WEBVTT
00:00:08.329 --> 00:00:15.048
Hello everyone And welcome to the switch statement podcast It's a podcast for investigations into miscellaneous tech topics
00:00:16.884 --> 00:00:20.484
This is episode three of"The Coming Wave" by Mustafa Suleiman
00:00:31.434 --> 00:00:32.473
Hey, John.
00:00:32.503 --> 00:00:32.823
How you doing?
00:00:33.808 --> 00:00:34.988
Hello, Matt.
00:00:35.279 --> 00:00:36.548
I am doing very well.
00:00:36.558 --> 00:00:37.168
How are you doing?
00:00:38.344 --> 00:00:39.264
I am doing all right.
00:00:39.274 --> 00:00:47.191
Actually, I'm on cloud nine right now because, you know, we got an amazing suggestion.
00:00:47.191 --> 00:00:52.680
You know, I was going through the mailbags, uh, of our fan mail
00:00:52.963 --> 00:00:54.162
Yeah, I saw this
00:00:54.856 --> 00:00:57.695
and Alex, I hope you're, you're okay.
00:00:57.695 --> 00:01:00.045
That's all we're gonna say without saying your name.
00:01:00.045 --> 00:01:09.525
But, um, he had a fantastic idea, which was that we should do a better job of incorporating the listeners in our book selection process.
00:01:10.135 --> 00:01:18.394
and so, you know, he said maybe we should do a poll or at least give a heads up a couple, you know, in anticipation of the next book.
00:01:18.745 --> 00:01:20.555
And I think those are both fantastic ideas.
00:01:20.555 --> 00:01:40.754
And what we're planning on doing, we don't have a, we don't have a domain or anything like that, but we think we're gonna build a little, a little web app where people can go and, you know, we're both coders, like we're just gonna vibe code some shitty web app that allows people to go and vote up our candidate books
00:01:41.777 --> 00:01:42.277
I love this idea.
00:01:42.346 --> 00:01:47.777
I love this idea because I feel like we often kind of struggle to pick a book.
00:01:47.826 --> 00:01:58.887
We're like,"Eh." And, and this'll, I don't know, this'll just be more fun for us, I feel,'cause we'll, we might have to do books that we've never heard of, that are interesting to us.
00:01:59.537 --> 00:02:00.307
So it'll be cool
00:02:01.055 --> 00:02:08.955
That is the part that I am most excited about, is getting exposed to books that we never would've found otherwise.
00:02:09.425 --> 00:02:23.634
Um, th- the other thing that's funny, whenever we're picking a book, it's pretty clear that both of us wanna spend a- the absolute minimum possible time we can picking the book, which is ironic given the fact that it will likely
00:02:24.188 --> 00:02:25.979
We spent so much time.
00:02:26.004 --> 00:02:26.604
lives.
00:02:27.109 --> 00:02:30.058
Yeah, we just arbitrarily pick Godel, Escher,
00:02:30.224 --> 00:02:33.324
we can spend 45 seconds picking this book?
00:02:33.344 --> 00:02:39.504
Like,'cause,'cause I got shit to do," even though it's like, uh, gonna have major downstream ramifications.
00:02:40.305 --> 00:02:43.175
Uh, not the least of which is on the listener.
00:02:43.824 --> 00:02:46.004
So I, you know, I've always been a fan of democracy.
00:02:46.525 --> 00:02:47.365
They're now...
00:02:47.384 --> 00:02:53.685
They are just, uh, downstream recipients of our oligarchy of two.
00:02:53.935 --> 00:02:56.865
Our, our du- duarchy?
00:02:56.995 --> 00:02:57.664
Is that a thing?
00:02:57.911 --> 00:03:06.661
it's like an inverse duarchy because they're basically forcing us to do their bidding, which is how a, you know, how a democracy should work
00:03:07.550 --> 00:03:08.100
Exactly.
00:03:08.100 --> 00:03:14.800
Well, I'm just saying like the, in the system today, it's a, it's a duopoly or du- duarchy
00:03:15.275 --> 00:03:15.604
Yeah
00:03:15.691 --> 00:03:19.181
a monarchy and, and we're gonna open it up to the masses.
00:03:19.441 --> 00:03:21.901
And now we may have, we may have some,
00:03:22.411 --> 00:03:23.170
Do Umver
00:03:23.281 --> 00:03:23.730
know, we'll...
00:03:24.251 --> 00:03:24.700
What's that?
00:03:24.711 --> 00:03:24.721
Duumvirate.
00:03:26.091 --> 00:03:26.631
That's good.
00:03:26.640 --> 00:03:26.871
That's
00:03:26.932 --> 00:03:28.502
I'm just thinking of the triumvirate.
00:03:28.522 --> 00:03:31.073
I don't know if there's a similar word for, for two
00:03:31.838 --> 00:03:39.058
I think we don't have a word for it'cause it's just a s- just a stupid idea to just have two leaders'cause y- like what happens when they disagree?
00:03:39.465 --> 00:03:43.324
It's so strange though,'cause hasn't there been a few triumvirates?
00:03:43.335 --> 00:03:44.485
Like, there's Augustus...
00:03:44.564 --> 00:03:49.104
Or though m- maybe you're pointing out the, the reason why there's never a duumvirate.
00:03:49.901 --> 00:03:51.121
That's exactly what I'm saying.
00:03:51.121 --> 00:03:52.950
Yeah, sorry, that-- if that wasn't clear.
00:03:53.040 --> 00:03:57.871
Like, a triumvirate makes sense because at least then you have two people, like you can break ties.
00:03:58.354 --> 00:03:58.754
Yeah
00:03:59.800 --> 00:04:16.141
uh, so, um, but yeah, so what I was gonna say is I think we might have some mechanism to override the will of the people in cases where, you know, maybe, um,"My Little Pony: Friendship is Magic," the, you know, the book...
00:04:16.420 --> 00:04:22.920
Not to-- Listen, not to throw any shade on"My Little Pony: Friendship is Magic," but, um, and I'm not even sure if there are
00:04:23.009 --> 00:04:27.069
Like how you know an exact title of a My Little Pony book?
00:04:28.069 --> 00:04:33.238
Well, that was the name of the, that was the name of the series that really blew up with like Derpy Hooves and all that stuff.
00:04:33.249 --> 00:04:34.389
But, um,
00:04:35.036 --> 00:04:37.146
Wow, I have no knowledge of this.
00:04:37.867 --> 00:04:50.216
Now, I feel like our fans are gonna be like,"Can you read the every digit of pi book?" And then we'll be analyzing, like, digits 10 million through 10,000,500, and
00:04:50.269 --> 00:04:53.809
Yeah, it's like, it's like, you know, this, this section was pretty interesting.
00:04:53.809 --> 00:05:02.428
There were only, uh, you know, point, you know, 9.8887% ones in this section, and I thought that was, I thought that was pretty interesting.
00:05:03.201 --> 00:05:03.661
Yeah.
00:05:04.678 --> 00:05:05.408
uh, okay.
00:05:05.408 --> 00:05:07.389
But anyway, so that's, that's the end of that section.
00:05:07.389 --> 00:05:12.769
This is the kind of end of the house- housekeeping section, which we have heretofore never had to do.
00:05:12.769 --> 00:05:13.538
It's pretty exciting.
00:05:13.738 --> 00:05:21.468
'Cause,'cause as much as I was joking, this is actually, I think, literally the first time we've ever had had anyone seriously write in, uh,
00:05:21.555 --> 00:05:26.415
Yeah, we, we have I was wondering whether we were gonna admit that or not.
00:05:26.415 --> 00:05:33.944
We have received other emails, but they've been kind of like, I don't know, scammy, like whatever, and it just kinda makes us depressed.
00:05:33.944 --> 00:05:36.634
So, so thank you, Alex, for brightening our, our days
00:05:37.478 --> 00:05:37.889
Yes.
00:05:37.918 --> 00:05:38.959
No, we really appreciate it.
00:05:39.259 --> 00:05:48.288
And you sound like our, our ideal listener, which is funny'cause we've never actually even thought about, uh, what, what an ideal listener would look like.
00:05:48.309 --> 00:05:54.829
But, you know, someone who's trying to work with the material and go through it a- and reason about it in, in a deep way.
00:05:55.389 --> 00:06:02.608
Um, you know, that's kind of always how I imagined our, our listeners or the class of listener that would benefit from our, our podcast, so.
00:06:03.158 --> 00:06:03.858
Um, all right.
00:06:03.879 --> 00:06:06.848
But talking about containment
00:06:08.194 --> 00:06:14.394
The containment problem certain things cannot be contained, and technology is one of them
00:06:15.605 --> 00:06:24.105
Well, yeah, I mean, and it sounds like almost every technology cannot be contained, or almost no technologies can be contained.
00:06:24.802 --> 00:06:25.343
Yes.
00:06:25.413 --> 00:06:29.673
Yeah, so that's, that's basically the theme of this chapter is...
00:06:29.812 --> 00:06:45.084
The last chapter sort of touched on this, where technology proliferates i- in a sort of ridiculous way, and he gave the example of the engine, and now we have engines in literally everything from lawnmowers to kids' toys to things that you wouldn't even imagine to have an engine.
00:06:45.675 --> 00:07:00.992
And he sort of discussed how the original inventors of the engine, you know, they were thinking about cars, sure, which is a primary use case for an engine, but now there's so many things with engines they never would've conceived of, of any of these things.
00:07:01.290 --> 00:07:02.478
and this is sort of what happens.
00:07:02.497 --> 00:07:11.408
It proliferates, it reaches into all these nooks and crannies of our day-to-day life, and it becomes impossible to contain.
00:07:12.257 --> 00:07:17.817
And he gives these examples of, like the ancient Ottomans, for example, with the printing press.
00:07:18.158 --> 00:07:19.687
Like you have, Gutenberg.
00:07:19.728 --> 00:07:32.850
He came up with this printing press because he wanted a simpler way of churning out Bibles, which is, I don't know, just a testament to how, you know, people don't really foresee the full consequences of what they're doing.
00:07:33.168 --> 00:07:36.307
And the Ottomans were trying to control religion.
00:07:36.377 --> 00:07:51.334
They didn't want, religious propaganda to sort of come into their territory and, you know, they had their own views on religion, and they wanted to sort of dictate to their populace what the primary religion would be and, and whatnot.
00:07:51.843 --> 00:08:00.958
and so they tried to prevent people from having printing presses and tried to prevent, information being produced from printing presses from entering their territory.
00:08:01.389 --> 00:08:02.524
but this utterly failed.
00:08:02.735 --> 00:08:11.074
You can imagine how hard it is to prevent something like paper with ink on it, you know, from permeating into your territory.
00:08:11.714 --> 00:08:13.995
Uh, and so it's kind of eventually backfired.
00:08:14.651 --> 00:08:24.161
Yeah, and so just to, just to touch on a point you just talked about where it's like there's all these unintended side effects as technology proliferates.
00:08:24.701 --> 00:08:44.313
He even goes a little farther, which is to say that generally speaking, it's kind of the rule that an inventor invents a technology, and then it's actually used to do something that di- directly conflicts with the inventor's intention for the technology.
00:08:44.846 --> 00:08:45.427
Dynamite
00:08:45.464 --> 00:08:56.494
a crucial a- element of what he's saying,'cause yes, there's all these unintended side effects, but he actually refers to it in this more stark language, revenge effects, which I think maybe goes a little bit too far.
00:08:56.797 --> 00:09:15.017
but basically that even in the case of these maybe like less sentient technologies, the tech- the, the propagation of the technology serves to kind of conflict with the original inventor's vision for the technology.
00:09:15.405 --> 00:09:15.666
Yeah.
00:09:16.096 --> 00:09:29.778
Yeah, and that I, I feel like the famous example of that is Nobel with dynamite, where he developed that for mining, for, you know, making mining simpler, being able to chisel into mountains m- more easily.
00:09:30.168 --> 00:09:38.447
And obviously, it ended up being used for warfare, and this is part of the reason he developed the Nobel Prize fund is because he felt guilt
00:09:39.261 --> 00:09:39.682
Yes.
00:09:39.741 --> 00:09:40.751
Yes, exactly.
00:09:41.162 --> 00:09:41.682
Um,
00:09:42.346 --> 00:09:42.735
Yeah
00:09:42.961 --> 00:09:53.532
um, so as you called out, right, we have all these c- mostly state actors, it seems like, who are the ones who really wanna control the technology.
00:09:53.542 --> 00:09:59.591
It's interesting'cause he talks about, like, the technology goes on to conflict with the inventor's original intent,
00:10:00.148 --> 00:10:00.158
Mm-hmm.
00:10:00.572 --> 00:10:02.261
but they aren't the primary ones.
00:10:02.261 --> 00:10:09.471
And maybe it, maybe they do attempt to control it, and the, it's just such a meager attempt that it doesn't bear even mentioning.
00:10:09.481 --> 00:10:16.971
Whereas at least when you have a government that's trying to control a technology, they do a, at least a good enough job that it's worth talking about.
00:10:17.001 --> 00:10:17.261
But,
00:10:17.658 --> 00:10:18.548
I wanna give a, uh...
00:10:18.749 --> 00:10:20.899
a quick pop culture reference.
00:10:21.895 --> 00:10:26.464
He mentions the Tokugawa shogunate, which is another example of a...
00:10:26.964 --> 00:10:38.082
Just like you're saying, you know, the Tokugawa shogu- shogunate sort of had control over a large swath of Japan, and they attempted to turn themselves into an extremely insular society.
00:10:38.113 --> 00:10:42.783
Like, they sort of had these brief dealings with the outside world, and they were like,"We don't like this.
00:10:42.812 --> 00:10:52.477
We're going to, you know, close the gates, make our society completely, self-contained and self-sufficient." And the, and Mustafa sort of talks about how this backfired.
00:10:52.847 --> 00:10:55.998
But what I wanted to bring up was, did you watch the s- show"Shogun"?
00:10:55.998 --> 00:10:58.187
The, uh-
00:10:58.282 --> 00:10:59.211
not, no
00:10:59.788 --> 00:11:13.160
Okay,'cause there is a scene in"Shogun" that I actually found this scene incredible, where the main character, who is, like, this, I think he was, like, a Portuguese cap- like, ship pilot or something.
00:11:13.831 --> 00:11:20.150
He was explaining to this shogun how, like, Spain owned Japan.
00:11:20.509 --> 00:11:33.219
Uh, and he was, like, drawing this map in the sand, and he was like,"Oh, don't you know, like, this random country that you've never heard of owns Japan?" Which it's like, you th- you think you're the shogun.
00:11:33.979 --> 00:11:34.568
And I don't know.
00:11:34.578 --> 00:11:35.798
It was just this kind of, like...
00:11:35.879 --> 00:11:40.094
And, uh, you know, obviously it, it sort of made no sense to the shogun.
00:11:40.094 --> 00:11:41.668
The shogun's like,"What the hell are you talking about?
00:11:41.668 --> 00:11:52.687
We've never heard of Spain." But it was also this interesting example of how, like, yeah, sure, you might try to contain yourself, but the outside world continues to exist, and people, you know, sort of continue to make these global decisions.
00:11:52.865 --> 00:11:58.335
And if you're not at the table, like, fine, but you're not gonna have a word in these decisions.
00:11:59.274 --> 00:11:59.924
Uh, so
00:12:00.104 --> 00:12:20.438
and this is actually, I'm really glad that you talked about that, about the rest of the world advancing,'cause there was one example that he calls or that he calls out Queen, Queen Elizabeth banning the knitting machine, and it sounds like she was actually successful in banning this knitting machine, at least within England.
00:12:20.678 --> 00:12:35.408
But then, like, when it came back, there had been a bunch of these advances in technology, and it was, like, the mechanical looms, like these enormous mechanical industrial looms that came up and just, like, bulldozed over it.
00:12:35.597 --> 00:12:54.734
So it's like, even if you do succeed in suppressing this technology for a time, it's like all of these other advances are gonna go on and, like, they're just gonna come back with a vengeance and, like, just completely trounce whoever the group of people you were intending to protect, with, you know, with this, uh, protectionism.
00:13:00.991 --> 00:13:16.523
he, brings up w- the one potentially defensible case where the world has up until now managed to suppress, I guess, the propagation of a technology, which is nuclear weapons.
00:13:16.991 --> 00:13:17.471
Yes.
00:13:17.501 --> 00:13:18.052
Yeah.
00:13:18.121 --> 00:13:25.662
And we're seeing the mechanisms of suppression today with the wa- the war in Iran, which, you
00:13:25.876 --> 00:13:26.496
Right, and I...
00:13:27.384 --> 00:13:45.453
and that kind of dovetails with, and this was before, obviously before the war in Iran, but, like, is part of his central thesis about the, why this might be the one technology that we're aware of that hasn't propagated is just'cause it's really fucking hard.
00:13:45.514 --> 00:13:56.818
You know, it's like, it's just enormously difficult to do, and unless you have, a, like a wealthy state that can fund the creation of this nuclear technology, you just can't do it.
00:13:56.889 --> 00:13:58.649
It's just not, you know, it's just not possible.
00:13:59.633 --> 00:14:11.192
He, he gives, he gives another really thought-provoking example of why it's able to be contained, which is just that the normal market forces don't exist for nuclear tech.
00:14:11.793 --> 00:14:15.182
You know, nuclear tech is a civ level...
00:14:15.673 --> 00:14:18.253
You know, just like you're saying, it's a civ level effort.
00:14:18.263 --> 00:14:20.712
Like, you require thousands of people.
00:14:20.712 --> 00:14:34.293
You require, you know, whatever it is, millions of centrifuges spinning uranium to enrich it, and y- it's just you require this ridiculous amount of resources, technology, people.
00:14:34.756 --> 00:14:37.466
But no one wants a nuclear bomb.
00:14:37.797 --> 00:14:46.176
You know, it's not like, it's not like ChatGPT where you open it a million times a day and ask it about, like, a mole on your, you know, eyebrow or something.
00:14:46.186 --> 00:14:47.326
I-it's just...
00:14:47.346 --> 00:14:56.538
And no one, no one wants this thing, and, and what that means is there's no real incentive for people to, you know, proliferate it.
00:14:56.548 --> 00:15:04.288
Like, obviously there's an incentive for nations to have it because it's this deterrent and, you know, you can sort of saber-rattle more effectively.
00:15:04.644 --> 00:15:20.138
But just on the, on the scale of, like, human society, there's not this, like, market, you know, market reason to sort of, like, proliferate it and improve it and make it more economical like there is for something like AI.
00:15:20.740 --> 00:15:21.091
Right.
00:15:21.240 --> 00:15:21.971
Exactly.
00:15:22.057 --> 00:15:25.886
and yeah, so, so that i- it does seem to be one of the few...
00:15:25.897 --> 00:15:34.919
A- and luckily, r- right, we've only used it twice in, in a real war, um, which, you the United States has that dubious honor.
00:15:35.644 --> 00:15:50.745
One, one point that I wanted to call out that he makes in the earlier part of this chapter, it seems like he's making this claim both that containment is necessary and impossible.
00:15:51.264 --> 00:16:05.327
Uh, like, he has this section where it's called, Containment is the Foundation." And in that section, he's basically like,"Well, you know, you can't have anything without containment." And then he goes on to say where it's like it's impossible to do.
00:16:05.368 --> 00:16:18.629
So it just L like the only conclusion i can possibly draw from that is that he's basically saying like,"Well, we're all doomed." And I'm not sure if there's another takeaway from that, but
00:16:23.403 --> 00:16:30.754
It's funny, I had a similar note, uh, or a similar bullet point in my notes where I do feel like a lot of times his writing just feels alarmist with no real, solution, I guess.
00:16:31.434 --> 00:16:33.705
Uh, it, you know, it, I feel like he's...
00:16:34.434 --> 00:16:44.846
The main thrust is that we need to respect that technology can't be contained, and we need to move forward with, with that knowledge.
00:16:45.186 --> 00:16:58.971
You know,'cause he's mentioned, like, these societies that try to be Luddite societies and how, like, sure, that can exist in microcosm, but it j- it, it's not something that can be, like, sustained at any large scale.
00:16:59.515 --> 00:17:08.136
and so it seems as though he's just sort of warning us that, like, okay, you know, you might think politicians are gonna put laws on these things, but it's just not gonna happen.
00:17:08.136 --> 00:17:14.659
It's not gonna work, and so we need a different way of approaching it, and I think that's, what he's driving towards.
00:17:14.669 --> 00:17:22.239
But, but yeah, it does sometimes come across just, like, a little alarmist, like sky is falling and there's no- nothing we can do about it.
00:17:22.886 --> 00:18:07.509
Well, and maybe this is why he brings up this nuclear example where he's like,"Hey, I mean, I wouldn't look to the fact that we managed to contain nuclear weapons as any reassuring example." Maybe this is exactly to your point where it's like, hey, even this one thing where we managed to contain, like you shouldn't derive too much comfort from that example because there's important realities to AI as a technology, as you called out, that indicate or indicate that it's gonna be significantly harder to contain in the same way because there's this tremendous utility at a small, like at a micro level to an individual.
00:18:08.029 --> 00:18:08.720
Exactly.
00:18:08.730 --> 00:18:20.912
Yeah, and I think that's really gonna be his, one of his primary theses is these two technologies, artificial intelligence and synthetic biology, have tremendous human value.
00:18:21.362 --> 00:18:22.172
Tremendous.
00:18:22.192 --> 00:18:27.571
And so if the United States, you know, passes some laws to block it, whatever, Singapore will do it.
00:18:27.612 --> 00:18:29.382
Some other, some other nation will do it.
00:18:29.731 --> 00:18:36.832
They'll, they'll h- you know, get all the best doctors, who by the way want to work on these things because that's what they do.
00:18:36.842 --> 00:18:44.699
You know, th- they've spent their lives working towards, you know, creating scientific breakthroughs and, and so they're going to be interested in these things.
00:18:44.864 --> 00:18:51.413
Like I work with a bunch of engineers who are interested in working on agentic harnesses, and that's just, that's just what we're doing.
00:18:51.616 --> 00:18:54.136
And sure, we're like obviating our own selves.
00:18:55.147 --> 00:19:00.727
You know, that's the, that's the like only logical conclusion to what we're doing.
00:19:00.974 --> 00:19:05.950
But we're fascinated by the problem, and it's just not gonna stop happening.
00:19:05.950 --> 00:19:16.848
And so, yeah, I, I think we need to move forward with the knowledge that market forces are going to cause this to happen regardless of what we think or feel or believe
00:19:17.526 --> 00:19:17.986
Yes.
00:19:18.026 --> 00:19:18.486
Yeah.
00:19:18.526 --> 00:19:19.586
And, and he...
00:19:20.055 --> 00:19:22.384
I think that's definitely true.
00:19:22.384 --> 00:19:34.906
I mean, I guess the, the question is, and he has a quote earlier in this chapter where he's like Are the creators absolved of their responsibility because it was always inevitable?
00:19:34.916 --> 00:19:46.876
I mean, his, his conclusion is no, and this is another place where it feels like there's these contradictions because he, he freely admits, like, this is something that the original inventor would never been able to stop anyway.
00:19:47.262 --> 00:19:57.932
But, uh, you know, he kind of has to come out with this, like, uh, on this pretty weak, like, middle ground of like,"Well, you should still try." And it's
00:19:57.989 --> 00:19:57.999
Yeah.
00:19:57.999 --> 00:19:58.088
It, he's almost undermining his own point, you know?
00:20:01.858 --> 00:20:06.249
It's like when w- you know, when Germany used chlorine gas in World War I.
00:20:06.269 --> 00:20:16.567
It's like the reason they did that is because they felt there was no other option, and that's sort of-- that's what, how people are going to feel'cause they know that the other guy is gonna do it.
00:20:16.795 --> 00:20:20.551
So it's sort of this, like, zero-sum game arms race.
00:20:20.605 --> 00:20:21.315
bottom, yeah
00:20:21.531 --> 00:20:28.669
Yeah, where it's like if the United States isn't doing it, China's doing it, and China will develop AGI, and then China will rule the world.
00:20:28.865 --> 00:20:34.535
W- not that that's necessarily a bad thing, but the United States isn't gonna stand idly by while that happens.
00:20:34.987 --> 00:20:35.376
Right.
00:20:35.376 --> 00:20:57.141
So your point is kind of like He says, there's-- He said, and, and I'm quoting here from the book,"Decisions technologists and societies make at the source can still shape outcomes." But it almost sounds like there's more-- There's actually less agency that these technologists and societies have than he's alluding to here.
00:20:57.542 --> 00:20:57.923
Yeah.
00:20:57.942 --> 00:21:11.531
To me, there is sort of this inevitability that Yeah, nation, you know, by utilizing the market forces at their disposal, nations are going to support these technologies as much as possible.
00:21:11.719 --> 00:21:31.731
And, and you can see this with, you know, Trump investing a ton of money in Intel or Trump and Biden both creating very favorable circumstances for chip development in the United States and, and new data centers to be created and, you know, sort of like cutting the legal red tape in order for those things to happen.
00:21:32.296 --> 00:21:47.259
Like, the United States wants random Microsoft, Google, and OpenAI engineers to develop these technologies, and so they're creating very favorable circumstances for those companies to hire the best minds and put them to work.
00:21:47.740 --> 00:21:48.161
I don't know.
00:21:48.161 --> 00:21:57.250
I, I think, you know, Mustafa is sort of telling us to tread lightly and, and treat this tech with humility and responsibility, which I hope people are doing.
00:21:57.330 --> 00:21:59.631
I mean, that's sort of what I have in the back of my mind.
00:21:59.836 --> 00:22:06.324
I mean, not that I'm important to this whole thing, but I think hopefully most engineers who work on this stuff feel that way.
00:22:06.696 --> 00:22:07.967
But it's going to happen.
00:22:08.267 --> 00:22:09.166
It's going to happen
00:22:09.769 --> 00:22:13.127
Okay, well, I don't, think I really had any other, points.
00:22:13.127 --> 00:22:15.637
Did you have any other things you wanted to call out in this chapter?
00:22:16.930 --> 00:22:17.779
I don't think so.
00:22:18.339 --> 00:22:22.809
Yeah, this is an important chapter because, you know, this is the chapter where we sort of...
00:22:23.299 --> 00:22:24.309
stuff gets real.
00:22:24.490 --> 00:22:26.180
It's like it's coming at you
00:22:27.809 --> 00:22:27.890
Yeah.
00:22:28.039 --> 00:22:40.426
Well, actually there was a, there was one point, I'm glad you kind of brought it back to today because we're actually starting to see the seeds of this, resistance, you know, saying no to technology.
00:22:40.426 --> 00:22:43.683
It feels like There is this growing movement of, resistance to AI.
00:22:44.453 --> 00:22:45.673
Like you're, you're saying like they're...
00:22:46.144 --> 00:22:46.653
What's that?
00:22:46.753 --> 00:22:47.223
The Pope?
00:22:47.903 --> 00:22:48.294
Yeah.
00:22:48.703 --> 00:22:50.814
Pope Leo, yeah, he threw out that treatise
00:22:51.430 --> 00:22:55.344
Yeah, I am of two minds about it because I'm kind of generally on the side of technology.
00:22:55.344 --> 00:23:00.653
Like, yes, there's negative effects of technology in the near term and positive in the long term.
00:23:00.913 --> 00:23:09.548
but I think there's very real concerns that everyone has, you know, college grad- graduates booing the, you know, booing these commencement speakers who are, like, saying deal with it,
00:23:09.846 --> 00:23:10.586
Eric Schmidt.
00:23:10.615 --> 00:23:11.536
Yeah, that was hilarious
00:23:12.577 --> 00:23:25.365
And I feel so, like, I f- I feel so, uh, conflicted, I guess, about how to feel about those because at the end of the day, I kind of agree with the speaker'cause they're like...
00:23:25.566 --> 00:23:26.355
They say deal with it.
00:23:26.365 --> 00:23:32.561
Maybe they could've done it more, tactfully, but it's gonna happen, you know?
00:23:32.741 --> 00:23:38.162
Uh, and like you were saying, and so I don't know.
00:23:38.162 --> 00:23:40.922
It, it's hard to-- I think it's a hard situation to square.
00:23:40.942 --> 00:23:45.221
I think, I mean, frankly, honestly, I think Bernie kind of has the right idea.
00:23:45.221 --> 00:24:10.249
I, I don't think it's ever gonna happen, but he was basically like,"All right, we're gonna have this AI sovereign wealth fund." And I think the best thing to do is be like,"Okay, well, we're not just gonna allow this incredibly small set of individuals to profit while a bunch of people are put out of work." Like, at the very least, we should be using that money to go back to the people to help them learn a career in the post-AI world looks like, so.
00:24:10.806 --> 00:24:11.125
Yeah.
00:24:11.576 --> 00:24:14.125
No, and I, I do think that's actually a really good idea.
00:24:14.720 --> 00:24:22.096
and I for one, e- even being an engineer who's sort of working in this space, I'm super happy there's all this pushback.
00:24:22.306 --> 00:24:27.598
And I read what the Pope said about AI and how, you know, technology is sort of...
00:24:27.803 --> 00:24:38.883
I felt like the Pope did a really good job of saying that, you know, technology is a neutral thing, but it can be used for evil purposes or for good purposes.
00:24:39.833 --> 00:24:50.113
And I feel like, you know, this type of, like, opposition-- Uh, opposition is not the word I'm looking for, but I'm, uh, I can't think of the word I'm looking for.
00:24:50.113 --> 00:25:15.218
So this type of opposition against a technology I think is helpful because it just makes people more-- think about it more and sort of, when they are making decisions in the day-to-day of like, Oh, should the AI be really forceful about how it responds or should it be more, you know, graceful and sort of nice?" And, uh, like these day-to-day decisions are so important.
00:25:15.298 --> 00:25:25.532
And so if someone is constantly barraged by like everyone hating AI, they're gonna like treat their day-to-day with more, more thought and respect, and I think that's a good thing
00:25:26.820 --> 00:25:27.101
Yeah.
00:25:28.191 --> 00:25:29.530
I, I agree.
00:25:29.570 --> 00:25:30.171
I agree.
00:25:30.580 --> 00:25:34.980
Um, all right, well, for real this time, uh, I do not have anything else.
00:25:35.228 --> 00:25:38.407
and the next chapter is called"The Technology of Intelligence."
00:25:39.301 --> 00:25:39.741
Yes.
00:25:39.892 --> 00:25:40.811
I look forward to it
00:25:41.375 --> 00:25:41.704
All right.
00:25:41.704 --> 00:25:42.394
See you next time, John
00:25:43.031 --> 00:25:43.771
See you next time, Matt
00:00:08.329 --> 00:00:15.048
Hello everyone And welcome to the switch statement podcast It's a podcast for investigations into miscellaneous tech topics
00:00:16.884 --> 00:00:20.484
This is episode three of"The Coming Wave" by Mustafa Suleiman
00:00:31.434 --> 00:00:32.473
Hey, John.
00:00:32.503 --> 00:00:32.823
How you doing?
00:00:33.808 --> 00:00:34.988
Hello, Matt.
00:00:35.279 --> 00:00:36.548
I am doing very well.
00:00:36.558 --> 00:00:37.168
How are you doing?
00:00:38.344 --> 00:00:39.264
I am doing all right.
00:00:39.274 --> 00:00:47.191
Actually, I'm on cloud nine right now because, you know, we got an amazing suggestion.
00:00:47.191 --> 00:00:52.680
You know, I was going through the mailbags, uh, of our fan mail
00:00:52.963 --> 00:00:54.162
Yeah, I saw this
00:00:54.856 --> 00:00:57.695
and Alex, I hope you're, you're okay.
00:00:57.695 --> 00:01:00.045
That's all we're gonna say without saying your name.
00:01:00.045 --> 00:01:09.525
But, um, he had a fantastic idea, which was that we should do a better job of incorporating the listeners in our book selection process.
00:01:10.135 --> 00:01:18.394
and so, you know, he said maybe we should do a poll or at least give a heads up a couple, you know, in anticipation of the next book.
00:01:18.745 --> 00:01:20.555
And I think those are both fantastic ideas.
00:01:20.555 --> 00:01:40.754
And what we're planning on doing, we don't have a, we don't have a domain or anything like that, but we think we're gonna build a little, a little web app where people can go and, you know, we're both coders, like we're just gonna vibe code some shitty web app that allows people to go and vote up our candidate books
00:01:41.777 --> 00:01:42.277
I love this idea.
00:01:42.346 --> 00:01:47.777
I love this idea because I feel like we often kind of struggle to pick a book.
00:01:47.826 --> 00:01:58.887
We're like,"Eh." And, and this'll, I don't know, this'll just be more fun for us, I feel,'cause we'll, we might have to do books that we've never heard of, that are interesting to us.
00:01:59.537 --> 00:02:00.307
So it'll be cool
00:02:01.055 --> 00:02:08.955
That is the part that I am most excited about, is getting exposed to books that we never would've found otherwise.
00:02:09.425 --> 00:02:23.634
Um, th- the other thing that's funny, whenever we're picking a book, it's pretty clear that both of us wanna spend a- the absolute minimum possible time we can picking the book, which is ironic given the fact that it will likely
00:02:24.188 --> 00:02:25.979
We spent so much time.
00:02:26.004 --> 00:02:26.604
lives.
00:02:27.109 --> 00:02:30.058
Yeah, we just arbitrarily pick Godel, Escher,
00:02:30.224 --> 00:02:33.324
we can spend 45 seconds picking this book?
00:02:33.344 --> 00:02:39.504
Like,'cause,'cause I got shit to do," even though it's like, uh, gonna have major downstream ramifications.
00:02:40.305 --> 00:02:43.175
Uh, not the least of which is on the listener.
00:02:43.824 --> 00:02:46.004
So I, you know, I've always been a fan of democracy.
00:02:46.525 --> 00:02:47.365
They're now...
00:02:47.384 --> 00:02:53.685
They are just, uh, downstream recipients of our oligarchy of two.
00:02:53.935 --> 00:02:56.865
Our, our du- duarchy?
00:02:56.995 --> 00:02:57.664
Is that a thing?
00:02:57.911 --> 00:03:06.661
it's like an inverse duarchy because they're basically forcing us to do their bidding, which is how a, you know, how a democracy should work
00:03:07.550 --> 00:03:08.100
Exactly.
00:03:08.100 --> 00:03:14.800
Well, I'm just saying like the, in the system today, it's a, it's a duopoly or du- duarchy
00:03:15.275 --> 00:03:15.604
Yeah
00:03:15.691 --> 00:03:19.181
a monarchy and, and we're gonna open it up to the masses.
00:03:19.441 --> 00:03:21.901
And now we may have, we may have some,
00:03:22.411 --> 00:03:23.170
Do Umver
00:03:23.281 --> 00:03:23.730
know, we'll...
00:03:24.251 --> 00:03:24.700
What's that?
00:03:24.711 --> 00:03:24.721
Duumvirate.
00:03:26.091 --> 00:03:26.631
That's good.
00:03:26.640 --> 00:03:26.871
That's
00:03:26.932 --> 00:03:28.502
I'm just thinking of the triumvirate.
00:03:28.522 --> 00:03:31.073
I don't know if there's a similar word for, for two
00:03:31.838 --> 00:03:39.058
I think we don't have a word for it'cause it's just a s- just a stupid idea to just have two leaders'cause y- like what happens when they disagree?
00:03:39.465 --> 00:03:43.324
It's so strange though,'cause hasn't there been a few triumvirates?
00:03:43.335 --> 00:03:44.485
Like, there's Augustus...
00:03:44.564 --> 00:03:49.104
Or though m- maybe you're pointing out the, the reason why there's never a duumvirate.
00:03:49.901 --> 00:03:51.121
That's exactly what I'm saying.
00:03:51.121 --> 00:03:52.950
Yeah, sorry, that-- if that wasn't clear.
00:03:53.040 --> 00:03:57.871
Like, a triumvirate makes sense because at least then you have two people, like you can break ties.
00:03:58.354 --> 00:03:58.754
Yeah
00:03:59.800 --> 00:04:16.141
uh, so, um, but yeah, so what I was gonna say is I think we might have some mechanism to override the will of the people in cases where, you know, maybe, um,"My Little Pony: Friendship is Magic," the, you know, the book...
00:04:16.420 --> 00:04:22.920
Not to-- Listen, not to throw any shade on"My Little Pony: Friendship is Magic," but, um, and I'm not even sure if there are
00:04:23.009 --> 00:04:27.069
Like how you know an exact title of a My Little Pony book?
00:04:28.069 --> 00:04:33.238
Well, that was the name of the, that was the name of the series that really blew up with like Derpy Hooves and all that stuff.
00:04:33.249 --> 00:04:34.389
But, um,
00:04:35.036 --> 00:04:37.146
Wow, I have no knowledge of this.
00:04:37.867 --> 00:04:50.216
Now, I feel like our fans are gonna be like,"Can you read the every digit of pi book?" And then we'll be analyzing, like, digits 10 million through 10,000,500, and
00:04:50.269 --> 00:04:53.809
Yeah, it's like, it's like, you know, this, this section was pretty interesting.
00:04:53.809 --> 00:05:02.428
There were only, uh, you know, point, you know, 9.8887% ones in this section, and I thought that was, I thought that was pretty interesting.
00:05:03.201 --> 00:05:03.661
Yeah.
00:05:04.678 --> 00:05:05.408
uh, okay.
00:05:05.408 --> 00:05:07.389
But anyway, so that's, that's the end of that section.
00:05:07.389 --> 00:05:12.769
This is the kind of end of the house- housekeeping section, which we have heretofore never had to do.
00:05:12.769 --> 00:05:13.538
It's pretty exciting.
00:05:13.738 --> 00:05:21.468
'Cause,'cause as much as I was joking, this is actually, I think, literally the first time we've ever had had anyone seriously write in, uh,
00:05:21.555 --> 00:05:26.415
Yeah, we, we have I was wondering whether we were gonna admit that or not.
00:05:26.415 --> 00:05:33.944
We have received other emails, but they've been kind of like, I don't know, scammy, like whatever, and it just kinda makes us depressed.
00:05:33.944 --> 00:05:36.634
So, so thank you, Alex, for brightening our, our days
00:05:37.478 --> 00:05:37.889
Yes.
00:05:37.918 --> 00:05:38.959
No, we really appreciate it.
00:05:39.259 --> 00:05:48.288
And you sound like our, our ideal listener, which is funny'cause we've never actually even thought about, uh, what, what an ideal listener would look like.
00:05:48.309 --> 00:05:54.829
But, you know, someone who's trying to work with the material and go through it a- and reason about it in, in a deep way.
00:05:55.389 --> 00:06:02.608
Um, you know, that's kind of always how I imagined our, our listeners or the class of listener that would benefit from our, our podcast, so.
00:06:03.158 --> 00:06:03.858
Um, all right.
00:06:03.879 --> 00:06:06.848
But talking about containment
00:06:08.194 --> 00:06:14.394
The containment problem certain things cannot be contained, and technology is one of them
00:06:15.605 --> 00:06:24.105
Well, yeah, I mean, and it sounds like almost every technology cannot be contained, or almost no technologies can be contained.
00:06:24.802 --> 00:06:25.343
Yes.
00:06:25.413 --> 00:06:29.673
Yeah, so that's, that's basically the theme of this chapter is...
00:06:29.812 --> 00:06:45.084
The last chapter sort of touched on this, where technology proliferates i- in a sort of ridiculous way, and he gave the example of the engine, and now we have engines in literally everything from lawnmowers to kids' toys to things that you wouldn't even imagine to have an engine.
00:06:45.675 --> 00:07:00.992
And he sort of discussed how the original inventors of the engine, you know, they were thinking about cars, sure, which is a primary use case for an engine, but now there's so many things with engines they never would've conceived of, of any of these things.
00:07:01.290 --> 00:07:02.478
and this is sort of what happens.
00:07:02.497 --> 00:07:11.408
It proliferates, it reaches into all these nooks and crannies of our day-to-day life, and it becomes impossible to contain.
00:07:12.257 --> 00:07:17.817
And he gives these examples of, like the ancient Ottomans, for example, with the printing press.
00:07:18.158 --> 00:07:19.687
Like you have, Gutenberg.
00:07:19.728 --> 00:07:32.850
He came up with this printing press because he wanted a simpler way of churning out Bibles, which is, I don't know, just a testament to how, you know, people don't really foresee the full consequences of what they're doing.
00:07:33.168 --> 00:07:36.307
And the Ottomans were trying to control religion.
00:07:36.377 --> 00:07:51.334
They didn't want, religious propaganda to sort of come into their territory and, you know, they had their own views on religion, and they wanted to sort of dictate to their populace what the primary religion would be and, and whatnot.
00:07:51.843 --> 00:08:00.958
and so they tried to prevent people from having printing presses and tried to prevent, information being produced from printing presses from entering their territory.
00:08:01.389 --> 00:08:02.524
but this utterly failed.
00:08:02.735 --> 00:08:11.074
You can imagine how hard it is to prevent something like paper with ink on it, you know, from permeating into your territory.
00:08:11.714 --> 00:08:13.995
Uh, and so it's kind of eventually backfired.
00:08:14.651 --> 00:08:24.161
Yeah, and so just to, just to touch on a point you just talked about where it's like there's all these unintended side effects as technology proliferates.
00:08:24.701 --> 00:08:44.313
He even goes a little farther, which is to say that generally speaking, it's kind of the rule that an inventor invents a technology, and then it's actually used to do something that di- directly conflicts with the inventor's intention for the technology.
00:08:44.846 --> 00:08:45.427
Dynamite
00:08:45.464 --> 00:08:56.494
a crucial a- element of what he's saying,'cause yes, there's all these unintended side effects, but he actually refers to it in this more stark language, revenge effects, which I think maybe goes a little bit too far.
00:08:56.797 --> 00:09:15.017
but basically that even in the case of these maybe like less sentient technologies, the tech- the, the propagation of the technology serves to kind of conflict with the original inventor's vision for the technology.
00:09:15.405 --> 00:09:15.666
Yeah.
00:09:16.096 --> 00:09:29.778
Yeah, and that I, I feel like the famous example of that is Nobel with dynamite, where he developed that for mining, for, you know, making mining simpler, being able to chisel into mountains m- more easily.
00:09:30.168 --> 00:09:38.447
And obviously, it ended up being used for warfare, and this is part of the reason he developed the Nobel Prize fund is because he felt guilt
00:09:39.261 --> 00:09:39.682
Yes.
00:09:39.741 --> 00:09:40.751
Yes, exactly.
00:09:41.162 --> 00:09:41.682
Um,
00:09:42.346 --> 00:09:42.735
Yeah
00:09:42.961 --> 00:09:53.532
um, so as you called out, right, we have all these c- mostly state actors, it seems like, who are the ones who really wanna control the technology.
00:09:53.542 --> 00:09:59.591
It's interesting'cause he talks about, like, the technology goes on to conflict with the inventor's original intent,
00:10:00.148 --> 00:10:00.158
Mm-hmm.
00:10:00.572 --> 00:10:02.261
but they aren't the primary ones.
00:10:02.261 --> 00:10:09.471
And maybe it, maybe they do attempt to control it, and the, it's just such a meager attempt that it doesn't bear even mentioning.
00:10:09.481 --> 00:10:16.971
Whereas at least when you have a government that's trying to control a technology, they do a, at least a good enough job that it's worth talking about.
00:10:17.001 --> 00:10:17.261
But,
00:10:17.658 --> 00:10:18.548
I wanna give a, uh...
00:10:18.749 --> 00:10:20.899
a quick pop culture reference.
00:10:21.895 --> 00:10:26.464
He mentions the Tokugawa shogunate, which is another example of a...
00:10:26.964 --> 00:10:38.082
Just like you're saying, you know, the Tokugawa shogu- shogunate sort of had control over a large swath of Japan, and they attempted to turn themselves into an extremely insular society.
00:10:38.113 --> 00:10:42.783
Like, they sort of had these brief dealings with the outside world, and they were like,"We don't like this.
00:10:42.812 --> 00:10:52.477
We're going to, you know, close the gates, make our society completely, self-contained and self-sufficient." And the, and Mustafa sort of talks about how this backfired.
00:10:52.847 --> 00:10:55.998
But what I wanted to bring up was, did you watch the s- show"Shogun"?
00:10:55.998 --> 00:10:58.187
The, uh-
00:10:58.282 --> 00:10:59.211
not, no
00:10:59.788 --> 00:11:13.160
Okay,'cause there is a scene in"Shogun" that I actually found this scene incredible, where the main character, who is, like, this, I think he was, like, a Portuguese cap- like, ship pilot or something.
00:11:13.831 --> 00:11:20.150
He was explaining to this shogun how, like, Spain owned Japan.
00:11:20.509 --> 00:11:33.219
Uh, and he was, like, drawing this map in the sand, and he was like,"Oh, don't you know, like, this random country that you've never heard of owns Japan?" Which it's like, you th- you think you're the shogun.
00:11:33.979 --> 00:11:34.568
And I don't know.
00:11:34.578 --> 00:11:35.798
It was just this kind of, like...
00:11:35.879 --> 00:11:40.094
And, uh, you know, obviously it, it sort of made no sense to the shogun.
00:11:40.094 --> 00:11:41.668
The shogun's like,"What the hell are you talking about?
00:11:41.668 --> 00:11:52.687
We've never heard of Spain." But it was also this interesting example of how, like, yeah, sure, you might try to contain yourself, but the outside world continues to exist, and people, you know, sort of continue to make these global decisions.
00:11:52.865 --> 00:11:58.335
And if you're not at the table, like, fine, but you're not gonna have a word in these decisions.
00:11:59.274 --> 00:11:59.924
Uh, so
00:12:00.104 --> 00:12:20.438
and this is actually, I'm really glad that you talked about that, about the rest of the world advancing,'cause there was one example that he calls or that he calls out Queen, Queen Elizabeth banning the knitting machine, and it sounds like she was actually successful in banning this knitting machine, at least within England.
00:12:20.678 --> 00:12:35.408
But then, like, when it came back, there had been a bunch of these advances in technology, and it was, like, the mechanical looms, like these enormous mechanical industrial looms that came up and just, like, bulldozed over it.
00:12:35.597 --> 00:12:54.734
So it's like, even if you do succeed in suppressing this technology for a time, it's like all of these other advances are gonna go on and, like, they're just gonna come back with a vengeance and, like, just completely trounce whoever the group of people you were intending to protect, with, you know, with this, uh, protectionism.
00:13:00.991 --> 00:13:16.523
he, brings up w- the one potentially defensible case where the world has up until now managed to suppress, I guess, the propagation of a technology, which is nuclear weapons.
00:13:16.991 --> 00:13:17.471
Yes.
00:13:17.501 --> 00:13:18.052
Yeah.
00:13:18.121 --> 00:13:25.662
And we're seeing the mechanisms of suppression today with the wa- the war in Iran, which, you
00:13:25.876 --> 00:13:26.496
Right, and I...
00:13:27.384 --> 00:13:45.453
and that kind of dovetails with, and this was before, obviously before the war in Iran, but, like, is part of his central thesis about the, why this might be the one technology that we're aware of that hasn't propagated is just'cause it's really fucking hard.
00:13:45.514 --> 00:13:56.818
You know, it's like, it's just enormously difficult to do, and unless you have, a, like a wealthy state that can fund the creation of this nuclear technology, you just can't do it.
00:13:56.889 --> 00:13:58.649
It's just not, you know, it's just not possible.
00:13:59.633 --> 00:14:11.192
He, he gives, he gives another really thought-provoking example of why it's able to be contained, which is just that the normal market forces don't exist for nuclear tech.
00:14:11.793 --> 00:14:15.182
You know, nuclear tech is a civ level...
00:14:15.673 --> 00:14:18.253
You know, just like you're saying, it's a civ level effort.
00:14:18.263 --> 00:14:20.712
Like, you require thousands of people.
00:14:20.712 --> 00:14:34.293
You require, you know, whatever it is, millions of centrifuges spinning uranium to enrich it, and y- it's just you require this ridiculous amount of resources, technology, people.
00:14:34.756 --> 00:14:37.466
But no one wants a nuclear bomb.
00:14:37.797 --> 00:14:46.176
You know, it's not like, it's not like ChatGPT where you open it a million times a day and ask it about, like, a mole on your, you know, eyebrow or something.
00:14:46.186 --> 00:14:47.326
I-it's just...
00:14:47.346 --> 00:14:56.538
And no one, no one wants this thing, and, and what that means is there's no real incentive for people to, you know, proliferate it.
00:14:56.548 --> 00:15:04.288
Like, obviously there's an incentive for nations to have it because it's this deterrent and, you know, you can sort of saber-rattle more effectively.
00:15:04.644 --> 00:15:20.138
But just on the, on the scale of, like, human society, there's not this, like, market, you know, market reason to sort of, like, proliferate it and improve it and make it more economical like there is for something like AI.
00:15:20.740 --> 00:15:21.091
Right.
00:15:21.240 --> 00:15:21.971
Exactly.
00:15:22.057 --> 00:15:25.886
and yeah, so, so that i- it does seem to be one of the few...
00:15:25.897 --> 00:15:34.919
A- and luckily, r- right, we've only used it twice in, in a real war, um, which, you the United States has that dubious honor.
00:15:35.644 --> 00:15:50.745
One, one point that I wanted to call out that he makes in the earlier part of this chapter, it seems like he's making this claim both that containment is necessary and impossible.
00:15:51.264 --> 00:16:05.327
Uh, like, he has this section where it's called, Containment is the Foundation." And in that section, he's basically like,"Well, you know, you can't have anything without containment." And then he goes on to say where it's like it's impossible to do.
00:16:05.368 --> 00:16:18.629
So it just L like the only conclusion i can possibly draw from that is that he's basically saying like,"Well, we're all doomed." And I'm not sure if there's another takeaway from that, but
00:16:23.403 --> 00:16:30.754
It's funny, I had a similar note, uh, or a similar bullet point in my notes where I do feel like a lot of times his writing just feels alarmist with no real, solution, I guess.
00:16:31.434 --> 00:16:33.705
Uh, it, you know, it, I feel like he's...
00:16:34.434 --> 00:16:44.846
The main thrust is that we need to respect that technology can't be contained, and we need to move forward with, with that knowledge.
00:16:45.186 --> 00:16:58.971
You know,'cause he's mentioned, like, these societies that try to be Luddite societies and how, like, sure, that can exist in microcosm, but it j- it, it's not something that can be, like, sustained at any large scale.
00:16:59.515 --> 00:17:08.136
and so it seems as though he's just sort of warning us that, like, okay, you know, you might think politicians are gonna put laws on these things, but it's just not gonna happen.
00:17:08.136 --> 00:17:14.659
It's not gonna work, and so we need a different way of approaching it, and I think that's, what he's driving towards.
00:17:14.669 --> 00:17:22.239
But, but yeah, it does sometimes come across just, like, a little alarmist, like sky is falling and there's no- nothing we can do about it.
00:17:22.886 --> 00:18:07.509
Well, and maybe this is why he brings up this nuclear example where he's like,"Hey, I mean, I wouldn't look to the fact that we managed to contain nuclear weapons as any reassuring example." Maybe this is exactly to your point where it's like, hey, even this one thing where we managed to contain, like you shouldn't derive too much comfort from that example because there's important realities to AI as a technology, as you called out, that indicate or indicate that it's gonna be significantly harder to contain in the same way because there's this tremendous utility at a small, like at a micro level to an individual.
00:18:08.029 --> 00:18:08.720
Exactly.
00:18:08.730 --> 00:18:20.912
Yeah, and I think that's really gonna be his, one of his primary theses is these two technologies, artificial intelligence and synthetic biology, have tremendous human value.
00:18:21.362 --> 00:18:22.172
Tremendous.
00:18:22.192 --> 00:18:27.571
And so if the United States, you know, passes some laws to block it, whatever, Singapore will do it.
00:18:27.612 --> 00:18:29.382
Some other, some other nation will do it.
00:18:29.731 --> 00:18:36.832
They'll, they'll h- you know, get all the best doctors, who by the way want to work on these things because that's what they do.
00:18:36.842 --> 00:18:44.699
You know, th- they've spent their lives working towards, you know, creating scientific breakthroughs and, and so they're going to be interested in these things.
00:18:44.864 --> 00:18:51.413
Like I work with a bunch of engineers who are interested in working on agentic harnesses, and that's just, that's just what we're doing.
00:18:51.616 --> 00:18:54.136
And sure, we're like obviating our own selves.
00:18:55.147 --> 00:19:00.727
You know, that's the, that's the like only logical conclusion to what we're doing.
00:19:00.974 --> 00:19:05.950
But we're fascinated by the problem, and it's just not gonna stop happening.
00:19:05.950 --> 00:19:16.848
And so, yeah, I, I think we need to move forward with the knowledge that market forces are going to cause this to happen regardless of what we think or feel or believe
00:19:17.526 --> 00:19:17.986
Yes.
00:19:18.026 --> 00:19:18.486
Yeah.
00:19:18.526 --> 00:19:19.586
And, and he...
00:19:20.055 --> 00:19:22.384
I think that's definitely true.
00:19:22.384 --> 00:19:34.906
I mean, I guess the, the question is, and he has a quote earlier in this chapter where he's like Are the creators absolved of their responsibility because it was always inevitable?
00:19:34.916 --> 00:19:46.876
I mean, his, his conclusion is no, and this is another place where it feels like there's these contradictions because he, he freely admits, like, this is something that the original inventor would never been able to stop anyway.
00:19:47.262 --> 00:19:57.932
But, uh, you know, he kind of has to come out with this, like, uh, on this pretty weak, like, middle ground of like,"Well, you should still try." And it's
00:19:57.989 --> 00:19:57.999
Yeah.
00:19:57.999 --> 00:19:58.088
It, he's almost undermining his own point, you know?
00:20:01.858 --> 00:20:06.249
It's like when w- you know, when Germany used chlorine gas in World War I.
00:20:06.269 --> 00:20:16.567
It's like the reason they did that is because they felt there was no other option, and that's sort of-- that's what, how people are going to feel'cause they know that the other guy is gonna do it.
00:20:16.795 --> 00:20:20.551
So it's sort of this, like, zero-sum game arms race.
00:20:20.605 --> 00:20:21.315
bottom, yeah
00:20:21.531 --> 00:20:28.669
Yeah, where it's like if the United States isn't doing it, China's doing it, and China will develop AGI, and then China will rule the world.
00:20:28.865 --> 00:20:34.535
W- not that that's necessarily a bad thing, but the United States isn't gonna stand idly by while that happens.
00:20:34.987 --> 00:20:35.376
Right.
00:20:35.376 --> 00:20:57.141
So your point is kind of like He says, there's-- He said, and, and I'm quoting here from the book,"Decisions technologists and societies make at the source can still shape outcomes." But it almost sounds like there's more-- There's actually less agency that these technologists and societies have than he's alluding to here.
00:20:57.542 --> 00:20:57.923
Yeah.
00:20:57.942 --> 00:21:11.531
To me, there is sort of this inevitability that Yeah, nation, you know, by utilizing the market forces at their disposal, nations are going to support these technologies as much as possible.
00:21:11.719 --> 00:21:31.731
And, and you can see this with, you know, Trump investing a ton of money in Intel or Trump and Biden both creating very favorable circumstances for chip development in the United States and, and new data centers to be created and, you know, sort of like cutting the legal red tape in order for those things to happen.
00:21:32.296 --> 00:21:47.259
Like, the United States wants random Microsoft, Google, and OpenAI engineers to develop these technologies, and so they're creating very favorable circumstances for those companies to hire the best minds and put them to work.
00:21:47.740 --> 00:21:48.161
I don't know.
00:21:48.161 --> 00:21:57.250
I, I think, you know, Mustafa is sort of telling us to tread lightly and, and treat this tech with humility and responsibility, which I hope people are doing.
00:21:57.330 --> 00:21:59.631
I mean, that's sort of what I have in the back of my mind.
00:21:59.836 --> 00:22:06.324
I mean, not that I'm important to this whole thing, but I think hopefully most engineers who work on this stuff feel that way.
00:22:06.696 --> 00:22:07.967
But it's going to happen.
00:22:08.267 --> 00:22:09.166
It's going to happen
00:22:09.769 --> 00:22:13.127
Okay, well, I don't, think I really had any other, points.
00:22:13.127 --> 00:22:15.637
Did you have any other things you wanted to call out in this chapter?
00:22:16.930 --> 00:22:17.779
I don't think so.
00:22:18.339 --> 00:22:22.809
Yeah, this is an important chapter because, you know, this is the chapter where we sort of...
00:22:23.299 --> 00:22:24.309
stuff gets real.
00:22:24.490 --> 00:22:26.180
It's like it's coming at you
00:22:27.809 --> 00:22:27.890
Yeah.
00:22:28.039 --> 00:22:40.426
Well, actually there was a, there was one point, I'm glad you kind of brought it back to today because we're actually starting to see the seeds of this, resistance, you know, saying no to technology.
00:22:40.426 --> 00:22:43.683
It feels like There is this growing movement of, resistance to AI.
00:22:44.453 --> 00:22:45.673
Like you're, you're saying like they're...
00:22:46.144 --> 00:22:46.653
What's that?
00:22:46.753 --> 00:22:47.223
The Pope?
00:22:47.903 --> 00:22:48.294
Yeah.
00:22:48.703 --> 00:22:50.814
Pope Leo, yeah, he threw out that treatise
00:22:51.430 --> 00:22:55.344
Yeah, I am of two minds about it because I'm kind of generally on the side of technology.
00:22:55.344 --> 00:23:00.653
Like, yes, there's negative effects of technology in the near term and positive in the long term.
00:23:00.913 --> 00:23:09.548
but I think there's very real concerns that everyone has, you know, college grad- graduates booing the, you know, booing these commencement speakers who are, like, saying deal with it,
00:23:09.846 --> 00:23:10.586
Eric Schmidt.
00:23:10.615 --> 00:23:11.536
Yeah, that was hilarious
00:23:12.577 --> 00:23:25.365
And I feel so, like, I f- I feel so, uh, conflicted, I guess, about how to feel about those because at the end of the day, I kind of agree with the speaker'cause they're like...
00:23:25.566 --> 00:23:26.355
They say deal with it.
00:23:26.365 --> 00:23:32.561
Maybe they could've done it more, tactfully, but it's gonna happen, you know?
00:23:32.741 --> 00:23:38.162
Uh, and like you were saying, and so I don't know.
00:23:38.162 --> 00:23:40.922
It, it's hard to-- I think it's a hard situation to square.
00:23:40.942 --> 00:23:45.221
I think, I mean, frankly, honestly, I think Bernie kind of has the right idea.
00:23:45.221 --> 00:24:10.249
I, I don't think it's ever gonna happen, but he was basically like,"All right, we're gonna have this AI sovereign wealth fund." And I think the best thing to do is be like,"Okay, well, we're not just gonna allow this incredibly small set of individuals to profit while a bunch of people are put out of work." Like, at the very least, we should be using that money to go back to the people to help them learn a career in the post-AI world looks like, so.
00:24:10.806 --> 00:24:11.125
Yeah.
00:24:11.576 --> 00:24:14.125
No, and I, I do think that's actually a really good idea.
00:24:14.720 --> 00:24:22.096
and I for one, e- even being an engineer who's sort of working in this space, I'm super happy there's all this pushback.
00:24:22.306 --> 00:24:27.598
And I read what the Pope said about AI and how, you know, technology is sort of...
00:24:27.803 --> 00:24:38.883
I felt like the Pope did a really good job of saying that, you know, technology is a neutral thing, but it can be used for evil purposes or for good purposes.
00:24:39.833 --> 00:24:50.113
And I feel like, you know, this type of, like, opposition-- Uh, opposition is not the word I'm looking for, but I'm, uh, I can't think of the word I'm looking for.
00:24:50.113 --> 00:25:15.218
So this type of opposition against a technology I think is helpful because it just makes people more-- think about it more and sort of, when they are making decisions in the day-to-day of like, Oh, should the AI be really forceful about how it responds or should it be more, you know, graceful and sort of nice?" And, uh, like these day-to-day decisions are so important.
00:25:15.298 --> 00:25:25.532
And so if someone is constantly barraged by like everyone hating AI, they're gonna like treat their day-to-day with more, more thought and respect, and I think that's a good thing
00:25:26.820 --> 00:25:27.101
Yeah.
00:25:28.191 --> 00:25:29.530
I, I agree.
00:25:29.570 --> 00:25:30.171
I agree.
00:25:30.580 --> 00:25:34.980
Um, all right, well, for real this time, uh, I do not have anything else.
00:25:35.228 --> 00:25:38.407
and the next chapter is called"The Technology of Intelligence."
00:25:39.301 --> 00:25:39.741
Yes.
00:25:39.892 --> 00:25:40.811
I look forward to it
00:25:41.375 --> 00:25:41.704
All right.
00:25:41.704 --> 00:25:42.394
See you next time, John
00:25:43.031 --> 00:25:43.771
See you next time, Matt
