episode
stringlengths
45
100
text
stringlengths
1
528
timestamp_link
stringlengths
56
56
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
sort of totally lost any notion of human autonomy. And, you know, so in effect, right. This would
https://karpathy.ai/lexicap/0009-large.html#01:17:07.440
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
happen like the slow boiling frog, right? We would gradually turn over more and more of the
https://karpathy.ai/lexicap/0009-large.html#01:17:17.680
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
management of our civilization to machines as we are already doing. And this, you know, if this
https://karpathy.ai/lexicap/0009-large.html#01:17:24.240
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
if this process continues, you know, we sort of gradually switch from sort of being the masters
https://karpathy.ai/lexicap/0009-large.html#01:17:29.520
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
of technology to just being the guests. Right. So we become guests on a cruise ship, you know,
https://karpathy.ai/lexicap/0009-large.html#01:17:37.920
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
which is fine for a week, but not not for the rest of eternity. You know, and it's almost
https://karpathy.ai/lexicap/0009-large.html#01:17:44.160
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
irreversible. Right. Once you once you lose the incentive to, for example, you know, learn to be
https://karpathy.ai/lexicap/0009-large.html#01:17:51.360
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
an engineer or a doctor or a sanitation operative or any other of the infinitely many ways that we
https://karpathy.ai/lexicap/0009-large.html#01:17:58.640
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
maintain and propagate our civilization. You know, if you if you don't have the incentive to do any
https://karpathy.ai/lexicap/0009-large.html#01:18:08.000
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
of that, you won't. And then it's really hard to recover. And of course, as just one of the
https://karpathy.ai/lexicap/0009-large.html#01:18:14.240
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
technologies that could that third failure mode result in that there's probably other
https://karpathy.ai/lexicap/0009-large.html#01:18:20.320
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
technology in general detaches us from it does a bit. But the difference is that in terms of
https://karpathy.ai/lexicap/0009-large.html#01:18:24.400
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
the knowledge to to run our civilization, you know, up to now, we've had no alternative but
https://karpathy.ai/lexicap/0009-large.html#01:18:31.120
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
to put it into people's heads. Right. And if you software with Google, I mean, so software in
https://karpathy.ai/lexicap/0009-large.html#01:18:38.240
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
general, so computers in general, but but the, you know, the knowledge of how, you know, how a
https://karpathy.ai/lexicap/0009-large.html#01:18:43.920
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
sanitation system works, you know, that's an AI has to understand that it's no good putting it
https://karpathy.ai/lexicap/0009-large.html#01:18:51.200
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
into Google. So, I mean, we we've always put knowledge in on paper, but paper doesn't run our
https://karpathy.ai/lexicap/0009-large.html#01:18:56.000
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
civilization and only runs when it goes from the paper into people's heads again. Right. So we've
https://karpathy.ai/lexicap/0009-large.html#01:19:02.560
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
always propagated civilization through human minds. And we've spent about a trillion person
https://karpathy.ai/lexicap/0009-large.html#01:19:07.120
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
years doing that. I literally write you, you can work it out. It's about right. There's about just
https://karpathy.ai/lexicap/0009-large.html#01:19:13.680
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
over 100 billion people who've ever lived. And each of them has spent about 10 years learning
https://karpathy.ai/lexicap/0009-large.html#01:19:19.440
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
stuff to keep their civilization going. And so that's a trillion person years we put into this
https://karpathy.ai/lexicap/0009-large.html#01:19:25.280
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
effort. Beautiful way to describe all civilization. And now we're, you know, we're in danger of
https://karpathy.ai/lexicap/0009-large.html#01:19:30.640
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
throwing that away. So this is a problem that AI can't solve. It's not a technical problem. It's
https://karpathy.ai/lexicap/0009-large.html#01:19:36.160
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
you know, if we do our job right, the AI systems will say, you know, the human race doesn't in the
https://karpathy.ai/lexicap/0009-large.html#01:19:40.880
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
long run want to be passengers in a cruise ship. The human race wants autonomy. This is part of
https://karpathy.ai/lexicap/0009-large.html#01:19:48.560
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
human preferences. So we, the AI systems are not going to do this stuff for you. You've got to do
https://karpathy.ai/lexicap/0009-large.html#01:19:54.560
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
it for yourself. Right. I'm not going to carry you to the top of Everest in an autonomous
https://karpathy.ai/lexicap/0009-large.html#01:20:01.200
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
helicopter. You have to climb it if you want to get the benefit and so on. So, but I'm afraid that
https://karpathy.ai/lexicap/0009-large.html#01:20:06.320
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
because we are short sighted and lazy, we're going to override the AI systems. And, and there's an
https://karpathy.ai/lexicap/0009-large.html#01:20:14.960
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
amazing short story that I recommend to everyone that I talked to about this called The Machine
https://karpathy.ai/lexicap/0009-large.html#01:20:22.400
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
Stops, written in 1909 by E.M. Forster, who, you know, wrote novels about the British Empire and
https://karpathy.ai/lexicap/0009-large.html#01:20:28.720
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
sort of things that became costume dramas on the BBC. But he wrote this one science fiction story,
https://karpathy.ai/lexicap/0009-large.html#01:20:37.520
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
which is an amazing vision of the future. It has basically iPads, it has video conferencing,
https://karpathy.ai/lexicap/0009-large.html#01:20:42.240
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
it has MOOCs, it has computer induced obesity. I mean, literally it's what people spend their
https://karpathy.ai/lexicap/0009-large.html#01:20:51.680
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
time doing is giving online courses or listening to online courses and talking about ideas,
https://karpathy.ai/lexicap/0009-large.html#01:21:00.320
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
but they never get out there in the real world. They don't really have a lot of face to face
https://karpathy.ai/lexicap/0009-large.html#01:21:05.920
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
contact. Everything is done online, you know, so all the things we're worrying about now
https://karpathy.ai/lexicap/0009-large.html#01:21:11.200
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
were described in the story. And, and then the human race becomes more and more dependent on
https://karpathy.ai/lexicap/0009-large.html#01:21:17.520
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
the machine, loses knowledge of how things really run and then becomes vulnerable to collapse. And
https://karpathy.ai/lexicap/0009-large.html#01:21:22.000
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
so it's a, it's a pretty unbelievably amazing story for someone writing in 1909 to imagine all
https://karpathy.ai/lexicap/0009-large.html#01:21:31.360
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
this. So there's very few people that represent artificial intelligence more than you Stuart
https://karpathy.ai/lexicap/0009-large.html#01:21:38.640
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
Russell. If you say it's okay, that's very kind. So it's all my fault. Right. You're often brought
https://karpathy.ai/lexicap/0009-large.html#01:21:45.760
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
up as the person, well, Stuart Russell, like the AI person is worried about this. That's why you
https://karpathy.ai/lexicap/0009-large.html#01:21:57.200
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
should be worried about it. Do you feel the burden of that? I don't know if you feel that at all,
https://karpathy.ai/lexicap/0009-large.html#01:22:03.680
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
but when I talk to people like from, you talk about people outside of computer science,
https://karpathy.ai/lexicap/0009-large.html#01:22:10.240
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
when they think about this, Stuart Russell is worried about AI safety. You should be worried
https://karpathy.ai/lexicap/0009-large.html#01:22:15.840
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
too. Do you feel the burden of that? I mean, in a practical sense, yeah, because I get, you know,
https://karpathy.ai/lexicap/0009-large.html#01:22:21.280
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
a dozen, sometimes 25 invitations a day to talk about it, to give interviews, to write press
https://karpathy.ai/lexicap/0009-large.html#01:22:29.840
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
articles and so on. So in that very practical sense, I'm seeing that people are concerned and
https://karpathy.ai/lexicap/0009-large.html#01:22:38.640
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
really interested about this. Are you worried that you could be wrong as all good scientists are?
https://karpathy.ai/lexicap/0009-large.html#01:22:46.160
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
Of course. I worry about that all the time. I mean, that's, that's always been the way that I,
https://karpathy.ai/lexicap/0009-large.html#01:22:52.320
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
I've worked, you know, is like I have an argument in my head with myself, right? So I have,
https://karpathy.ai/lexicap/0009-large.html#01:22:57.920
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
I have some idea and then I think, okay, how could that be wrong? Or did someone else already have
https://karpathy.ai/lexicap/0009-large.html#01:23:03.440
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
that idea? So I'll go and, you know, search in as much literature as I can to see whether someone
https://karpathy.ai/lexicap/0009-large.html#01:23:10.560
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
else already thought of that or, or even refuted it. So, you know, I, right now I'm, I'm reading a
https://karpathy.ai/lexicap/0009-large.html#01:23:16.720
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
lot of philosophy because, you know, in, in the form of the debates over, over utilitarianism and,
https://karpathy.ai/lexicap/0009-large.html#01:23:23.680
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
and other kinds of moral, moral formulas, shall we say, people have already thought through
https://karpathy.ai/lexicap/0009-large.html#01:23:32.800
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
some of these issues. But, you know, what, one of the things I'm, I'm not seeing in a lot of
https://karpathy.ai/lexicap/0009-large.html#01:23:42.800
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
these debates is this specific idea about the importance of uncertainty in the objective
https://karpathy.ai/lexicap/0009-large.html#01:23:47.680
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
that this is the way we should think about machines that are beneficial to humans. So this
https://karpathy.ai/lexicap/0009-large.html#01:23:56.400
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
idea of provably beneficial machines based on explicit uncertainty in the objective,
https://karpathy.ai/lexicap/0009-large.html#01:24:01.920
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
you know, it seems to be, you know, my gut feeling is this is the core of it. It's going to have to
https://karpathy.ai/lexicap/0009-large.html#01:24:10.000
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
be elaborated in a lot of different directions and there are a lot of beneficial. Yeah. But there,
https://karpathy.ai/lexicap/0009-large.html#01:24:17.600
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
there are, I mean, it has to be right. We can't afford, you know, hand wavy beneficial because
https://karpathy.ai/lexicap/0009-large.html#01:24:23.600
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
there are, you know, whenever we do hand wavy stuff, there are loopholes. And the thing about
https://karpathy.ai/lexicap/0009-large.html#01:24:30.640
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
super intelligent machines is they find the loopholes, you know, just like, you know, tax
https://karpathy.ai/lexicap/0009-large.html#01:24:34.800
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
evaders. If you don't write your tax law properly, people will find the loopholes and end up paying
https://karpathy.ai/lexicap/0009-large.html#01:24:40.560
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
no tax. And, and so you should think of it this way and, and getting those definitions right,
https://karpathy.ai/lexicap/0009-large.html#01:24:46.400
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
you know, it is really a long process, you know, so you can, you can define mathematical frameworks
https://karpathy.ai/lexicap/0009-large.html#01:24:56.480
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
and within that framework, you can prove mathematical theorems that yes, this will,
https://karpathy.ai/lexicap/0009-large.html#01:25:04.400
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
you know, this, this theoretical entity will be provably beneficial to that theoretical entity,
https://karpathy.ai/lexicap/0009-large.html#01:25:08.560
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
but that framework may not match the real world in some crucial way. So it's a long process,
https://karpathy.ai/lexicap/0009-large.html#01:25:13.680
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
thinking through it, iterating and so on. Last question. Yep. You have 10 seconds to answer it.
https://karpathy.ai/lexicap/0009-large.html#01:25:20.160
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
What is your favorite sci fi movie about AI? I would say interstellar has my favorite robots.
https://karpathy.ai/lexicap/0009-large.html#01:25:27.120
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
Oh, beats space. Yeah. Yeah. Yeah. So, so Tars, the robots, one of the robots in interstellar is
https://karpathy.ai/lexicap/0009-large.html#01:25:34.480
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
the way robot should behave. And, uh, I would say ex machina is in some ways, the one,
https://karpathy.ai/lexicap/0009-large.html#01:25:42.160
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
the one that makes you think, uh, in a nervous kind of way about, about where we're going.
https://karpathy.ai/lexicap/0009-large.html#01:25:51.520
Stuart Russell: Long-Term Future of Artificial Intelligence | Lex Fridman Podcast #9
Well Stuart, thank you so much for talking today. Pleasure.
https://karpathy.ai/lexicap/0009-large.html#01:25:58.080
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
The following is a conversation with Peter Abbeel.
https://karpathy.ai/lexicap/0010-large.html#00:00:00.000
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
He's a professor at UC Berkeley
https://karpathy.ai/lexicap/0010-large.html#00:00:03.120
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
and the director of the Berkeley Robotics Learning Lab.
https://karpathy.ai/lexicap/0010-large.html#00:00:04.840
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
He's one of the top researchers in the world
https://karpathy.ai/lexicap/0010-large.html#00:00:07.840
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
working on how we make robots understand
https://karpathy.ai/lexicap/0010-large.html#00:00:10.080
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
and interact with the world around them,
https://karpathy.ai/lexicap/0010-large.html#00:00:13.080
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
especially using imitation and deep reinforcement learning.
https://karpathy.ai/lexicap/0010-large.html#00:00:15.360
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
This conversation is part of the MIT course
https://karpathy.ai/lexicap/0010-large.html#00:00:19.720
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
on Artificial General Intelligence
https://karpathy.ai/lexicap/0010-large.html#00:00:22.360
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
and the Artificial Intelligence podcast.
https://karpathy.ai/lexicap/0010-large.html#00:00:24.080
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
If you enjoy it, please subscribe on YouTube,
https://karpathy.ai/lexicap/0010-large.html#00:00:26.400
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
iTunes, or your podcast provider of choice,
https://karpathy.ai/lexicap/0010-large.html#00:00:29.060
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
or simply connect with me on Twitter at Lex Friedman,
https://karpathy.ai/lexicap/0010-large.html#00:00:31.680
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
spelled F R I D.
https://karpathy.ai/lexicap/0010-large.html#00:00:34.840
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
And now, here's my conversation with Peter Abbeel.
https://karpathy.ai/lexicap/0010-large.html#00:00:36.920
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
You've mentioned that if there was one person
https://karpathy.ai/lexicap/0010-large.html#00:00:41.400
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
you could meet, it would be Roger Federer.
https://karpathy.ai/lexicap/0010-large.html#00:00:44.120
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
So let me ask, when do you think we'll have a robot
https://karpathy.ai/lexicap/0010-large.html#00:00:46.200
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
that fully autonomously can beat Roger Federer at tennis?
https://karpathy.ai/lexicap/0010-large.html#00:00:50.120
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
Roger Federer level player at tennis?
https://karpathy.ai/lexicap/0010-large.html#00:00:54.760
Pieter Abbeel: Deep Reinforcement Learning | Lex Fridman Podcast #10
Well, first, if you can make it happen for me to meet Roger,
https://karpathy.ai/lexicap/0010-large.html#00:00:57.520