Latest / Elon Musk Thinking / Elon Musk Discuss AI SAFETY PANEL!

Transcript

Elon Musk Discuss AI SAFETY PANEL! · Elon Musk Thinking

0:00
Some great guests here. I think you guys know me and opposite Prime Minister,
0:05
but if you guys could introduce yourselves and say a bit about you and
0:08
you know you've done both done amazing things, so please let me shy.
0:12
Sure thing. So I'm Greg Brockman. I'm the president and co founder of
0:15
Open Aie. But we started Opening Eye out of my apartment in twenty fifteen,
0:20
twenty sixteen. So yeah, it's been a really exciting time to just,
0:24
you know, be able to help not just move this field forward,
0:27
but try to help steer it in a positive direction. I think that's really
0:29
foundational to how I think about AI. What got me excited about this field
0:33
is the fact it's going to be so transformative and I think it can benefit
0:36
everyone in huge ways, but we also really need to mitigate the downsides.
0:40
And so I think that's something that really resonated with me about the earlier conversation.
0:43
I hope we'll talk about a lot upcoming minutes. And since Greg was
0:47
too modest to say so, I'm going to add something to your introduction of
0:50
yourself there, which is just a few years ago when Greg and his colleagues
0:55
were saying, yo, in a few years we can act. We believe
0:59
we can actually build AI that can master human language and common knowledge. A
1:04
lot of my colleagues and AI were like, that's crazy science fiction. It's
1:08
going to take decades, but you did it. We literally had an intern
1:14
in twenty six teams our very first summer who we had this conversation about how
1:19
to solve the touring test, and we had all these ideas and whatever,
1:22
and we were like, all right, three years, we'll solve the touring
1:23
test. And that intern told me I'm out, like I'm not joining full
1:26
time, like you guys were crazy. And three years later we had GPT
1:30
two, so we didn't quite solve the Turing desk. So he was right,
1:32
But I think that we had far more progress than I think most people
1:34
thought possible. So I'm Max tech Mark, professor doing AI research at MIT,
1:41
and I think it was fought on what we heard earlier here from you,
1:46
that artificial intelligence will be both a blessing and a curse. I've very
1:51
for a long time and very excited about the blessing part, how we can
1:55
use it to solve all the problems that we haven't been smart enough to solve
1:57
before. And go for beyond and help life flourished for billions of years on
2:02
Earth and beyond. And I also I'm very concerned about the curse part.
2:07
So nine years ago I founded this nonprofit called the Future Life Institute. Together
2:12
with Elon we launched the first ever attempt to the mainstream AI safety by launching
2:17
a worldwide grants program and educate and do conferences and so on that you Greg
2:22
participated also to make sure we can steer this every more powerful technology away from
2:28
the curse towards the blessing direction. Well, can I say that I want
2:32
to plug this book? He wrote this book and I read it. It's
2:37
a good book. It's a very good book. Was it chat guptwo or
2:39
I have an alibi because not even Greg was fast enough to have it written
2:44
by the time chat gupt ain't good enough yet to do this. But from
2:49
what I hear you guys saying, it's good. Yeah, give it a
2:52
minute. It was thirty seconds. Well you see, Well let's talk about
2:54
that, because you know, I think that the kind of things that you
2:58
ring here are I think a monumental. You know, when I see the
3:01
talent on this stage, excluding myself, I'm reminded of what JFK. John
3:06
Kennedy said, you know, when he brought his Gifted team into the White
3:09
House. He said, this is the most gifted team ever since that's sat
3:14
here, ever since Thomas Jefferson dined alone. So I think you're about to
3:19
surpass that, and you are changing history. And I'm sure you know when
3:22
such power comes enormous responsibility. And I think that's the crux of what we're
3:25
talking about here is how do we inject a measure of responsibility and ethics into
3:31
this, into this exponentially changing development. How do you do that? That's
3:37
really what I'd like to know from you. I think that this should be
3:40
a very interesting discussion. And you know, I think I know almost everyone
3:45
in AI, and I think I really can't think of anyone who's got bad
3:50
motivations and AI. I think everyone I know an AI has good motivations.
3:53
They want to create technology that helps the world. But one you can also
3:58
say, if you think of someone like Einstein, who was obviously a very
4:00
sort of peace loving person, and you know, he didn't think that there
4:05
would be a nuclear bomb created as a result of his discoveries in physics.
4:10
He wasn't thinking, let me, let's make super weapons. He was thinking,
4:13
let's try to understand that the truth of the universe. So it's just
4:15
important too. I always want to just preface, you know, any criticism
4:20
of AI development with it not being a criticism of the people doing it,
4:25
but just fair in mind that just as Einstein didn't expect his work in physics
4:30
to lead to nuclear weapons, we need to be cautious that even with good
4:33
intentions, the best of intentions, that we could create something bad, that
4:38
that that's one of the possible outcomes, and we want to try to take
4:41
whatever actions we can to ensure that the future of humanity is good and that
4:46
is much more blessing than That's definitely why I work on it. I think
4:51
that not for evil it turns out, But I think that balance is really
4:55
key, and I think that one of the things that is important is to
4:59
really calibrate the choices with respect to the technology, right because I think this
5:03
technology it has this way of being very surprising, the way that people pictured
5:08
like you read asthmav you read kind of all the science fiction you read Ellie
5:12
as you're a Kowski, and kind of all this thinking people are doing about
5:15
very smart machines that are smarter than humans, and somehow none of it really
5:17
predicted GPT three, right. Somehow the technology that we're building, it's just
5:21
surprising. It has very different properties. And I think that even from you
5:26
think about like chess, people used to think chess would be we solve that,
5:29
we solve human intelligence, and it turns out that's the first thing to
5:30
go. And so I think that what we're really seeing is that we need
5:33
to be calibrated and see not just what the current systems are doing. And
5:36
this is one of the reasons we deployed chat GPT is it's just really important
5:40
to not get in our heads and think, oh, this is how it's
5:42
all going to play out, this is how people are going to misuse it.
5:44
But you need to actually see that. You need to actually have people
5:46
spending a lot of time thinking about where the slots in, where it works,
5:49
where it doesn't, but also to really try to peer ahead. And
5:53
so we've been putting a lot of effort into capability prediction. One of the
5:55
things that I think was kind of undersung in our GPT four release is that
5:59
we actually had very good predictions of exactly how good GPD four would be before
6:02
we even trained it, and that's something we can never do with previous models.
6:05
And I think that this and doing that it's a sidal scale, Like
6:10
I think that's one answer for how do we actually get this right? Because
6:12
if you can see around the corner, then you have an answer and an
6:15
AI. It's always been the opposite. We've always been taken by surprise.
6:17
Well, I want to ask you because there are sort of layers of questions
6:21
here. The Max's book takes you to the existential question of whether you know
6:28
you project basically machine intelligence or human intelligence into the cosmos, human intelligence turned
6:33
into machine intelligence into the cosmos and so on. That's a big philosophical question.
6:39
I'd like to think we have about six years for that, okay,
6:42
But the other questions that I'm dealing with as a leader of a country that
6:45
is blessed with a lot of AI talent rated about number two and talent relative
6:50
to population size and AI. But I wanted to be among the top three
6:55
because it's important to advance the country. But I think also for it raises
6:59
questions that we raised before. But here's another one. Okay, as chat
7:03
GUP four or chat. GPT eight comes into being, and you have Hollywood
7:08
strikes already about screenwriters and directors and actors. Okay, what do you think
7:13
will happen to the job market? I mean, we had this conversation,
7:16
what do you think will happen? I mean, you know, you can
7:19
tell me, you can comfort me with all previous revolutions that there were more
7:24
jobs, many more jobs created than lost. But you really believe that now?
7:27
Yeah, I think it's I think it's far from obvious. I think
7:29
it is a correct question. And I also think that it's important to kind
7:32
of look at the whole picture of for sure, their jobs are going to
7:35
go away, what's going to be the balance of like, you know,
7:39
marginal job creation? And you can tell the story of oh, it's going
7:41
to be very comforting you look at every previous technology. I do think AI
7:45
is not like every previous technology, right that this is usually we kind of
7:47
are seeing automation of mechanical parts of creation, and here it's almost this creative
7:54
aspect of creation. But I think that we want to even go one step
7:57
further from where we've been, right and not just okay, like you know,
8:01
I could have been an artist and now an AI is a competing artist,
8:03
but you want AI that can help us come up with ideas and solve
8:05
problems that we just couldn't before. And so what does that world look like?
8:09
And you know, if your post scarcity world but people don't have to
8:11
work to survive, you know what set of people? Because people I identify
8:16
themselves, you know, your whole identity is around your work. You identify
8:18
yourself by your job, yes, exactly, And so what will happen then?
8:22
And what does that look like? And so I don't come with an
8:24
answer to this question of exactly what the balance will be like. I think
8:28
there will be fundamental changes and how we even relate to work and to meaning
8:35
that will happen and may result in a landscape that just looks very different.
8:37
That the answer to that question is, you know, if you're a hunter
8:41
gatherer and someone's like, you know, what's going to happen to our hunter
8:43
gatherer job post agriculture of revolution, it's just a weird question, right,
8:46
It's like the framework changes on you. And so that's that's not a cop
8:50
out, right. I think that it might be the case that actually many
8:52
more jobs go away than get created, and there's a lot of chaos and
8:54
turmoil, but I think we got to be ahead of it. And the
8:58
more that we can predict it, the more that we can greet answers.
9:00
And we have teams that try, right, But I think that we're just
9:03
getting started and we need help. Well, it'll change our economic models.
9:05
Okay. I'm a clear free market guy. I liberated and I helped liberate
9:11
the Israeli economy to make it a sort of a technological powerhouse. But that's
9:15
because of free markets, and free markets means open competition. And it's fine,
9:18
but you stop at the monopolies and monopolies edge. Okay, well you
9:22
do that. I had a conversation with one of your colleagues, Peter Tail,
9:24
and he said to me, oh, it's all scale advantages, it's
9:28
all monopolies. I said, well, yeah, I believe that, but
9:31
I think we have to stop it at a certain point because we don't want
9:33
it to press competition. Okay. Well, Ai is producing you know these
9:39
this wall and you have these trillion dollar companies that are produced to one overnight,
9:45
okay, and they concentrate enormous wealth and power with smaller and smaller number
9:48
of people. Okay, and the question is what do you do about that
9:54
monopoly power? What do you do about competition? And if you are going
9:58
to if you are going to cannibalize a lot more jobs than you create,
10:01
then we have to change the structure of our economic policies and our political policies
10:07
to take care of the people who are not going to find jobs. We're
10:11
not going to contribute added value to the economy. We have to make sure
10:13
that they have a living, a decent one, and they have all the
10:16
services. Will probably have the money to do that, but it requires a
10:20
challenging model, certainly for a free market guy, a free market disciple like
10:24
me. And that's coming very fast. It's not going to come slowly,
10:28
and we have time to adjust. I like how you challenge the question what's
10:31
going to happen? I like to continue on that thing because I think if
10:35
we asked this very passive kind of question, what's going to happen to us,
10:39
let's just sit here and eat popcorn and wait for the future to happen,
10:43
then we're steading straight for the curse. I think this is a sort
10:46
of very passive approach which I think is just gonna you know, We've done
10:50
this experiment before when a new species showed up on the planet that was smarter
10:54
than all the other ones, and what happened was then the underthal automentics think,
10:58
if we just very passively go into this, I think it's very likely
11:01
humans will go extinct. Also, I think that's and I think that's the
11:05
wrong attitude. Even in economics in a country, you don't just sit back
11:09
and wait to see what happens and get child labor and the super monopoly of
11:15
doom. No, you ask yourself what do you want to happen, and
11:18
then you put in place institutions, you ban child labor, you put an
11:22
anti trust. And in this case, I think the analogous question is how
11:24
do we make sure that with this ever more powerful tech, it's we who
11:30
control the AI rather than the other way around. Right, So I was
11:33
honored that you had this book. If this book still hasn't put you to
11:37
sleep, because this is a technical nerd paper I wrote with this one's written
11:39
with technological approach to how we can control machines that are even smarter than us
11:45
by actually having other AIS formally prove that the I is safe. And the
11:48
reason why I'm so excited about the proving part is, you know, the
11:52
only way you can trust something much smarter than you to still do what you
11:56
want is if you can prove it, Because no one, how clever it
12:00
is, it can never do what's provably impossible, and it turns out it's
12:03
super It's much harder to come up with proofs about math or about what a
12:07
program is going to do than it is to check that the proof is correct.
12:11
You can write a three hundred line computer program which will check any proof
12:16
claiming anything about anything else. We can understand that if we force AI to
12:20
prove things to ourselves, to us humans that's going to meet our specifications,
12:26
then we actually have a reason a way we can control things that are smarter
12:30
than us. How do you know? How do you know that the proof
12:33
is not disputable? That's the beauty of mathematics. You can actually check even
12:37
if it's a brilliant proof, rather than I couldn't come up with myself.
12:41
I can just see I just plug this into this and this, Yeah,
12:45
you know it checks out. And there's a big science in doing this with
12:48
proving things about programs for cybersecurity because companies are so fed up with losing millions
12:54
of dollars by begetting hacked all the time because yet another bug was found and
13:00
or a shell or whatever. That there's been years of investment. Greg and
13:03
I had a fun conversation actually just before we went on stage here about how
13:07
this field of trying to prove that things are safe is still in the stone
13:11
age and there's pre large language model phase where it's most the humans trying to
13:16
make the proofs. I think it's ripe for a revolution where the kind of
13:18
technology that Open Eyes building can totally turbocharge our ability to prove things about the
13:24
code. And I think there's a bigger picture too, because in my mind,
13:28
if you think about cybersecurity, it's always a cat and mouse game.
13:31
You have attackers, you have defenders. Each one's leveling up defeating the other.
13:35
Is there ever going to be a time when the defenders totally win?
13:37
Is there going to be a time when the attackers totally win? And I
13:41
think with AI that balance can change. And I think that formal verifications an
13:43
example where maybe you can have defenders just win once and for all. And
13:48
I think there's a lot of asterisks there. I think it's definitely a maybe,
13:50
But I think that at least at a technical level, formal verification has
13:54
been this idea of if you could mathematically prove these systems are safe and good
13:58
and that you don't have bugs, and you're linings, colonel, then all
14:01
these bugs we've seen we were talking about heart bleed and shell shock and all
14:03
these these bugs from the past ten years, those would be solved, those
14:07
would go away. You would know there'd be no future one of those.
14:09
Now you need to be very careful because you need to make sure that your
14:11
undergoing machine model is like actually correct in a bunch of details like that.
14:16
But it's just interesting to think that these fundamental things we believe about how security
14:20
works might just be up it. And we've never had an opportunity for this.
14:22
That will give us a lot of free time exactly. And it's worth
14:26
adding for the nerds listening to this to like the code. You know,
14:28
normally, debugging, all that ever does is is all it can never accomplish
14:35
is proved that there is something wrong. If you don't find any bugs,
14:37
is there one more? What when you prove it? When it's what's called
14:43
formal verification, As Greg said, that means guaranteed you'll seal the deal.
14:48
Seal the deal, all right. That assumes, by the way, that
14:52
mathematics is, you know, that's it. It's the law of the universe.
14:56
It's not going to change. You're not going to get alien mathematics.
15:00
You don't need to assume that. Actually, well you're sort of getting the
15:01
girls incompleteness theorem and things like that. That has comes in a big way
15:05
when you're dealing with infinities of various kinds. All the computers, even Gregg's
15:09
very biggest servers, it's all finite. I would say, you still got
15:13
to watch out for the alien mathematics. But you can totally formalize what it
15:16
means that it's impossible to hack into his laptop, and if you have a
15:20
rigorous proof, I think that's heals the deal. So I think that beyond
15:24
these, you know, these enormous questions that you raise, I think from
15:28
me, the immediate problem is to fashion a national policy, which I tend
15:33
to announce in a few months, and you guys are simulated me. I
15:37
have to say this visit, but you sort of think, Okay, what
15:41
is it that you what is it that you were trying to achieve. You're
15:43
trying to better all the systems transportation, health, education, agriculture, production,
15:50
everything for government, for the private sector, for philanthropy, for academia
15:56
and so on. And what is the role of the government in this?
15:58
And you know, various kind trees are seeking solutions to this. What do
16:02
you do? What kind of instruments do you build? I think the larger
16:04
question is you still have the monopolization of AI power in small concentrations because of
16:11
what you said before, Elon, you said, basically, it's still big.
16:15
You know, you need the big data, you need the big you
16:18
know, the big computational power, you need the talent for creating the algorithms,
16:22
and that's concentrated in a few countries. We have another test, not
16:26
only to make sure that we take care of ourselves, but how we have
16:30
these benefits percolate to you know, the rest the scientific and technological have nots.
16:36
And I think that's that's another real issue. But so far, you
16:38
know the advances that we've made, at least in Israel, I don't know
16:42
anything from disconquy. Remember that you used to have something like that, And
16:47
I don't know that a drug, the camera pill that goes into your intestine
16:51
or any other thing that you have, from ICQ to cherry tomatoes that benefits
16:55
everybody. Drip irogation, okay, in Ai, it seems to me that
16:59
you're gonna have concentration of power that will create a bigger and bigger distance between
17:04
the haves and the have nots. And that's another thing that causes tremendous instability
17:10
in our world. And I don't know if you have an idea how you
17:12
overcome that. Well, first of all, I want to say just on
17:15
the talent density front, we have a number of Israeli engineers, scientist managers
17:21
and they are top notch, no question. We have a manager who runs
17:23
one of our core teams who is ex Special Forces and there's nothing that phases
17:27
him. So I think there's something about the Israeli way that I think is
17:32
actually very conducive to this field. And I think that it is really important
17:36
to counteract this rich, get richer phenomenon, at if it comes at the
17:40
expense of everyone else. Right, I think that there's a the world we
17:42
should shoot for is one where all the boats are rising, and I think
17:45
that there will be some sort of inequities in the landscape, and I think
17:48
that's something to address, and that's something that I think that leaders such as
17:52
yourself, I think need to be very cognizant about. You know, I
17:53
think that we've talked about UBI like opening actually has run. I think the
17:56
world's most comprehensive UBI. Something people talk about universal basic income, the idea
18:03
of just the government just distributes money, right, that's just what it does.
18:04
If most people are not working, producing, you know, dollars themselves,
18:10
and there's a world of plenty. Is that a mechanism that can work?
18:12
Maybe it can, maybe it can't. There's a lot of assumptions that
18:15
would go into a world where it works. It's clearly I think also has
18:18
a lot of downsides to it. But I think that we need to have
18:21
creative solutions because again, I think that just like defense versus offense might fundamentally
18:26
change economic production and what it takes to actually build even a stable society I
18:30
think is going to fundamentally change. Well, Ellen, you said when I
18:34
asked you this question one of our nighttime conversations. I asked you, well,
18:37
what are you gonna do when people don't have jobs and we shell out
18:41
money for them and you said. I think you said, oh, what's
18:45
bad and living in paradise? Right, I think you said that, Yeah,
18:48
I mean it's I mean this is a question of like, you know,
18:52
the concept of heaven. Now, in heaven, I don't think you
18:55
need to work in heaven. Haven't read that anywhere, So they're not laughing,
19:00
you know if you want so, you know, the very positive scenario
19:04
of AI is actually in a lot of ways the description of heaven in that
19:10
really nobody would need to work. I wouldn't even quote universal basic income.
19:12
I'd say it's probably universal high income. I'm describing the best case scenario here,
19:17
so I'm not saying this is definitely what occurred. There's a range of
19:19
scenarios from very negative to very positive. Very positive scenario is basically sounds like
19:25
heaven. You can have whatever you want, you don't need to work,
19:27
you have no obligations. Any illness you have can be cured. And when
19:32
do we die? Well, that's a good question. I think it'll be
19:33
a choice. I think it probably ends up being somewhat of a choice.
19:36
So you want this world, well, I'm not saying I don't know.
19:40
There's some there's a sort of a philosophical debate is the concept of heaven as
19:44
it is normally described. Actually, what do you want? I mean,
19:47
do you want to be in a future or world where you know the in
19:51
the AI sense that a I could do everything better than you by far?
19:55
That's that's because you know you have, like man, search for meaning questions
20:00
thing good work and what meaning if you if, like I said, if
20:03
I can do everything better than you do, we live a life of hedonism.
20:07
I don't know, it's hard to say, but that is where things
20:11
ahead. In a positive scenario is that there would be no scarcity of goods
20:15
and services. The robots and computers can make as much as much of whatever
20:18
you want, for as many people as you want, with really no limitation.
20:22
Any scarcity we have would be artificial, meaning like if you define that
20:26
in particular artwork done by a human is unique and special, then it's artificial
20:32
scarcity. But there will be no actual scarcity unless we define it arbitrarily,
20:37
define something to be scarce. So I think for a lot of people that
20:41
is a good, good future. I might be personally a bit frustrated because
20:45
I'm like, well, what am I supposed to do? Now? You'll
20:47
think of think of something. I think that this is a very interesting set
20:51
of challenges. What would we do if people have no jobs? But I
20:53
think it's important to remember that we're not automatically going to end up in this
20:59
situation in the way we're thinking about it now. We could also, I
21:02
think, very likely end up in a situation where people have no jobs because
21:06
there are no people anymore, where we literally go extinct. You know,
21:08
if we had a statement cut that came out in May, right that I
21:12
think Sam Olmen and you both signed right saying and ours left under Lyon came
21:17
out with it a couple of days ago saying we have to take seriously human
21:19
extinction. And if this sounds like very crazy science fiction, you know we
21:23
were joking. But just a few years ago all your success sounds like science
21:27
fiction. Two and you said that we don't have a lot of time.
21:30
These are things that could happen relatively soon. And there's how it's very important
21:36
to me, because I want the good future, to think about how we
21:40
steer away from those sort of scenarios. One way in which we could end
21:42
up and this kind of bad future is pretty obvious. The idea that we
21:48
can lose control to something more intelligent is not new at all. That Alan
21:52
Turing himself said this in the fifties. That's kind of the natural thing to
21:56
expect. That's what always happened on our planet before. When a news more
22:00
speechies came in, it took control over the older ones, and then you
22:03
know, bye bye and Neanderthals by by mammoths whatever. More concretely, I
22:07
think, even though Hollywood is full of scenarios where someone somehow loses control of
22:14
AI and it takes over, there are more sneaky ways in which we could
22:18
end up in this sort of bad state, even more willingly. So for
22:21
example, right now, I'm a big fan of the free market too,
22:25
you know, But it's the fact that if people are competing against other people,
22:30
they will often seed control voluntarily at the machines to get an edge on
22:33
the other guys. So companies who replace workers with machines out compete companies who
22:38
don't. Armies who seed control to machines out compete armies who don't. Once
22:44
we have really powerful any of that, that's better at running companies and countries
22:48
and humans, you know, than countries that replaced their prime ministers by AI
22:52
might outcompete others, and then we end up in this future where all these
22:56
things are happening, but it's not our future anymore, and we're like,
22:59
wait a minute, where did we go along wrong? Here? It was
23:02
supposed to be for us. But that's the point, you know, forgiving
23:04
for borrowing a page from your book, literally a page from your book,
23:08
I'm confessing I'm a speciest Okay, Okay, So I'm for the human species.
23:15
Okay. We can talk about other evolutions and you know, machine launched
23:18
intelligence and so on, but I'm you know, I'm for you know,
23:22
in the Jewish tradition, there's a saying, how do we translate this?
23:26
The impoverished of your own city, of your own town, proceed the impoverished
23:30
of other towns. So the human species, as far as I'm concerned,
23:33
is something that I'm vitally interested in maintaining. And I'll tell you that the
23:37
problem we have is not merely extinction, which I hope you can, maybe
23:41
you can extemporaries just elaborate on. But here's something else. Even if we
23:47
don't go extinct, how do we define ourselves? You talked about that,
23:51
Greg, Okay, so we defined ourselves. We had this myth in the
23:53
Bible, we had heaven, you know, paradise you can just pick off
23:59
the trees, you know, pick off the fruit from the trees. You'd
24:00
have to do nothing, and then the serpent comes and messes things up,
24:06
and then we're condemned to toil for our labors. Life is a struggle.
24:10
It's defined as a struggle to define all the time, as a struggle where
24:14
you're competing with the forces of nature, or with other human beings or with
24:17
animals, and you constantly better your position. This is how the human race
24:22
has defined itself, and our self definition is based on that, both as
24:25
individuals, as nations, as humanity as a whole. Now that could be
24:30
challenged, and it is challenged. So it's both our self definition throughout human
24:34
history and evolution that is being challenged, and also the question of our continued
24:40
existence, which you know you can please explain why you think we're in danger
24:44
of extension. Well, well, I was gonna say. I mean,
24:45
I think this whole arc in my mind is all about paradigm shift, right,
24:49
And I think that even the question of what would that heaven push agi
24:55
positive future look like? I think even that is hard for us to imagine
24:57
what the true upside could be because it's not just material abundance. It's not
25:02
just we all get nice cars or something hopefully I'll tell us, but also
25:06
that everyone gets a great education, right, you know, everyone has that
25:08
story of that one teacher that they had who paid attention to them and inspire
25:11
them on a specific subject. Imagine if you had that teacher for anything you
25:15
wanted constantly, and just how much better if people would we be? How
25:18
much better would we relate to others? Like that is just one example of
25:22
the kinds of upsides that are possible. And I think the question of well
25:26
what does it mean to be a human? You know, if you back
25:27
to how do you predict what's going to come next? Actually the thinker who
25:32
I think had the best foresight about how the AI revolution was going to play
25:36
out is actually Ray Kurtzwile you know he yeah, and his book Singularity is
25:41
near gets like a lot of shit. I think that people that you know,
25:44
kind of assume it is going to be this almost like just sort of
25:45
religious text, but instead it's a very dry analytical text. And he just
25:48
looks at the compute curves and he says this is the fundamental unlock or of
25:52
intelligence. Everyone thought that was crazy, and now it's basically true. It's
25:56
basically common wisdom. And part of what he says is, look, what's
25:59
going to happen is in twenty thirties. First of all, he says,
26:00
AGI twenty twenty nine. Yeah, I keep telling people it seems to be
26:03
almost exactly right. It's spooky. It's spooky. Twenty thirties is when the
26:07
merge happens. So we've got neurallink coming, and you know, maybe other
26:11
systems like that, And what does it mean once you actually are kind of
26:15
merging with an intelligence. Well, neuralalink is necessarily moves slower than AI because
26:21
when when have you put a device in a human? You have to be
26:22
incredibly careful. So I think it's not clear to me that the neuralink will
26:26
be ready before a GI. I think AGI is probably going to happen first,
26:30
but you know, we'll have the sort of AGI singularity, you know,
26:34
so as digital superintelligence is called like a singularity, like a black hole,
26:38
because just like with a black hole, is difficult to predict what happens
26:41
after you pass the event. Arison of black hole, and we are currently
26:45
so you know, circling the event horizon of the black hole, that is
26:49
at digital superintelligence, the event horizon. So I bet part of the reason
26:55
for neurallink, although initially we're very helpful to people who have brain or spine
26:59
injuries, just doing basic stuff like enabling people who are tetraplegics or like a
27:03
Stephen Hawking to communicate actually even faster than I'm communicating with a fully functional body.
27:10
So and ultimately to improve the bandwidth between the cortex and your AI version
27:18
of yourself. In fact, I think a lot of people perhaps don't quite
27:21
realize that they're already a syebalk. So you've got your lymic system, your
27:25
sort of basic drives, your cortex, which is the thinking and planning,
27:29
and then you have tertiary layer, which is your computers, your devices,
27:30
your phones, laptops, all the servers that exist, the applications. And
27:36
in fact, I think probably a lot of people have found that if you
27:40
leave your cell phone, take it away, if you get if you forget
27:42
your cell phone, it's like missing Limb syndrome. You know, you've like
27:47
read that thing, go Losing your cell phone is like missing Lin syndrome.
27:49
So, because it is your cell phone is extension extension of yourself, the
27:53
limitation is badwidth. So you the rate which you can input or i should
27:59
say output in nation into your phone or computer is very slow. So the
28:03
phone, it's really just the speed of your thumb movements, and with you
28:07
know, best case scenario, you're a speed typist on a keyboard. But
28:11
even that data rate is very slow, which we're talking about tens, maybe
28:15
hundreds of bits per second, whereas a computer can communicate in trillions of bits
28:19
per second. So so the and this is admittedly somewhat of a you know,
28:25
hail Mary shot or whatever lung shot, is that if you can improve
28:27
the bad with between your cortex and the your digital tertiary self, then you
28:33
can achieve better cohesion between what humans want and what AI does. At least
28:38
that's one theory. I'm not saying this is a short thing. It's just
28:41
one potential iron in the fire. If ultimately, you know, one hundreds
28:45
of millions or billions people get a high bound with interface to their digital tertiary
28:51
self, their AI self effectively, then that that seems like that probably leads
28:56
to a better future for humanity. So that's this is I'm getting somewhat esoteric.
29:00
Hopefully some of this is making sense, and like I said, it's
29:03
a long shot. But you know, what I find interesting is that I've
29:07
I've not found any human who wishes to delete either their cortex or their limits
29:11
system. I probably I suspect nobody in this room wants to do that.
29:14
It's a role or their cell phone. Well, now I've met people who
29:18
want to who don't use a cell phone. It's rare. Okay, Well,
29:21
I confess for years I avoided this, and I avoided it. I
29:25
didn't even have Not only did I not have a cell phone within my immediate
29:30
radius, I didn't have a television set. My phones were thirty five year
29:34
old phones because I was Prime Minister Visual and I knew what this means.
29:38
Okay, I mean maybe your audience doesn't know, but we knew. Now,
29:44
having spent some time in the opposition, I lapsed into the use of
29:48
the cell phone, and I find it very hard to dissociate from it.
29:51
For the reasons that you said. It's become an extension of, you know,
29:53
a repository of memory, the ability to get information very fast, which
29:59
you say correctly, is very slow, but compared to what it was before.
30:03
Remember you had research assistance, have to go to libraries, things like
30:07
that. Where is that that's gone? But this will be seen as very
30:11
vastly primitive compared to what you're talking about. Yeah, I'll just add something
30:15
is that to max? But the I mean, it's it's sort of a
30:18
funny thing like if you assume, like a best case scenario, imagine if
30:21
if you're the AI and you're trying to you just want the human to tell
30:25
you what it wants, just please spit it out. But it's speaking so
30:30
slowly like a tree. Okay, Like trees communicate, Okay, they if
30:34
you watch a tree, like a you know, sped up a version of
30:37
a tree growing, it's actually communicating. It's communicating with the soil. It's
30:41
trying to find the sunlight. You know, it's reacting to other trees and
30:45
that kind of thing very slowly. But from a tree standpoint, it's you
30:49
know, not that slow. So so what I'm saying is we don't want
30:52
to be a tree. That's that's the idea behind a high band with the
30:56
neural interface is just even when the AI it desperately wants to do good,
31:00
good things for us that we can actually communicate several orders of attitude fast,
31:04
and we currently can. I love that you mentioned the trees there, because
31:07
if anyone is still struggling to understand why it's so likely that we could get
31:11
wiped out by AI, you know, as superintelligent AI, we are like
31:17
trees. They are as much faster than us, and they're thinking than we
31:22
are. The trees. So if some trees in the rainforest are a little
31:23
bit worried that some humans are going to come chop them down, you know,
31:27
and they're like, Oh, don't worry, We're so smart, we'll
31:30
stop those humans. Good luck with that, right. I love how you
31:33
went philosophical here and talked about what it means to be human, and I
31:37
wanted to follow up on that, and but before that, maybe just continue
31:40
that. Okay, So, how you're assuming that the danger of human extinction
31:45
is that you have machines, supermachines, superintelligent machines that outpass humans by orders
31:51
and orders of magnitude who will not necessarily value the continued human existence as we
31:57
know it, and basically program is out. That's basically what you're saying,
32:01
so, if we were to be so dumb as to create the entities that
32:07
are much more powerful than us that don't share our goals, yeah, that's
32:09
a really bad idea. I mean, in Israel, I think you understand
32:13
really well that if you have a bunch of other beings who are intelligent and
32:17
don't share your values, you know, it can go really badly, right,
32:22
and let's not make that mistake by giving too much power to machines don't
32:25
share our values. But of course we can tackle that in various ways.
32:30
So coming back to what can we concretely do, I just want to say
32:34
one thing about the philosophical point about being human, just for myself. One
32:37
of the things that I really cherish the most about being human is the agency
32:42
that I can actually make choices and make a difference. And one of the
32:45
most beautiful things about what you were all doing with technology is giving us more
32:51
agency. Back in the Middle Ages, we had a life expectancy of freaking
32:54
thirty years and knowing that we might starve to death next year because of the
32:59
weather. There, you know what kind of it's so much more inspiring now
33:02
that we're becoming more of the captain of our own ship with technology able to
33:06
actually have more agency as a species. So I say, let's use that
33:08
agency to create a wonderful future, not to throw it away. And three
33:13
things I wanted to just suggest for concrete action items we can do here.
33:16
I'll say a little more about each of them. One has to do with
33:20
pausing risky stuff, one has to do with controlling the machines, and one
33:23
has to do with inspiration for the future. So for the controlling, we
33:29
already talked quite a lot about the importance of staying in control over these machines,
33:32
so I don't need to doell on it anymore. But we are I
33:37
think it's saying things like, oh, what is going to happen to us
33:40
bring out the popcorn. That's too passive for me. Let's celebrate our agency
33:44
and ask ourselves not what we think will happen, but what we want to
33:47
happen. And clearly we want to control these machines we're building them. Second
33:52
thing, so, I, you know, agree with Gregg on many things,
33:57
and I greatly respect your work. One thing where we have a friendly
34:00
disagreement is you know, I signed this pause letter saying I think we should
34:04
pause the riskiest things, and you don't. I think that it would be
34:07
a good strategy for us to emulate what biotech does and just have clear safety
34:12
standards. You know, if you want to release a new drug, you
34:14
can't just sell it in the supermarket. You have to do a clinical trial,
34:17
persuade some experts somehow that this is safe. And I would like to
34:21
see really high stakes, very powerful systems, some sort of system like this
34:25
with standards, and then the companies have and it's on them to, you
34:30
know, prove that they're safe before they get released it. Who does the
34:31
standards? Well, and we've solved this, It's been solved in Israel.
34:37
It's been solved in the US for biotech, not for biotech. But here,
34:40
you know, it's a's a do you want it? Regulators? Who
34:44
do you want it? Legislatures? I mean, it's not the only choice,
34:46
but you know, it's a pretty pretty clear dichotomy. I don't think
34:51
we need to reinvent this wheel. The people who decide whether a drug gets
34:53
approved in Israel is not people in parliament in the Kinnessets, right, it's
34:58
Israeli scientists, but the government involved in approving them. Exactly we can emulate
35:02
that, and Elon can tell us all about how we deal with the cars.
35:06
I don't want to dwell on, and I just think this is not
35:08
rocket science. We can put these standards in place and then we can continue
35:14
doing all sorts of wonderful stuff with a tech. Some stuff will wait a
35:16
little longer for but it'll be worth the wait because it'll be safe. The
35:21
last thing I just wanted to say is I think for us to really seize
35:25
this agency and work together, not just between companies but also across borders,
35:30
we really need a shared positive vision. And this is something I really appreciate
35:36
so much about You, Elon have the guts more than most to talk about
35:38
like really audacious positive visions for the future. What is it that makes people
35:45
collaborate and put little squabbles aside. It's always a shared positive vision where people
35:51
realize that if we collaborate, we'll all be better off. You know,
35:53
technology is not a zero some game, right. You can take countries like
36:00
Sweden and Denmark that had been killing each other for centuries and flat GDP,
36:06
you know, and then technology came along and they both got dramatically better off.
36:09
It's so obvious that the technology that you're building. You're building you know,
36:14
can make all humans just spectacularly better off with AI, and not just
36:19
on this planet. I like the list before we started you had there about
36:22
how it can revolutionize healthcare and so many other things. And I would add
36:25
to that, you know, why limit ourselves to this planet when there's this
36:29
beautiful universe where life can flourish for billions of years. If the more we
36:32
can can just appreciate how huge the upside is if we do this right,
36:37
you know, the more incentive we will have to actually be a little patient
36:43
where we need to be a little bit patient, make everything safe and therefore
36:46
get this wonderful future. Well, maybe we're sort of lagging behind in our
36:52
social and political development. This is like having nuclear technology and the Stone Age,
36:57
what do you do with it? Which I think is you know,
37:00
because more That's the thing that I talked about in terms of the pace of
37:02
development, pacing basically what solutions we need to put in place to maximize the
37:07
benefits and limits the risks. I think we're there. I think we should
37:12
have a meeting of minds. You were in you were in the Senate the
37:15
other day. I presume you were talking about this. I mean, yeah,
37:17
you don't have to divulge what you can divulge, but divulge would you
37:22
can? Well, a lot of people in the meeting, so I don't
37:23
think it was secret. It was just a discussion on AI regulation. And
37:28
actually Sam Often from open I was there and a number of eye leaders were
37:31
were there. And I think generally it would be a good idea to have
37:36
some kind of AI regulatory agency and start off with a team that gathers insight
37:43
to get a maximum understanding. Then you have some proposed rulemaking, and then
37:47
eventually you have regulations that are put in place. And this is something we
37:51
have for everything that is potential danger to the public. So if it is
37:53
you know, if we're in drugs in drug administration, we got aircraft with
37:58
fa and rockets. You know, there's ever anything that is a danger to
38:01
the public. Over time, we have learned that often the hard way,
38:06
after many people have died, to have a regulatory agency to protect public safety.
38:12
I'm not someone who thinks that regulations some panacea where it's only good.
38:15
Of course, there are some downsides to regulation and that things move a bit
38:19
slower, and sometimes you get regulatory capture and that kind of thing. But
38:22
but on balance, I think the public would not want to get rid of
38:28
most regulatory agencies. And you can think of it also as like the regulatory
38:32
agency being like a referee. You know, what sports game doesn't have a
38:36
referee? You need someone to make sure that that people are playing fairly,
38:40
not breaking the rules. And that's why basically every sport has a referee of
38:45
one kind or another. So that's the rationale for AI safety. And I've
38:50
been pushing this all around the world, and when I was in China a
38:52
few months ago meeting with some of the senior leadership, that my primary topic
38:57
was AI safety and regulation and they, after we had a long discussion,
39:01
agreed that this merit to AI regulation and immediately took action in this regard.
39:07
So so sometimes it will get this comment of like, well, if the
39:12
West does AI regulation surely, then what about what if China doesn't and then
39:16
leaves ahead. And I think they are also taking it very seriously, because
39:21
you know the opposite of what of a moral constraints you program into the system?
39:24
I mean, Greg, I don't know what you what are your thoughts
39:27
on this? Yeah, I mean I think again it's just like this technology
39:30
is almost backwards from how you'd expect it to work. Like I think that
39:32
a lot of people talk about the paper clips problem of if you think of
39:36
that as if we literally program into a machine make as many paper clips as
39:39
possible, and it gets really smart, and it's like, well, there's
39:42
all these houses and factories and other things, they're not producing paper clips.
39:45
They're all going to be just paper clip production. Now, all these humans
39:47
that are in the way paper clip would be a better arrangement of their atoms.
39:51
And that's just not the technology path we actually seem to be on.
39:54
You look at GPT three, and it is a system that gets the nuance,
39:58
It gets the common sense. It has other problems, right, it
40:00
makes mistakes that no human ever would like kind of loses this chain of thought
40:05
and says all these weird things and sometimes make things up like we have real
40:07
problems we have to encounter. But I think that for me, that the
40:10
hard thing that I think gets lost in some of the conversations is the nuance
40:15
right of really being coupled to what's the technology we actually see and particularly how
40:19
does it depart from our intuitions about where we thought it would go. I
40:21
actually think it's even interesting. You know, Max mentioned the pause letter.
40:23
The pause letter, one of the key parts of it was to not train
40:27
GPD five for the next six months, and we looked at that and we're
40:30
like, well, we're not training GPT five for the next six months anyway.
40:34
And so you know, it's just like again if you kind of miss
40:36
the sort of nuances of what's going on at a mechanical level. And some
40:39
of this does require getting like really in the weeds and coming and talking to
40:44
us and also opening up and showing kind of in progress work. And I
40:46
think that's something that will be very important as you think about moving to these
40:51
very powerful systems. And I also think there's another blessing and curse of how
40:54
the technology is playing out, where we talk about digital superintelligence and we also
40:59
talk about chat. These are very different things, but somehow there's going to
41:02
be a smooth continuing between them. And so I think that the blessing is
41:07
that we see a little bit into the future. We have a little taste
41:08
of a system that you know, I talked about a personalized tutor for everyone.
41:13
You kind of have it now. It's not a perfect tutor. It's
41:15
got a lot of problems, but like you know, you can go use
41:17
shat GPT for free to ask about anything and it'll give you something. Sometimes
41:21
it's actually be pretty insightful. Well fair enough, I'll withdraw the seven and
41:25
a half, but but you'll get there. You'll get to a ten.
41:28
We will, we will, we will. Definitely, that's right, And
41:30
I think it's great for people to see it while we can still laugh about
41:34
it. Right. I think that when it starts coming up with the new
41:36
ideas, you type in some problem you're having at home, trying to figure
41:38
out how to do something, and it just comes back with, oh,
41:40
you should say this to this person, and you should do that, you
41:43
should you know, here's say you should revenge your schedule. And it's just
41:45
like great ideas, Like that's gonna feel so different. And then one day,
41:49
when we're actually having a system that can actually just autonomously go off and
41:52
do a bunch of research and say, here's this crazy chemical formula, like
41:55
please mix them together in these ways. You want to make sure that thing
42:00
trustworthy right before we start mixing those chemicals. And so I think that this
42:02
question of how do we separate out the sort of diffusion of today's AI technology
42:07
and application and really, you know, all the debates that we're having,
42:10
let those happen, open source technologies like everyone's building. We should not squelch
42:15
those, we should actually promote those. But I think that also looking ahead,
42:19
being ahead of the game here and saying there is going to be a
42:22
point and it's where we're hitting that wall, where we're going to be building
42:24
the most massive machines that humans have ever built. Like Youan talks about these
42:29
massive rows of data centers with all these computers, like like it's one of
42:32
those things where you kind of have to see it to really feel it.
42:36
Like I've walked through some of these data center halls and you just get a
42:38
sense of the scale of what's being built. And it's very different when it's
42:42
abstractly just like you're looking at, oh is this many GPUs and you know,
42:45
looking at numbers, it's like you just realize the mechanical force that is
42:47
behind what you're building for cognitive intelligence. And I think that kind of thing
42:52
we need international answers for. I fully agree with you. I think,
42:57
Chris, of all you know, with all due respect of the pause button,
43:00
we ain't going to get it. It's not going to happen. I
43:02
think you can't put the genie back into the bottle, and I'm not sure
43:06
you can slow it down. The only thing you can do is just to
43:08
accelerate our own efforts. And I think you're right. It has to be
43:12
international. I think there. I think you start with a like minded and
43:15
like valued states. Start with that. And frankly, I think you know,
43:19
obviously, you know what AI stands for. It stands for America and
43:21
Israel. Obviously, I'm just plugging my own country. But it's a it's
43:25
not an empty plug, because I think that there's a concentration of talent an
43:30
Israel that makes it more than a mouse compared to the American elephant. It's
43:32
more than that. It's a very nimble mammal, very fast compact, and
43:37
it's it's also got this feed of military AI which moves very fast and takes
43:45
a very large part of the population. The talents there screens it and it
43:49
trains it very quickly. So I think that you start with a like minded
43:52
states, you definitely should have a very robust conversation. First of all,
43:58
decide among us what is something there are things that we wouldn't want to share,
44:02
and it's obvious related to our security. There are standards that we have,
44:07
ethical standards and other regulations that we have to achieve quickly among between industry
44:12
and governments quickly. And then I think we have to conduct a robust discussion
44:16
with the other powers of the world based on their self interest, as you
44:22
began to do. And I think that's a pioneering work, and I think
44:24
we have a shot maybe at getting to some degree of control over our future,
44:30
which could be amazing. You know, I'm not sure I want to
44:34
live in heaven. It could be boring. I don't mean the boring company.
44:37
I mean it could be boring, But I think we're headed. I'd
44:39
much rather have a boring heaven than an interesting hell. So here at Tesla,
44:45
they have a different name for pause button. They call it the break.
44:49
And when we launched this pause letter, right, I personally didn't think
44:53
that there was immediately going to cause a pause for me. The real purpose
44:57
of it was the mainstream the conversation about these issues, which really did happen.
45:01
And you know, I really appreciate you and Sam and your colleagues for
45:06
saying very openly that there may come a time when you will want to pause
45:09
certain things if you meet certain criteria. Right now, if Elon says that
45:15
may come a time when he's going to slow down his Tesla, I would
45:19
feel I would trust that more if I knew that Tesla actually had breaks in
45:22
it, so that they had like a mechanism by which you could actually slow
45:25
it down. So this is something I'm very curious to actually to hear it
45:29
from you about if what criteria and would you say that you all have can
45:35
can you say it in the future, say it already now, Like what
45:37
are some things which would make you want to pause? And and then won't
45:39
it be difficult with the actual mechanism, like to resist commercial pressures to continue
45:45
even if you and Sam want to pause. Well, I actually kind of
45:47
agree with both of you right that, I think that my picture of how
45:52
the future is going to play out in terms of AI development is not as
45:55
simple as oh, you're scaling models and then you see something a little bit
45:59
weird and then you like stop all scaling and you all go home or something,
46:01
right'. It's it's definitely not going to be that. I think that
46:04
we need to be making maximum velocity efforts towards a good future, right,
46:09
And I think that is something that you don't want to pause. You actually
46:12
want to put the foot on the pedal. Now, of course, the
46:15
question is what actions actually get you towards there, which ones are against it,
46:19
you know, Like I think that nuance is something that I think is
46:21
what we should be debating. But that's why I actually bounce a little bit
46:23
off of the framing of a pause, because I think even to the point
46:27
of GPD four, doing all this prediction work ahead of time, like if
46:30
that prediction work was a little bit wobbly, weren't quite there, and we
46:32
actually do this all the time, that there's like some experiment that came back
46:36
a little bit strange, we're not really sure we understand it. We actually
46:37
spend a lot of time to fix that. And it's not even you know,
46:40
even if you forget about the safety reasons, and actually one of the
46:44
reasons we do it is safety motivated. But if you even forget about that,
46:45
you're just not going to build a good system. And so I think
46:47
that there's a lot of places in AI where there's actually great alignment between capability
46:53
and safety, between building something useful and something that is safe. It's not
46:57
perfect, but I think it is actually surprising where there are divergences. Actually,
47:00
the single biggest divergence we've seen has not been owe too much pressure to
47:05
deploy quickly because I actually think that getting things out, Like you can see,
47:07
the market it hates hallucinations, It hates it when the model makes something
47:10
up. All the jokes we made about oh it doesn't give you good insights,
47:13
the market hates that stuff, right, And actually I think it's very
47:15
aligned with safety to fix them. But what the market actually pushes us against
47:20
is actually when it comes into conflict with other values, for example privacy.
47:22
Right, So that the way we launch the initial gpt APIs was that we
47:27
said, hey, we're going to monitor everything. We're going to record everything,
47:29
like if something sketchy comes up, we'll go and look and we'll be
47:31
able to look back in the logs. We'll be able to see what you
47:34
did. People hate that, right, like you know you kind of want
47:37
that for everyone else, but for yourself you want privacy. And so how
47:39
do you balance those two factors? And we actually see this very much in
47:44
terms of to actually win deals to decide to be able to give companies the
47:47
assurance that hey, we're not going to retain any of your data, will
47:50
delete it immediately. That removes our ability to do any investigation post talk.
47:53
And so this is actually a place where I think government can be very helpful,
47:55
where norms can be very helpful to say, hey, even though you
48:00
know that this is what would be you know commercially most viable, that this
48:04
is what the rule is going to be, and everyone has to play with
48:06
it. You don't ever race to the bottom. So I think that again,
48:08
it's all going to be nuanced, it's all going to be I think,
48:12
like I've just accepted that this field it's totally counterintuitive, Like everything we
48:15
learn this is like it just would make you just like if you told yourself
48:20
ten years ago you have said, there's no way. That sounds totally crazy,
48:22
and somehow we have to navigate through that. Well, the only thing
48:25
that I can add to this conversation is to invite you all to as well.
48:30
I'd like to, in fact continue this conversation, in fact, bring
48:34
a few more of your colleagues and a few leaders to discuss this international strategy
48:38
for like minded states to begin with that, you know, first like minded
48:44
states, then the world. Okay, how about that as a plan of
48:46
action. But I think that I think that the work that you're doing is
48:52
amazing. Nothink short of it. You know, Elon was saying, what
48:55
is a human being? It's sort of this two legged thing ape like true,
49:00
with a cortex and a computer and its wants, desires and a brain.
49:05
But the brain is developing in ways that we couldn't possibly imagine, and
49:07
it just doesn't stop. Well, I mean, from my other comments,
49:12
like the thing that paps should provide some hope for digital not being controllable in
49:16
a very very specific way but acting in a way that is beneficial to humans
49:21
is that our cortex and our limc system, the cortex is actually in service
49:25
to the limits system. So you know, the cortex, the thinking part,
49:29
is constantly trying to make the instinctual part happy, even though it is
49:32
much smarter than the instinctual part. And so I think there's hope there that
49:37
digital superintelligence, even though it is much smarter than our cortex, will none
49:40
less want to make the cortex happy and ultimately LIMICS system happy. I mean
49:45
not to get too off color here, but the sheer amount of thinking and
49:49
about two that people have done to have sex is a lot, but not
49:53
sex will procreation, just for the to make the limics system happy. So
49:59
you know, given how much you know thought has gone into getting laid essentially,
50:02
but for the sole purpose of getting laid, not procreation, and all
50:06
that does is make the limbic system, you know, give us block of
50:08
joy that gives. I think that's actually suggests that there is a good future
50:14
where even if we don't understand the digital superintelligence, just as the Olympic system
50:19
does not under understated the cortex, it might still actually at the end of
50:22
the day, if you're trying to make our limbic system and cortex happy,
50:24
is that a happy note to end? I think it's a great note to
50:27
end on. Thank you, Thank you everyone, Thank you athletes. So
50:30
that's really amid