Sean Carroll: The Nature of the Universe, Life, and Intelligence
物理与宇宙学生物与进化音乐与艺术哲学与宗教AI 与机器学习
📋 章节目录
暂无章节信息
🔑 关键词
universedonintelligentspacepodcastintelligenceunderstandinghumandegreesfreedomconsciousnesssciencepossiblesimulationtalkartificialphysicsargumenthardfuture
💬 精彩语录
暂无语录
🎙️ 完整对话（844 条）
Lex Fridman (00:00.000)
The following is a conversation with Sean Carroll.
Lex Fridman (00:02.700)
He's a theoretical physicist at Caltech
Lex Fridman (00:04.900)
specializing in quantum mechanics, gravity, and cosmology.
Lex Fridman (00:08.780)
He's the author of several popular books,
Lex Fridman (00:11.620)
one on the arrow of time called From Eternity to Here,
Sean Carroll (00:15.340)
one on the Higgs boson called Particle
Lex Fridman (00:17.820)
at the End of the Universe,
Lex Fridman (00:19.140)
and one on science and philosophy called The Big Picture
Lex Fridman (00:22.540)
on the Origins of Life, Meaning, and the Universe Itself.
Sean Carroll (00:26.340)
He has an upcoming book on quantum mechanics
Lex Fridman (00:28.700)
that you can preorder now called Something Deeply Hidden.
Sean Carroll (00:32.660)
He writes one of my favorite blogs on his website,
Lex Fridman (00:36.060)
preposterousuniverse.com.
Sean Carroll (00:37.980)
I recommend clicking on the Greatest Hits link
Lex Fridman (00:40.460)
that lists accessible, interesting posts
Sean Carroll (00:43.340)
on the arrow of time, dark matter, dark energy,
Lex Fridman (00:45.660)
the Big Bang, general relativity,
Sean Carroll (00:47.620)
string theory, quantum mechanics,
Lex Fridman (00:49.580)
and the big meta questions about the philosophy of science,
Sean Carroll (00:53.180)
God, ethics, politics, academia, and much, much more.
Lex Fridman (00:57.620)
Finally, and perhaps most famously,
Sean Carroll (01:00.300)
he's the host of a podcast called Mindscape
Lex Fridman (01:03.660)
that you should subscribe to and support on Patreon.
Sean Carroll (01:06.940)
Along with the Joe Rogan experience,
Lex Fridman (01:08.820)
Sam Harris's Making Sense,
Lex Fridman (01:10.500)
and Dan Carlin's Hardcore History,
Lex Fridman (01:13.100)
Sean's Mindscape podcast is one of my favorite ways
Sean Carroll (01:15.860)
to learn new ideas or explore different perspectives
Lex Fridman (01:18.820)
and ideas that I thought I understood.
Sean Carroll (01:22.140)
It was truly an honor to meet
Lex Fridman (01:24.660)
and spend a couple hours with Sean.
Sean Carroll (01:27.240)
It's a bit heartbreaking to say
Lex Fridman (01:28.980)
that for the first time ever,
Sean Carroll (01:30.540)
the audio recorder for this podcast
Lex Fridman (01:32.500)
died in the middle of our conversation.
Sean Carroll (01:34.940)
There's technical reasons for this,
Lex Fridman (01:36.320)
having to do with phantom power
Sean Carroll (01:38.380)
that I now understand and will avoid.
Lex Fridman (01:41.060)
It took me one hour to notice and fix the problem.
Sean Carroll (01:44.220)
So, much like the universe is 68% dark energy,
Lex Fridman (01:48.340)
roughly the same amount from this conversation was lost,
Sean Carroll (01:51.340)
except in the memories of the two people involved
Lex Fridman (01:54.220)
and in my notes.
Sean Carroll (01:56.320)
I'm sure we'll talk again and continue this conversation
Lex Fridman (01:59.940)
on this podcast or on Sean's.
Lex Fridman (02:02.420)
And of course, I look forward to it.
Lex Fridman (02:05.300)
This is the Artificial Intelligence podcast.
Sean Carroll (02:07.820)
If you enjoy it, subscribe on YouTube, iTunes,
Lex Fridman (02:11.060)
support it on Patreon,
Sean Carroll (02:12.520)
or simply connect with me on Twitter at Lex Friedman.
Lex Fridman (02:16.660)
And now, here's my conversation with Sean Carroll.
Lex Fridman (02:21.380)
What do you think is more interesting and impactful,
Lex Fridman (02:23.540)
understanding how the universe works at a fundamental level
Lex Fridman (02:26.860)
or understanding how the human mind works?
Lex Fridman (02:29.180)
You know, of course this is a crazy,
Sean Carroll (02:31.960)
meaningless, unanswerable question in some sense,
Lex Fridman (02:33.940)
because they're both very interesting
Lex Fridman (02:35.140)
and there's no absolute scale of interestingness
Lex Fridman (02:37.500)
that we can rate them on.
Sean Carroll (02:39.180)
There's a glib answer that says the human brain
Lex Fridman (02:41.140)
is part of the universe, right?
Lex Fridman (02:43.060)
And therefore, understanding the universe
Lex Fridman (02:44.420)
is more fundamental than understanding the human brain.
Lex Fridman (02:47.020)
But do you really believe that once we understand
Lex Fridman (02:49.580)
the fundamental way the universe works
Sean Carroll (02:51.500)
at the particle level, the forces,
Lex Fridman (02:53.740)
we would be able to understand how the mind works?
Sean Carroll (02:55.820)
No, certainly not.
Lex Fridman (02:56.660)
We cannot understand how ice cream works
Lex Fridman (02:58.740)
just from understanding how particles work, right?
Lex Fridman (03:01.060)
So I'm a big believer in emergence.
Sean Carroll (03:02.740)
I'm a big believer that there are different ways
Lex Fridman (03:05.300)
of talking about the world
Sean Carroll (03:07.900)
beyond just the most fundamental microscopic one.
Lex Fridman (03:11.180)
You know, when we talk about tables and chairs
Lex Fridman (03:13.860)
and planets and people,
Lex Fridman (03:15.120)
we're not talking the language of particle physics
Lex Fridman (03:17.300)
and cosmology.
Lex Fridman (03:18.380)
So, but understanding the universe,
Lex Fridman (03:20.860)
you didn't say just at the most fundamental level, right?
Lex Fridman (03:24.060)
So understanding the universe at all levels
Sean Carroll (03:26.740)
is part of that.
Lex Fridman (03:28.200)
I do think, you know, to be a little bit more fair
Sean Carroll (03:29.940)
to the question, there probably are general principles
Lex Fridman (03:33.980)
of complexity, biology, information processing,
Sean Carroll (03:38.500)
memory, knowledge, creativity
Lex Fridman (03:41.820)
that go beyond just the human brain, right?
Lex Fridman (03:45.620)
And maybe one could count understanding those
Lex Fridman (03:47.800)
as part of understanding the universe.
Sean Carroll (03:49.140)
The human brain, as far as we know,
Lex Fridman (03:50.480)
is the most complex thing in the universe.
Lex Fridman (03:54.300)
So there's, it's certainly absurd to think
Lex Fridman (03:57.420)
that by understanding the fundamental laws
Sean Carroll (03:58.860)
of particle physics,
Lex Fridman (04:00.340)
you get any direct insight on how the brain works.
Lex Fridman (04:02.860)
But then there's this step from the fundamentals
Lex Fridman (04:05.940)
of particle physics to information processing,
Sean Carroll (04:08.700)
which a lot of physicists and philosophers
Lex Fridman (04:10.820)
may be a little bit carelessly take
Sean Carroll (04:12.460)
when they talk about artificial intelligence.
Lex Fridman (04:14.620)
Do you think of the universe
Lex Fridman (04:18.020)
as a kind of a computational device?
Lex Fridman (04:21.300)
No, to be like, the honest answer there is no.
Sean Carroll (04:24.140)
There's a sense in which the universe
Lex Fridman (04:26.300)
processes information, clearly.
Sean Carroll (04:29.140)
There's a sense in which the universe
Lex Fridman (04:30.700)
is like a computer, clearly.
Lex Fridman (04:33.880)
But in some sense, I think,
Lex Fridman (04:36.500)
I tried to say this once on my blog
Lex Fridman (04:38.540)
and no one agreed with me,
Lex Fridman (04:39.360)
but the universe is more like a computation
Sean Carroll (04:42.360)
than a computer because the universe happens once.
Lex Fridman (04:45.060)
A computer is a general purpose machine, right?
Sean Carroll (04:46.900)
That you can ask it different questions,
Lex Fridman (04:48.700)
even a pocket calculator, right?
Lex Fridman (04:50.140)
And it's set up to answer certain kinds of questions.
Lex Fridman (04:52.980)
The universe isn't that.
Lex Fridman (04:54.340)
So information processing happens in the universe,
Lex Fridman (04:57.360)
but it's not what the universe is.
Lex Fridman (04:59.220)
And I know your MIT colleague, Seth Lloyd,
Lex Fridman (05:01.580)
feels very differently about this, right?
Sean Carroll (05:03.820)
Well, you're thinking of the universe as a closed system.
Lex Fridman (05:07.220)
I am.
Lex Fridman (05:08.060)
So what makes a computer more like a PC,
Lex Fridman (05:12.980)
like a computing machine is that there's a human
Sean Carroll (05:15.500)
that every once comes up to it and moves the mouse around.
Lex Fridman (05:19.100)
So input.
Sean Carroll (05:19.940)
Gives it input.
Lex Fridman (05:20.760)
Gives it input.
Lex Fridman (05:23.500)
And that's why you're saying it's just a computation,
Lex Fridman (05:26.300)
a deterministic thing that's just unrolling.
Lex Fridman (05:29.260)
But the immense complexity of it
Lex Fridman (05:32.220)
is nevertheless like processing.
Sean Carroll (05:34.420)
There's a state and then it changes with good rules.
Lex Fridman (05:40.140)
And there's a sense for a lot of people
Sean Carroll (05:41.660)
that if the brain operates,
Lex Fridman (05:44.420)
the human brain operates within that world,
Sean Carroll (05:46.460)
then it's simply just a small subset of that.
Lex Fridman (05:49.340)
And so there's no reason we can't build
Sean Carroll (05:52.500)
arbitrarily great intelligences.
Lex Fridman (05:55.560)
Yeah.
Lex Fridman (05:56.400)
Do you think of intelligence in this way?
Lex Fridman (05:58.660)
Intelligence is tricky.
Sean Carroll (05:59.580)
I don't have a definition of it offhand.
Lex Fridman (06:01.660)
So I remember this panel discussion that I saw on YouTube.
Sean Carroll (06:05.460)
I wasn't there, but Seth Lloyd was on the panel.
Lex Fridman (06:07.620)
And so was Martin Rees, the famous astrophysicist.
Lex Fridman (06:10.540)
And Seth gave his shtick for why the universe is a computer
Lex Fridman (06:13.780)
and explained this.
Lex Fridman (06:14.820)
And Martin Rees said, so what is not a computer?
Lex Fridman (06:19.140)
And Seth was like, oh, that's a good question.
Sean Carroll (06:22.000)
I'm not sure.
Lex Fridman (06:22.840)
Because if you have a sufficiently broad definition
Lex Fridman (06:24.960)
of what a computer is, then everything is, right?
Lex Fridman (06:28.360)
And the simile or the analogy gains force
Sean Carroll (06:32.140)
when it excludes some things.
Lex Fridman (06:34.380)
You know, is the moon going around the earth
Lex Fridman (06:36.260)
performing a computation?
Lex Fridman (06:38.620)
I can come up with definitions in which the answer is yes,
Lex Fridman (06:41.320)
but it's not a very useful computation.
Lex Fridman (06:43.820)
I think that it's absolutely helpful
Sean Carroll (06:46.140)
to think about the universe in certain situations,
Lex Fridman (06:49.620)
certain contexts, as an information processing device.
Sean Carroll (06:53.020)
I'm even guilty of writing a paper
Lex Fridman (06:54.860)
called Quantum Circuit Cosmology,
Sean Carroll (06:56.820)
where we modeled the whole universe as a quantum circuit.
Lex Fridman (06:59.260)
As a circuit.
Sean Carroll (07:00.100)
As a circuit, yeah.
Lex Fridman (07:01.340)
With qubits kind of thing?
Sean Carroll (07:02.860)
With qubits basically, right, yeah.
Lex Fridman (07:05.040)
So, and qubits becoming more and more entangled.
Lex Fridman (07:07.440)
So do we wanna digress a little bit?
Lex Fridman (07:09.660)
Let's do it.
Sean Carroll (07:10.500)
It's kind of fun.
Lex Fridman (07:11.340)
So here's a mystery about the universe
Sean Carroll (07:13.700)
that is so deep and profound that nobody talks about it.
Lex Fridman (07:16.880)
Space expands, right?
Lex Fridman (07:19.080)
And we talk about, in a certain region of space,
Lex Fridman (07:21.940)
a certain number of degrees of freedom,
Sean Carroll (07:23.620)
a certain number of ways that the quantum fields
Lex Fridman (07:25.540)
and the particles in that region can arrange themselves.
Sean Carroll (07:28.800)
That number of degrees of freedom in a region of space
Lex Fridman (07:32.220)
is arguably finite.
Sean Carroll (07:33.820)
We actually don't know how many there are,
Lex Fridman (07:36.660)
but there's a very good argument
Sean Carroll (07:37.820)
that says it's a finite number.
Lex Fridman (07:39.420)
So as the universe expands and space gets bigger,
Lex Fridman (07:44.900)
are there more degrees of freedom?
Lex Fridman (07:46.780)
If it's an infinite number, it doesn't really matter.
Sean Carroll (07:48.540)
Infinity times two is still infinity.
Lex Fridman (07:49.980)
But if it's a finite number, then there's more space,
Lex Fridman (07:53.480)
so there's more degrees of freedom.
Lex Fridman (07:54.420)
So where did they come from?
Sean Carroll (07:55.740)
That would mean the universe is not a closed system.
Lex Fridman (07:58.020)
There's more degrees of freedom popping into existence.
Lex Fridman (08:01.500)
So what we suggested was
Lex Fridman (08:03.460)
that there are more degrees of freedom,
Lex Fridman (08:05.340)
and it's not that they're not there to start,
Lex Fridman (08:07.980)
but they're not entangled to start.
Lex Fridman (08:10.860)
So the universe that you and I know of,
Lex Fridman (08:12.820)
the three dimensions around us that we see,
Sean Carroll (08:15.440)
we said those are the entangled degrees of freedom
Lex Fridman (08:18.100)
making up space time.
Lex Fridman (08:19.620)
And as the universe expands,
Lex Fridman (08:20.920)
there are a whole bunch of qubits in their zero state
Sean Carroll (08:25.180)
that become entangled with the rest of space time
Lex Fridman (08:28.140)
through the action of these quantum circuits.
Lex Fridman (08:31.180)
So what does it mean that there's now more
Lex Fridman (08:35.580)
degrees of freedom as they become more entangled?
Sean Carroll (08:39.300)
Yeah, so.
Lex Fridman (08:40.300)
As the universe expands.
Sean Carroll (08:41.660)
That's right, so there's more and more degrees of freedom
Lex Fridman (08:43.300)
that are entangled, that are playing part,
Sean Carroll (08:46.420)
playing the role of part
Lex Fridman (08:47.360)
of the entangled space time structure.
Lex Fridman (08:49.600)
So the basic, the underlying philosophy is
Lex Fridman (08:51.980)
that space time itself arises from the entanglement
Sean Carroll (08:54.620)
of some fundamental quantum degrees of freedom.
Lex Fridman (08:57.560)
Wow, okay, so at which point
Lex Fridman (09:00.820)
is most of the entanglement happening?
Lex Fridman (09:05.260)
Are we talking about close to the Big Bang?
Lex Fridman (09:07.460)
Are we talking about throughout the time of the life?
Lex Fridman (09:11.820)
Throughout history, yeah.
Lex Fridman (09:12.660)
So the idea is that at the Big Bang,
Lex Fridman (09:15.140)
almost all the degrees of freedom
Sean Carroll (09:16.780)
that the universe could have were there,
Lex Fridman (09:19.700)
but they were unentangled with anything else.
Lex Fridman (09:22.420)
And that's a reflection of the fact
Lex Fridman (09:23.900)
that the Big Bang had a low entropy.
Sean Carroll (09:25.620)
It was a very simple, very small place.
Lex Fridman (09:28.020)
And as space expands, more and more degrees of freedom
Sean Carroll (09:31.420)
become entangled with the rest of the world.
Lex Fridman (09:34.300)
Well, I have to ask John Carroll,
Lex Fridman (09:35.940)
what do you think of the thought experiment
Lex Fridman (09:37.880)
from Nick Bostrom that we're living in a simulation?
Lex Fridman (09:41.580)
So I think, let me contextualize that a little bit more.
Lex Fridman (09:44.980)
I think people don't actually take this thought experiments.
Sean Carroll (09:48.340)
I think it's quite interesting.
Lex Fridman (09:50.460)
It's not very useful, but it's quite interesting.
Sean Carroll (09:52.900)
From the perspective of AI,
Lex Fridman (09:54.500)
a lot of the learning that can be done usually happens
Sean Carroll (09:58.020)
in simulation from artificial examples.
Lex Fridman (10:00.580)
And so it's a constructive question to ask,
Lex Fridman (10:04.900)
how difficult is our real world to simulate?
Lex Fridman (10:08.240)
Right.
Sean Carroll (10:09.360)
Which is kind of a dual part of,
Lex Fridman (10:12.180)
if we're living in a simulation
Lex Fridman (10:14.100)
and somebody built that simulation,
Lex Fridman (10:16.420)
if you were to try to do it yourself, how hard would it be?
Lex Fridman (10:18.860)
So obviously we could be living in a simulation.
Lex Fridman (10:21.100)
If you just want the physical possibility,
Sean Carroll (10:23.000)
then I completely agree that it's physically possible.
Lex Fridman (10:25.420)
I don't think that we actually are.
Lex Fridman (10:27.380)
So take this one piece of data into consideration.
Lex Fridman (10:30.300)
You know, we live in a big universe, okay?
Sean Carroll (10:35.140)
There's two trillion galaxies in our observable universe
Lex Fridman (10:38.500)
with 200 billion stars in each galaxy, et cetera.
Sean Carroll (10:41.660)
It would seem to be a waste of resources
Lex Fridman (10:44.940)
to have a universe that big going on
Sean Carroll (10:46.540)
just to do a simulation.
Lex Fridman (10:47.540)
So in other words, I want to be a good Bayesian.
Sean Carroll (10:50.140)
I want to ask under this hypothesis,
Lex Fridman (10:52.940)
what do I expect to see?
Lex Fridman (10:54.960)
So the first thing I would say is I wouldn't expect
Lex Fridman (10:56.780)
to see a universe that was that big, okay?
Sean Carroll (11:00.340)
The second thing is I wouldn't expect the resolution
Lex Fridman (11:02.540)
of the universe to be as good as it is.
Lex Fridman (11:05.020)
So it's always possible that if our superhuman simulators
Lex Fridman (11:08.740)
only have finite resources,
Lex Fridman (11:09.900)
that they don't render the entire universe, right?
Lex Fridman (11:12.420)
That the part that is out there,
Sean Carroll (11:14.340)
the two trillion galaxies,
Lex Fridman (11:16.300)
isn't actually being simulated fully, okay?
Lex Fridman (11:19.640)
But then the obvious extrapolation of that
Lex Fridman (11:22.740)
is that only I am being simulated fully.
Lex Fridman (11:25.500)
Like the rest of you are just non player characters, right?
Lex Fridman (11:29.220)
I'm the only thing that is real.
Sean Carroll (11:30.520)
The rest of you are just chat bots.
Lex Fridman (11:32.780)
Beyond this wall, I see the wall,
Lex Fridman (11:34.320)
but there is literally nothing
Lex Fridman (11:36.020)
on the other side of the wall.
Sean Carroll (11:37.360)
That is sort of the Bayesian prediction.
Lex Fridman (11:39.300)
That's what it would be like
Sean Carroll (11:40.180)
to do an efficient simulation of me.
Lex Fridman (11:42.240)
So like none of that seems quite realistic.
Sean Carroll (11:45.700)
I don't see, I hear the argument that it's just possible
Lex Fridman (11:50.900)
and easy to simulate lots of things.
Sean Carroll (11:53.300)
I don't see any evidence from what we know
Lex Fridman (11:55.780)
about our universe that we look like a simulated universe.
Sean Carroll (11:59.340)
Now, maybe you can say,
Lex Fridman (12:00.180)
well, we don't know what it would look like,
Lex Fridman (12:01.980)
but that's just abandoning your Bayesian responsibilities.
Lex Fridman (12:04.520)
Like your job is to say under this theory,
Sean Carroll (12:07.660)
here's what you would expect to see.
Lex Fridman (12:09.500)
Yeah, so certainly if you think about simulation
Sean Carroll (12:11.660)
as a thing that's like a video game
Lex Fridman (12:14.340)
where only a small subset is being rendered.
Lex Fridman (12:17.740)
But say the entire, all the laws of physics,
Lex Fridman (12:22.740)
the entire closed system of the quote unquote universe,
Sean Carroll (12:26.540)
it had a creator.
Lex Fridman (12:27.780)
Yeah, it's always possible.
Sean Carroll (12:29.320)
Right, so that's not useful to think about
Lex Fridman (12:32.280)
when you're thinking about physics.
Sean Carroll (12:34.020)
The way Nick Bostrom phrases it,
Lex Fridman (12:36.220)
if it's possible to simulate a universe,
Sean Carroll (12:39.100)
eventually we'll do it.
Lex Fridman (12:40.500)
Right.
Sean Carroll (12:42.700)
You can use that by the way for a lot of things.
Lex Fridman (12:44.860)
Well, yeah.
Lex Fridman (12:45.700)
But I guess the question is,
Lex Fridman (12:48.540)
how hard is it to create a universe?
Sean Carroll (12:52.340)
I wrote a little blog post about this
Lex Fridman (12:53.820)
and maybe I'm missing something,
Lex Fridman (12:55.460)
but there's an argument that says not only
Lex Fridman (12:57.680)
that it might be possible to simulate a universe,
Lex Fridman (13:00.500)
but probably if you imagine that you actually attribute
Lex Fridman (13:05.980)
consciousness and agency to the little things
Sean Carroll (13:08.860)
that we're simulating, to our little artificial beings,
Lex Fridman (13:12.020)
there's probably a lot more of them
Sean Carroll (13:13.420)
than there are ordinary organic beings in the universe
Lex Fridman (13:15.500)
or there will be in the future, right?
Lex Fridman (13:17.420)
So there's an argument that not only
Lex Fridman (13:18.500)
is being a simulation possible,
Sean Carroll (13:20.760)
it's probable because in the space
Lex Fridman (13:23.560)
of all living consciousnesses,
Lex Fridman (13:24.960)
most of them are being simulated, right?
Lex Fridman (13:26.620)
Most of them are not at the top level.
Sean Carroll (13:28.860)
I think that argument must be wrong
Lex Fridman (13:30.540)
because it follows from that argument that,
Sean Carroll (13:33.100)
if we're simulated, but we can also simulate other things,
Lex Fridman (13:36.920)
well, but if we can simulate other things,
Lex Fridman (13:38.840)
they can simulate other things, right?
Lex Fridman (13:41.840)
If we give them enough power and resolution
Lex Fridman (13:44.320)
and ultimately we'll reach a bottom
Lex Fridman (13:45.980)
because the laws of physics in our universe have a bottom,
Sean Carroll (13:49.140)
we're made of atoms and so forth,
Lex Fridman (13:51.000)
so there will be the cheapest possible simulations.
Lex Fridman (13:55.100)
And if you believe the original argument,
Lex Fridman (13:57.700)
you should conclude that we should be
Sean Carroll (13:59.340)
in the cheapest possible simulation
Lex Fridman (14:00.940)
because that's where most people are.
Lex Fridman (14:02.660)
But we don't look like that.
Lex Fridman (14:03.620)
It doesn't look at all like we're at the edge of resolution,
Sean Carroll (14:06.860)
that we're 16 bit things.
Lex Fridman (14:09.540)
It seems much easier to make much lower level things
Sean Carroll (14:13.020)
than we are.
Lex Fridman (14:14.980)
And also, I questioned the whole approach
Sean Carroll (14:18.220)
to the anthropic principle
Lex Fridman (14:19.460)
that says we are typical observers in the universe.
Sean Carroll (14:22.340)
I think that that's not actually,
Lex Fridman (14:23.660)
I think that there's a lot of selection that we can do
Sean Carroll (14:27.340)
that we're typical within things we already know,
Lex Fridman (14:30.180)
but not typical within all of the universe.
Lex Fridman (14:32.280)
So do you think there's intelligent life,
Lex Fridman (14:35.800)
however you would like to define intelligent life,
Lex Fridman (14:37.860)
out there in the universe?
Lex Fridman (14:39.940)
My guess is that there is not intelligent life
Sean Carroll (14:44.660)
in the observable universe other than us, simply
Lex Fridman (14:48.820)
on the basis of the fact that the likely number
Sean Carroll (14:52.540)
of other intelligent species in the observable universe,
Lex Fridman (14:56.340)
there's two likely numbers, zero or billions.
Lex Fridman (15:01.500)
And if there had been billions,
Lex Fridman (15:02.580)
you would have noticed already.
Sean Carroll (15:05.300)
For there to be literally like a small number,
Lex Fridman (15:07.340)
like, you know, Star Trek,
Sean Carroll (15:09.380)
there's a dozen intelligent civilizations in our galaxy,
Lex Fridman (15:13.300)
but not a billion, that's weird.
Sean Carroll (15:17.340)
That's sort of bizarre to me.
Lex Fridman (15:18.500)
It's easy for me to imagine that there are zero others
Sean Carroll (15:21.020)
because there's just a big bottleneck
Lex Fridman (15:22.620)
to making multicellular life
Sean Carroll (15:24.980)
or technological life or whatever.
Lex Fridman (15:27.020)
It's very hard for me to imagine
Sean Carroll (15:28.580)
that there's a whole bunch out there
Lex Fridman (15:30.140)
that have somehow remained hidden from us.
Sean Carroll (15:32.300)
The question I'd like to ask
Lex Fridman (15:34.700)
is what would intelligent life look like?
Lex Fridman (15:38.140)
What I mean by that question and where it's going
Lex Fridman (15:40.500)
is what if intelligent life is just in some very big ways
Lex Fridman (15:47.260)
different than the one that has on Earth?
Lex Fridman (15:51.500)
That there's all kinds of intelligent life
Sean Carroll (15:53.900)
that operates at different scales
Lex Fridman (15:55.420)
of both size and temporal.
Sean Carroll (15:57.300)
Right, that's a great possibility
Lex Fridman (15:59.300)
because I think we should be humble
Sean Carroll (16:00.800)
about what intelligence is, what life is.
Lex Fridman (16:02.640)
We don't even agree on what life is,
Lex Fridman (16:04.020)
much less what intelligent life is, right?
Lex Fridman (16:07.020)
So that's an argument for humility,
Sean Carroll (16:08.980)
saying there could be intelligent life
Lex Fridman (16:10.860)
of a very different character, right?
Sean Carroll (16:13.620)
Like you could imagine the dolphins are intelligent
Lex Fridman (16:18.060)
but never invent space travel
Sean Carroll (16:20.500)
because they live in the ocean
Lex Fridman (16:21.460)
and they don't have thumbs, right?
Lex Fridman (16:24.180)
So they never invent technology, they never invent smelting.
Lex Fridman (16:27.860)
Maybe the universe is full of intelligent species
Lex Fridman (16:32.020)
that just don't make technology, right?
Lex Fridman (16:34.060)
That's compatible with the data, I think.
Lex Fridman (16:36.320)
And I think maybe what you're pointing at
Lex Fridman (16:39.840)
is even more out there versions of intelligence,
Sean Carroll (16:44.440)
intelligence in intermolecular clouds
Lex Fridman (16:47.560)
or on the surface of a neutron star
Sean Carroll (16:49.440)
or in between the galaxies in giant things
Lex Fridman (16:51.760)
where the equivalent of a heartbeat is 100 million years.
Sean Carroll (16:56.440)
On the one hand, yes,
Lex Fridman (16:58.080)
we should be very open minded about those things.
Sean Carroll (16:59.860)
On the other hand, all of us share the same laws of physics.
Lex Fridman (17:04.860)
There might be something about the laws of physics,
Sean Carroll (17:08.240)
even though we don't currently know
Lex Fridman (17:09.400)
exactly what that thing would be,
Sean Carroll (17:10.920)
that makes meters and years
Lex Fridman (17:16.160)
the right length and timescales for intelligent life.
Sean Carroll (17:19.880)
Maybe not, but we're made of atoms,
Lex Fridman (17:22.240)
atoms have a certain size,
Sean Carroll (17:23.780)
we orbit stars or stars have a certain lifetime.
Lex Fridman (17:27.280)
It's not impossible to me that there's a sweet spot
Sean Carroll (17:30.300)
for intelligent life that we find ourselves in.
Lex Fridman (17:32.200)
So I'm open minded either way,
Sean Carroll (17:33.800)
I'm open minded either being humble
Lex Fridman (17:35.280)
and there's all sorts of different kinds of life
Sean Carroll (17:37.080)
or no, there's a reason we just don't know it yet
Lex Fridman (17:39.280)
why life like ours is the kind of life that's out there.
Sean Carroll (17:42.080)
Yeah, I'm of two minds too,
Lex Fridman (17:43.320)
but I often wonder if our brains is just designed
Sean Carroll (17:47.200)
to quite obviously to operate and see the world
Lex Fridman (17:52.720)
in these timescales and we're almost blind
Lex Fridman (17:56.360)
and the tools we've created for detecting things are blind
Lex Fridman (18:01.200)
to the kind of observation needed
Sean Carroll (18:02.760)
to see intelligent life at other scales.
Lex Fridman (18:04.920)
Well, I'm totally open to that,
Lex Fridman (18:07.040)
but so here's another argument I would make,
Lex Fridman (18:09.240)
we have looked for intelligent life,
Lex Fridman (18:11.520)
but we've looked at for it in the dumbest way we can,
Lex Fridman (18:14.120)
by turning radio telescopes to the sky.
Lex Fridman (18:16.600)
And why in the world would a super advanced civilization
Lex Fridman (18:21.040)
randomly beam out radio signals wastefully
Lex Fridman (18:24.040)
in all directions into the universe?
Lex Fridman (18:25.440)
That just doesn't make any sense,
Sean Carroll (18:27.280)
especially because in order to think
Lex Fridman (18:29.100)
that you would actually contact another civilization,
Sean Carroll (18:32.020)
you would have to do it forever,
Lex Fridman (18:33.840)
you have to keep doing it for millions of years,
Sean Carroll (18:35.840)
that sounds like a waste of resources.
Lex Fridman (18:38.280)
If you thought that there were other solar systems
Sean Carroll (18:43.120)
with planets around them,
Lex Fridman (18:44.520)
where maybe intelligent life didn't yet exist,
Lex Fridman (18:47.000)
but might someday,
Lex Fridman (18:48.600)
you wouldn't try to talk to it with radio waves,
Sean Carroll (18:51.380)
you would send a spacecraft out there
Lex Fridman (18:53.600)
and you would park it around there
Lex Fridman (18:55.560)
and it would be like, from our point of view,
Lex Fridman (18:57.360)
it'd be like 2001, where there was a monolith.
Sean Carroll (19:00.700)
Monolith.
Lex Fridman (19:01.540)
There could be an artifact,
Lex Fridman (19:02.380)
in fact, the other way works also, right?
Lex Fridman (19:04.520)
There could be artifacts in our solar system
Sean Carroll (19:08.440)
that have been put there
Lex Fridman (19:10.480)
by other technologically advanced civilizations
Lex Fridman (19:12.280)
and that's how we will eventually contact them.
Lex Fridman (19:14.640)
We just haven't explored the solar system well enough yet
Sean Carroll (19:16.840)
to find them.
Lex Fridman (19:18.580)
The reason why we don't think about that
Lex Fridman (19:20.000)
is because we're young and impatient, right?
Lex Fridman (19:21.520)
Like, it would take more than my lifetime
Sean Carroll (19:24.000)
to actually send something to another star system
Lex Fridman (19:26.080)
and wait for it and then come back.
Sean Carroll (19:27.800)
So, but if we start thinking on hundreds of thousands
Lex Fridman (19:30.800)
of years or million year time scales,
Sean Carroll (19:32.720)
that's clearly the right thing to do.
Lex Fridman (19:34.600)
Are you excited by the thing
Lex Fridman (19:36.800)
that Elon Musk is doing with SpaceX in general?
Lex Fridman (19:39.360)
Space, but the idea of space exploration,
Lex Fridman (19:41.620)
even though your, or your species is young and impatient?
Lex Fridman (19:45.360)
Yeah.
Sean Carroll (19:46.200)
No, I do think that space travel is crucially important,
Lex Fridman (19:49.200)
long term.
Sean Carroll (19:50.800)
Even to other star systems.
Lex Fridman (19:52.500)
And I think that many people overestimate the difficulty
Sean Carroll (19:57.500)
because they say, look, if you travel 1% the speed of light
Lex Fridman (20:00.940)
to another star system,
Lex Fridman (20:02.020)
we'll be dead before we get there, right?
Lex Fridman (20:04.060)
And I think that it's much easier.
Lex Fridman (20:06.180)
And therefore, when they write their science fiction stories,
Lex Fridman (20:08.120)
they imagine we'd go faster than the speed of light
Lex Fridman (20:09.580)
because otherwise they're too impatient, right?
Lex Fridman (20:11.700)
We're not gonna go faster than the speed of light,
Lex Fridman (20:13.600)
but we could easily imagine that the human lifespan
Lex Fridman (20:16.020)
gets extended to thousands of years.
Lex Fridman (20:18.100)
And once you do that,
Lex Fridman (20:19.140)
then the stars are much closer effectively, right?
Lex Fridman (20:21.180)
And then what's a hundred year trip, right?
Lex Fridman (20:23.260)
So I think that that's gonna be the future,
Sean Carroll (20:25.820)
the far future, not my lifetime once again,
Lex Fridman (20:28.700)
but baby steps.
Sean Carroll (20:30.380)
Unless your lifetime gets extended.
Lex Fridman (20:32.420)
Well, it's in a race against time, right?
Sean Carroll (20:34.740)
A friend of mine who actually thinks about these things
Lex Fridman (20:37.340)
said, you know, you and I are gonna die,
Lex Fridman (20:40.460)
but I don't know about our grandchildren.
Lex Fridman (20:43.060)
That's, I don't know, predicting the future is hard,
Lex Fridman (20:45.940)
but that's at least a plausible scenario.
Lex Fridman (20:47.900)
And so, yeah, no, I think that as we discussed earlier,
Lex Fridman (20:51.820)
there are threats to the earth, known and unknown, right?
Lex Fridman (20:56.780)
Having spread humanity and biology elsewhere
Sean Carroll (21:02.580)
is a really important longterm goal.
Lex Fridman (21:04.940)
What kind of questions can science not currently answer,
Lex Fridman (21:08.900)
but might soon?
Lex Fridman (21:11.480)
When you think about the problems and the mysteries
Sean Carroll (21:13.860)
before us that may be within reach of science.
Lex Fridman (21:17.840)
I think an obvious one is the origin of life.
Sean Carroll (21:20.300)
We don't know how that happened.
Lex Fridman (21:22.780)
There's a difficulty in knowing how it happened historically
Sean Carroll (21:25.300)
actually, you know, literally on earth,
Lex Fridman (21:27.240)
but starting life from non life is something
Lex Fridman (21:30.500)
I kind of think we're close to, right?
Lex Fridman (21:32.420)
We're really.
Lex Fridman (21:33.240)
You really think so?
Lex Fridman (21:34.080)
Like how difficult is it to start life?
Sean Carroll (21:36.740)
Well, I've talked to people,
Lex Fridman (21:39.260)
including on the podcast about this.
Sean Carroll (21:41.780)
You know, life requires three things.
Lex Fridman (21:43.340)
Life as we know it.
Lex Fridman (21:44.220)
So there's a difference with life,
Lex Fridman (21:45.500)
which who knows what it is,
Lex Fridman (21:47.060)
and life as we know it,
Lex Fridman (21:48.140)
which we can talk about with some intelligence.
Lex Fridman (21:50.780)
So life as we know it requires compartmentalization.
Lex Fridman (21:53.840)
You need like a little membrane around your cell.
Sean Carroll (21:56.660)
Metabolism, you need to take in food and eat it
Lex Fridman (21:58.980)
and let that make you do things.
Lex Fridman (22:01.020)
And then replication, okay?
Lex Fridman (22:02.620)
So you need to have some information about who you are
Sean Carroll (22:04.620)
that you pass down to future generations.
Lex Fridman (22:07.880)
In the lab, compartmentalization seems pretty easy.
Sean Carroll (22:11.780)
Not hard to make lipid bilayers
Lex Fridman (22:13.780)
that come into little cellular walls pretty easily.
Sean Carroll (22:16.760)
Metabolism and replication are hard,
Lex Fridman (22:20.160)
but replication we're close to.
Sean Carroll (22:21.900)
People have made RNA like molecules in the lab
Lex Fridman (22:24.960)
that I think the state of the art is,
Sean Carroll (22:28.840)
they're not able to make one molecule
Lex Fridman (22:30.660)
that reproduces itself,
Lex Fridman (22:32.060)
but they're able to make two molecules
Lex Fridman (22:33.600)
that reproduce each other.
Lex Fridman (22:35.260)
So that's okay.
Lex Fridman (22:36.100)
That's pretty close.
Sean Carroll (22:38.060)
Metabolism is harder, believe it or not,
Lex Fridman (22:41.060)
even though it's sort of the most obvious thing,
Lex Fridman (22:42.900)
but you want some sort of controlled metabolism
Lex Fridman (22:44.940)
and the actual cellular machinery in our bodies
Sean Carroll (22:47.500)
is quite complicated.
Lex Fridman (22:48.660)
It's hard to see it just popping into existence
Sean Carroll (22:50.940)
all by itself.
Lex Fridman (22:51.780)
It probably took a while,
Lex Fridman (22:53.740)
but we're making progress.
Lex Fridman (22:56.100)
And in fact, I don't think we're spending
Sean Carroll (22:57.240)
nearly enough money on it.
Lex Fridman (22:58.580)
If I were the NSF, I would flood this area with money
Sean Carroll (23:01.780)
because it would change our view of the world
Lex Fridman (23:05.220)
if we could actually make life in the lab
Lex Fridman (23:06.780)
and understand how it was made originally here on earth.
Lex Fridman (23:09.420)
And I'm sure it'd have some ripple effects
Sean Carroll (23:11.160)
that help cure disease and so on.
Lex Fridman (23:12.940)
I mean, just that understanding.
Lex Fridman (23:14.380)
So synthetic biology is a wonderful big frontier
Lex Fridman (23:16.700)
where we're making cells.
Sean Carroll (23:18.940)
Right now, the best way to do that
Lex Fridman (23:21.100)
is to borrow heavily from existing biology, right?
Sean Carroll (23:23.620)
Well, Craig Venter several years ago
Lex Fridman (23:25.380)
created an artificial cell, but all he did was,
Sean Carroll (23:28.220)
not all he did, it was a tremendous accomplishment,
Lex Fridman (23:29.860)
but all he did was take out the DNA from a cell
Lex Fridman (23:33.180)
and put in entirely new DNA and let it boot up and go.
Lex Fridman (23:37.200)
What about the leap to creating intelligent life on earth?
Sean Carroll (23:43.420)
Yeah.
Lex Fridman (23:44.260)
Again, we define intelligence, of course,
Lex Fridman (23:45.860)
but let's just even say Homo sapiens,
Lex Fridman (23:49.860)
the modern intelligence in our human brain.
Lex Fridman (23:55.340)
Do you have a sense of what's involved in that leap
Lex Fridman (23:58.660)
and how big of a leap that is?
Lex Fridman (24:00.420)
So AI would count in this, or do you really want life?
Lex Fridman (24:03.300)
Do you want really an organism in some sense?
Sean Carroll (24:06.420)
AI would count, I think.
Lex Fridman (24:07.540)
Okay.
Sean Carroll (24:08.980)
Yeah, of course, of course AI would count.
Lex Fridman (24:11.020)
Well, let's say artificial consciousness, right?
Lex Fridman (24:13.460)
So I do not think we are on the threshold
Lex Fridman (24:15.500)
of creating artificial consciousness.
Sean Carroll (24:16.760)
I think it's possible.
Lex Fridman (24:18.180)
I'm not, again, very educated about how close we are,
Lex Fridman (24:20.300)
but my impression is not that we're really close
Lex Fridman (24:22.100)
because we understand how little we understand
Sean Carroll (24:24.820)
of consciousness and what it is.
Lex Fridman (24:26.460)
So if we don't have any idea what it is,
Sean Carroll (24:28.440)
it's hard to imagine we're on the threshold
Lex Fridman (24:29.780)
of making it ourselves.
Lex Fridman (24:32.500)
But it's doable, it's possible.
Lex Fridman (24:34.500)
I don't see any obstacles in principle.
Lex Fridman (24:35.960)
So yeah, I would hold out some interest
Lex Fridman (24:38.160)
in that happening eventually.
Sean Carroll (24:40.220)
I think in general, consciousness,
Lex Fridman (24:42.700)
I think we would be just surprised
Lex Fridman (24:44.420)
how easy consciousness is once we create intelligence.
Lex Fridman (24:49.060)
I think consciousness is a thing
Sean Carroll (24:50.540)
that's just something we all fake.
Lex Fridman (24:55.540)
Well, good.
Sean Carroll (24:56.380)
No, actually, I like this idea that in fact,
Lex Fridman (24:57.680)
consciousness is way less mysterious than we think
Sean Carroll (25:00.500)
because we're all at every time, at every moment,
Lex Fridman (25:02.620)
less conscious than we think we are, right?
Sean Carroll (25:04.500)
We can fool things.
Lex Fridman (25:05.460)
And I think that plus the idea
Sean Carroll (25:07.780)
that you not only have artificial intelligent systems,
Lex Fridman (25:11.180)
but you put them in a body, right,
Sean Carroll (25:12.980)
give them a robot body,
Lex Fridman (25:15.620)
that will help the faking a lot.
Sean Carroll (25:18.460)
Yeah, I think creating consciousness
Lex Fridman (25:20.980)
in artificial consciousness is as simple
Sean Carroll (25:25.140)
as asking a Roomba to say, I'm conscious,
Lex Fridman (25:30.020)
and refusing to be talked out of it.
Sean Carroll (25:32.780)
Could be, it could be.
Lex Fridman (25:33.820)
And I mean, I'm almost being silly,
Lex Fridman (25:36.740)
but that's what we do.
Lex Fridman (25:39.660)
That's what we do with each other.
Sean Carroll (25:40.940)
This is the kind of,
Lex Fridman (25:42.020)
that consciousness is also a social construct.
Lex Fridman (25:44.500)
And a lot of our ideas of intelligence is a social construct.
Lex Fridman (25:47.860)
And so reaching that bar involves something that's beyond,
Sean Carroll (25:52.820)
that doesn't necessarily involve
Lex Fridman (25:54.940)
the fundamental understanding of how you go
Sean Carroll (25:57.720)
from electrons to neurons to cognition.
Lex Fridman (26:02.500)
No, actually, I think that is an extremely good point.
Lex Fridman (26:05.060)
And in fact, what it suggests is,
Lex Fridman (26:08.660)
so yeah, you referred to Kate Darling,
Sean Carroll (26:10.540)
who I had on the podcast,
Lex Fridman (26:11.940)
and who does these experiments with very simple robots,
Lex Fridman (26:16.440)
but they look like animals,
Lex Fridman (26:18.060)
and they can look like they're experiencing pain,
Lex Fridman (26:20.740)
and we human beings react very negatively
Lex Fridman (26:23.380)
to these little robots
Sean Carroll (26:24.400)
looking like they're experiencing pain.
Lex Fridman (26:26.300)
And what you wanna say is, yeah, but they're just robots.
Lex Fridman (26:29.980)
It's not really pain, right?
Lex Fridman (26:31.700)
It's just some electrons going around.
Lex Fridman (26:33.080)
But then you realize, you and I are just electrons
Lex Fridman (26:36.300)
going around, and that's what pain is also.
Lex Fridman (26:38.380)
And so what I would have an easy time imagining
Lex Fridman (26:43.060)
is that there is a spectrum
Sean Carroll (26:44.740)
between these simple little robots that Kate works with
Lex Fridman (26:47.420)
and a human being,
Sean Carroll (26:49.420)
where there are things that sort of
Lex Fridman (26:50.940)
by some strict definition,
Sean Carroll (26:52.840)
Turing test level thing are not conscious,
Lex Fridman (26:55.460)
but nevertheless walk and talk like they're conscious.
Lex Fridman (26:58.580)
And it could be that the future is,
Lex Fridman (27:00.220)
I mean, Siri is close, right?
Lex Fridman (27:02.460)
And so it might be the future
Lex Fridman (27:04.540)
has a lot more agents like that.
Lex Fridman (27:07.100)
And in fact, rather than someday going,
Lex Fridman (27:08.860)
aha, we have consciousness,
Sean Carroll (27:10.700)
we'll just creep up on it with more and more
Lex Fridman (27:13.180)
accurate reflections of what we expect.
Lex Fridman (27:15.220)
And in the future, maybe the present,
Lex Fridman (27:18.320)
for example, we haven't met before,
Lex Fridman (27:20.800)
and you're basically assuming that I'm human as it's a high
Lex Fridman (27:25.300)
probability at this time because the yeah,
Lex Fridman (27:28.560)
but in the future,
Lex Fridman (27:30.200)
there might be question marks around that, right?
Sean Carroll (27:32.000)
Yeah, no, absolutely.
Lex Fridman (27:33.340)
Certainly videos are almost to the point
Sean Carroll (27:35.740)
where you shouldn't trust them already.
Lex Fridman (27:36.740)
Photos you can't trust, right?
Sean Carroll (27:39.060)
Videos is easier to trust,
Lex Fridman (27:41.700)
but we're getting worse that,
Lex Fridman (27:44.020)
we're getting better at faking them, right?
Lex Fridman (27:46.540)
Yeah, so physical embodied people,
Lex Fridman (27:48.780)
what's so hard about faking that?
Lex Fridman (27:51.020)
So this is very depressing,
Sean Carroll (27:51.980)
this conversation we're having right now.
Lex Fridman (27:53.420)
So I mean,
Sean Carroll (27:54.340)
To me, it's exciting.
Lex Fridman (27:55.180)
To me, you're doing it.
Lex Fridman (27:56.300)
So it's exciting to you,
Lex Fridman (27:57.780)
but it's a sobering thought.
Lex Fridman (27:59.060)
We're very bad, right?
Lex Fridman (28:00.420)
At imagining what the next 50 years are gonna be like
Sean Carroll (28:02.820)
when we're in the middle of a phase transition
Lex Fridman (28:04.220)
as we are right now.
Sean Carroll (28:05.260)
Yeah, and I, in general,
Lex Fridman (28:06.740)
I'm not blind to all the threats.
Sean Carroll (28:09.220)
I am excited by the power of technology to solve,
Lex Fridman (28:14.540)
to protect us against the threats as they evolve.
Sean Carroll (28:18.060)
I'm not as much as Steven Pinker optimistic about the world,
Lex Fridman (28:22.340)
but in everything I've seen,
Sean Carroll (28:23.740)
all of the brilliant people in the world that I've met
Lex Fridman (28:27.300)
are good people.
Lex Fridman (28:29.160)
So the army of the good
Lex Fridman (28:30.800)
in terms of the development of technology is large.
Sean Carroll (28:33.400)
Okay, you're way more optimistic than I am.
Lex Fridman (28:37.820)
I think that goodness and badness
Sean Carroll (28:39.060)
are equally distributed among intelligent
Lex Fridman (28:40.900)
and unintelligent people.
Sean Carroll (28:42.700)
I don't see much of a correlation there.
Lex Fridman (28:44.660)
Interesting.
Sean Carroll (28:46.060)
Neither of us have proof.
Lex Fridman (28:47.300)
Yeah, exactly.
Lex Fridman (28:48.420)
Again, opinions are free, right?
Lex Fridman (28:50.660)
Nor definitions of good and evil.
Sean Carroll (28:52.540)
We come without definitions or without data opinions.
Lex Fridman (28:57.460)
So what kind of questions can science not currently answer
Lex Fridman (29:01.980)
and may never be able to answer in your view?
Lex Fridman (29:04.380)
Well, the obvious one is what is good and bad?
Lex Fridman (29:06.940)
What is right and wrong?
Lex Fridman (29:07.860)
I think that there are questions that,
Sean Carroll (29:09.460)
science tells us what happens,
Lex Fridman (29:11.300)
what the world is and what it does.
Sean Carroll (29:13.260)
It doesn't say what the world should do
Lex Fridman (29:14.740)
or what we should do,
Sean Carroll (29:15.580)
because we're part of the world.
Lex Fridman (29:17.800)
But we are part of the world
Lex Fridman (29:19.200)
and we have the ability to feel like something's right,
Lex Fridman (29:21.460)
something's wrong.
Lex Fridman (29:22.740)
And to make a very long story very short,
Lex Fridman (29:25.660)
I think that the idea of moral philosophy
Sean Carroll (29:28.000)
is systematizing our intuitions
Lex Fridman (29:30.100)
of what is right and what is wrong.
Lex Fridman (29:31.700)
And science might be able to predict ahead of time
Lex Fridman (29:34.580)
what we will do,
Lex Fridman (29:36.180)
but it won't ever be able to judge
Lex Fridman (29:38.000)
whether we should have done it or not.
Sean Carroll (29:39.600)
So, you're kind of unique in terms of scientists.
Lex Fridman (29:43.620)
Listen, it doesn't have to do with podcasts,
Lex Fridman (29:45.520)
but even just reaching out,
Lex Fridman (29:47.660)
I think you referred to as sort of
Sean Carroll (29:49.080)
doing interdisciplinary science.
Lex Fridman (29:51.300)
So you reach out and talk to people
Sean Carroll (29:54.100)
that are outside of your discipline,
Lex Fridman (29:55.980)
which I always hope that's what science was for.
Sean Carroll (30:00.140)
In fact, I was a little disillusioned
Lex Fridman (30:02.300)
when I realized that academia is very siloed.
Sean Carroll (30:06.420)
Yeah.
Lex Fridman (30:07.260)
And so the question is,
Sean Carroll (30:10.700)
how, at your own level,
Lex Fridman (30:13.020)
how do you prepare for these conversations?
Lex Fridman (30:15.380)
How do you think about these conversations?
Lex Fridman (30:16.900)
How do you open your mind enough
Lex Fridman (30:18.300)
to have these conversations?
Lex Fridman (30:20.220)
And it may be a little bit broader,
Lex Fridman (30:21.940)
how can you advise other scientists
Lex Fridman (30:24.380)
to have these kinds of conversations?
Sean Carroll (30:26.260)
Not at the podcast,
Lex Fridman (30:28.180)
the fact that you're doing a podcast is awesome,
Sean Carroll (30:29.860)
other people get to hear them,
Lex Fridman (30:31.380)
but it's also good to have it without mics in general.
Sean Carroll (30:34.700)
It's a good question, but a tough one to answer.
Lex Fridman (30:37.460)
I think about a guy I know who's a personal trainer,
Lex Fridman (30:40.980)
and he was asked on a podcast,
Lex Fridman (30:43.240)
how do we psych ourselves up to do a workout?
Lex Fridman (30:45.700)
How do we make that discipline to go and work out?
Lex Fridman (30:48.340)
And he's like, why are you asking me?
Sean Carroll (30:50.300)
I can't stop working out.
Lex Fridman (30:52.340)
I don't need to psych myself up.
Sean Carroll (30:54.380)
So, and likewise, he asked me,
Lex Fridman (30:57.340)
how do you get to have interdisciplinary conversations
Sean Carroll (30:59.740)
on all sorts of different things,
Lex Fridman (31:00.700)
all sorts of different people?
Lex Fridman (31:01.660)
I'm like, that's what makes me go, right?
Lex Fridman (31:04.860)
Like that's, I couldn't stop doing that.
Sean Carroll (31:07.380)
I did that long before any of them were recorded.
Lex Fridman (31:09.660)
In fact, a lot of the motivation for starting recording it
Sean Carroll (31:12.380)
was making sure I would read all these books
Lex Fridman (31:14.420)
that I had purchased, right?
Sean Carroll (31:15.460)
Like all these books I wanted to read,
Lex Fridman (31:17.700)
not enough time to read them.
Lex Fridman (31:18.900)
And now if I have the motivation,
Lex Fridman (31:20.700)
cause I'm gonna interview Pat Churchland,
Sean Carroll (31:23.220)
I'm gonna finally read her book.
Lex Fridman (31:25.180)
You know, and it's absolutely true
Lex Fridman (31:29.460)
that academia is extraordinarily siloed, right?
Lex Fridman (31:31.700)
We don't talk to people.
Sean Carroll (31:32.780)
We rarely do.
Lex Fridman (31:34.260)
And in fact, when we do, it's punished.
Sean Carroll (31:36.460)
You know, like the people who do it successfully
Lex Fridman (31:38.820)
generally first became very successful
Sean Carroll (31:41.420)
within their little siloed discipline.
Lex Fridman (31:43.100)
And only then did they start expanding out.
Sean Carroll (31:46.380)
If you're a young person, you know,
Lex Fridman (31:47.660)
I have graduate students.
Sean Carroll (31:48.940)
I try to be very, very candid with them about this,
Lex Fridman (31:52.980)
that it's, you know, most graduate students
Lex Fridman (31:55.580)
are to not become faculty members, right?
Lex Fridman (31:57.420)
It's a tough road.
Lex Fridman (31:59.020)
And so live the life you wanna live,
Lex Fridman (32:03.140)
but do it with your eyes open
Sean Carroll (32:04.620)
about what it does to your job chances.
Lex Fridman (32:06.900)
And the more broad you are
Lex Fridman (32:09.580)
and the less time you spend hyper specializing
Lex Fridman (32:12.900)
in your field, the lower your job chances are.
Sean Carroll (32:15.780)
That's just an academic reality.
Lex Fridman (32:17.060)
It's terrible, I don't like it, but it's a reality.
Lex Fridman (32:20.060)
And for some people, that's fine.
Lex Fridman (32:22.540)
Like there's plenty of people who are wonderful scientists
Sean Carroll (32:24.660)
who have zero interest in branching out
Lex Fridman (32:27.140)
and talking to things, to anyone outside their field.
Lex Fridman (32:30.740)
But it is disillusioning to me.
Lex Fridman (32:33.740)
Some of the, you know, romantic notion I had
Sean Carroll (32:36.180)
of the intellectual academic life
Lex Fridman (32:38.220)
is belied by the reality of it.
Sean Carroll (32:39.940)
The idea that we should reach out beyond our discipline
Lex Fridman (32:43.500)
and that is a positive good is just so rare
Sean Carroll (32:48.500)
in universities that it may as well not exist at all.
Lex Fridman (32:53.900)
But that said, even though you're saying you're doing it
Sean Carroll (32:57.660)
like the personal trainer, because you just can't help it,
Lex Fridman (33:00.300)
you're also an inspiration to others.
Sean Carroll (33:02.940)
Like I could speak for myself.
Lex Fridman (33:05.780)
You know, I also have a career I'm thinking about, right?
Lex Fridman (33:09.540)
And without your podcast,
Lex Fridman (33:12.060)
I may have not have been doing this at all, right?
Lex Fridman (33:15.060)
So it makes me realize that these kinds of conversations
Lex Fridman (33:19.540)
is kind of what science is about in many ways.
Sean Carroll (33:23.340)
The reason we write papers, this exchange of ideas,
Lex Fridman (33:27.460)
is it's much harder to do interdisciplinary papers,
Sean Carroll (33:30.540)
I would say.
Lex Fridman (33:31.380)
And conversations are easier.
Lex Fridman (33:35.140)
So conversations is the beginning.
Lex Fridman (33:36.820)
And in the field of AI, it's obvious
Sean Carroll (33:41.180)
that we should think outside of pure computer vision
Lex Fridman (33:45.580)
competitions on a particular data sets.
Sean Carroll (33:47.540)
We should think about the broader impact
Lex Fridman (33:49.660)
of how this can be, you know, reaching out to physics,
Sean Carroll (33:53.740)
to psychology, to neuroscience and having these
Lex Fridman (33:57.220)
conversations so that you're an inspiration.
Lex Fridman (34:00.580)
And so never know how the world changes.
Lex Fridman (34:05.220)
I mean, the fact that this stuff is out there
Lex Fridman (34:08.540)
and I've a huge number of people come up to me,
Lex Fridman (34:12.300)
grad students, really loving the podcast, inspired by it.
Lex Fridman (34:16.100)
And they will probably have that,
Lex Fridman (34:18.660)
they'll be ripple effects when they become faculty
Lex Fridman (34:20.740)
and so on and so on.
Lex Fridman (34:21.580)
We can end on a balance between pessimism and optimism.
Lex Fridman (34:25.300)
And Sean, thank you so much for talking to me, it was awesome.
Lex Fridman (34:27.780)
No, Lex, thank you very much for this conversation.
Sean Carroll (34:29.460)
It was great.
🔗 相关节目