And

Martin Rees (52:16.320)

secondly, to use cyborg techniques to implant something in our brain or indeed think about

Martin Rees (52:23.600)

downloading, et cetera. And those techniques will, one hopes, be heavily regulated on Earth on

Martin Rees (52:30.080)

prudentials and ethical grounds. And of course, we are pretty well adapted to the Earth, so we

Martin Rees (52:36.000)

don't have the incentive to do these things in the way they were there. So our argument is that

Martin Rees (52:42.560)

it'll be those crazy pioneers on Mars using all these scientific advances, which will be

Martin Rees (52:51.040)

controlled here, away from the regulators, they will transition into a new post human species.

Lex Fridman (52:58.720)

And so if they do that, and if they transition into something which is electronic,

Martin Rees (53:04.800)

eventually, because there may be some limits to the capacity of flesh and blood brains anyways,

Martin Rees (53:10.240)

then those electronic entities may not want to stay on a planet like Mars, they may want to go

Martin Rees (53:16.400)

away. And so they'll be the precursors of the future evolution of life and intelligence coming

Martin Rees (53:24.400)

from the Earth. And of course, there's one point which perhaps astronomers are more aware of than

Martin Rees (53:29.840)

most people. Most people are aware that we are the outcome of 4 billion years of evolution.

Martin Rees (53:36.800)

Most of them nonetheless, probably think that we humans are somehow the culmination,

Martin Rees (53:42.960)

the top of the tree. But yes, no astronomers can believe that because astronomers know

Martin Rees (53:49.760)

that the Earth is four and a half billion years old. The sun has been shiny for that length of

Martin Rees (53:56.240)

time. But the sun has got 6 billion years more to go before it flares up and engulfs the inner

Martin Rees (54:01.760)

planet. So the sun is less than halfway through its life. And the expanding universe goes on far

Martin Rees (54:08.160)

longer still, maybe forever. And I like to quote Woody Allen, who said, eternity is very long,

Martin Rees (54:13.040)

especially towards the end. So we shouldn't think of ourselves as maybe even the halfway stage

Martin Rees (54:19.840)

in the emergence of cosmic complexity. And so these entities who are postcursors,

Martin Rees (54:27.920)

they will go beyond the solar system. And of course, even if there's nothing else out there

Martin Rees (54:32.960)

already, then they could populate the rest of the galaxy. And maybe eventually meet the others who

Martin Rees (54:40.800)

are out there expanding as well. Expanding, populating with expanded capacity for life

Lex Fridman (54:48.480)

and intelligence, all those kinds of things. Well, they might. But again, all better off.

Martin Rees (54:55.440)

We can't conceive what they'd be like. They won't be green men and women with eyes on stalks.

Martin Rees (55:03.600)

Maybe something quite different. We just don't know. But there is an interesting question,

Martin Rees (55:08.560)

actually, which comes up when I've sometimes spoken to audiences about this topic,

Lex Fridman (55:12.480)

but the question of consciousness and self awareness. Because going back to philosophical

Lex Fridman (55:17.680)

questions, whether an electronic robot would be a zombie, or would it be conscious and self aware?

Lex Fridman (55:27.280)

And I think there's no way of answering this empirically. And some people think that

Martin Rees (55:34.400)

consciousness and self awareness is an emergent property in any sufficiently complicated networks

Martin Rees (55:39.680)

that they would be. Others say, well, maybe it's something special to the flesh and blood that we're

Martin Rees (55:44.720)

made of. We don't know. And in a sense, this may not matter to the way things behave because

Martin Rees (55:52.720)

they could be zombies and still behave as though they were intelligent. But I remember

Martin Rees (55:59.040)

after one of my talks, someone came up and said, wouldn't it be sad if these future entities,

Martin Rees (56:06.960)

which were the main intelligent in the universe, had no self awareness? So there was nothing

Martin Rees (56:11.680)

which could appreciate the wonder and mystery of the universe and the beauty of the universe

Martin Rees (56:17.440)

in the way that we do. And so it does perhaps affect one's perspective of whether you welcome

Martin Rees (56:24.880)

or deplore this possible future scenario, depending on whether you think the future

Martin Rees (56:30.800)

post human entities are conscious and have an aesthetic sense or whether they're just zombies.

Lex Fridman (56:36.800)

And of course, you have to be humble to realize that self awareness may not be the

Martin Rees (56:45.360)

highest form of being, that humans have a very strong ego and a very strong sense of identity,

Martin Rees (56:54.480)

like personal identity connected to this particular brain. It's not so obvious to me

Martin Rees (57:00.240)

that that is somehow the highest achievement of a life form, that maybe this kind of...

Lex Fridman (57:08.800)

Do you think something collective would be?

Martin Rees (57:10.720)

It's possible that, well, I think from an alien perspective, when you look at Earth,

Martin Rees (57:17.200)

it's not so obvious to me that individual humans are the atoms of intelligence. It could be the

Martin Rees (57:24.080)

entire organism together, the collective intelligence. And so we humans think of

Martin Rees (57:28.320)

ourselves as individuals, we dress up, we wear ties and suits, and we'll give each other prizes.

Lex Fridman (57:33.280)

But in reality, the intelligence, the things we create that are beautiful emerges from our

Martin Rees (57:39.920)

interaction with each other. And that may be where the intelligence is, ideas jumping from

Lex Fridman (57:44.960)

one person to another over generations.

Martin Rees (57:47.360)

Yes, but we have experiences where we can appreciate beauty and wonder and all that.

Lex Fridman (57:54.800)

And a zombie may not have those experiences.

Martin Rees (57:59.840)

Yeah, or it may have a very different, a very black and white,

Martin Rees (58:04.080)

harsh description of a philosophical zombie that could be just a very different way to experience.

Martin Rees (58:13.920)

And, you know, in terms of the explorers that colonize Mars,

Martin Rees (58:18.800)

I mean, there's several things I want to mention. One, it's just at a high level. To me,

Martin Rees (58:25.200)

that's one of the most inspiring things humans can do, is reach out into the unknown. That's

Lex Fridman (58:31.120)

in the space of ideas, in the space of science, but also the explorers.

Martin Rees (58:34.640)

Yes, no, I agree with that.

Lex Fridman (58:35.920)

And that inspires people here on Earth more. I mean, it did in their, you know,

Martin Rees (58:42.400)

when going to the moon or going out to space in the 20th century, that inspired me.

Martin Rees (58:46.320)

I think that also could be used to inspire a generation of new scientists in the 21st century

Martin Rees (58:54.160)

by reaching out towards Mars. So in that sense, I think what Elon Musk and others are doing is

Martin Rees (59:00.160)

actually quite inspiring. It's not, it's not a recreational thing. It's actually has a deep

Martin Rees (59:07.200)

humanitarian purpose of really inspiring the world. And then, you know, I think, you know,

Martin Rees (59:12.880)

I don't think Elon says we want to escape Earth's problems. It's more that we should allocate some

Martin Rees (59:22.240)

small percentage of resources to have a backup plan. And because you yourself have spoken about

Lex Fridman (59:30.880)

and written about all the ways we clever humans destroy ourselves.

Martin Rees (59:35.680)

You could go on.

Lex Fridman (59:36.320)

And I'm not sure, it does seem when you look at the long arc of human history, it seems almost

Martin Rees (59:45.680)

obvious that we need to become a multi planetary species over a period. If we are to survive many

Martin Rees (59:51.840)

centuries, it seems that as we get clever and clever with the ways we can destroy ourselves,

Martin Rees (59:59.440)

Earth is going to become less and less safe. So in that sense, this is one of the things,