Annalee Newitz<\/a>, journalist and author of Automatic Noodle and other books<\/p>\n\nEpisode Transcript\n<\/p>\n
This is a computer-generated transcript. While our team has reviewed it, there may be errors.<\/p>\n
Annalee Newitz (00:00):
What if it turns out language is just kind of nothing big? What if it\u2019s just like, oh, the reason why LLMs can do this so easily is because it turns out language is no big deal. And it\u2019s actually not what makes us cool. What makes us cool is something else. Because think about how bullshit written language really is. As somebody who I write a lot \u2013<\/p>\n
Robin Sloan (00:19):
Hold on. McMillan, we have a new strap line to print across the front of all of Annalee Newitz\u2019s novels.<\/p>\n
Alexis Madrigal (00:27):
I\u2019m Alexis Madrigal.<\/p>\n
Robin Sloan (00:30):
I\u2019m Robin Sloan.<\/p>\n
Alexis Madrigal (00:31):
And this is KQED\u2019s Dream Machines.<\/p>\n
Robin Sloan (00:34):
A podcast about AI, how it works, what it means. How it feels. Yeah, how it feels. And importantly, it\u2019s rooted here in San Francisco.<\/p>\n
Alexis Madrigal (00:45):
So of course, first episode of Dream Machines, we are talking about how AI took root here in San Francisco, Anthropic, OpenAI, and like 400 million other companies between here and Hayes Valley. The dream of San Francisco and the broader Bay Area as the place that is this incredible mechanism for bringing these audacious dreams forward, whether or not you kind of want them to, whether or not you think they\u2019re a good idea or a bad idea, just incontrovertibly going back in time, this is the place. Semiconductors all the way through the PC, all the different components of the PC are happening here. The internet, of course, social media, the great horrors of social media, as well as all the Uber for everything companies. Each one of these things has these roots in this area. And it is one of those times where you\u2019re like, was it just the gold rush?<\/p>\n
(01:39):
Did the gold rush somehow imbue some magic dust?<\/p>\n
Robin Sloan (01:42):
So for our first conversation, we wanted to find the ultimate<\/p>\n
Alexis Madrigal (01:47):
San<\/p>\n
Robin Sloan (01:48):
Franciscan.<\/p>\n
Alexis Madrigal (01:48):
The avatar<\/p>\n
Robin Sloan (01:50):
Of<\/p>\n
Alexis Madrigal (01:50):
San Francisco. Who<\/p>\n
Robin Sloan (01:51):
Can speak? Who can speak about this place in the long run? And across all these dimensions we\u2019re talking about, cultural and economic and technical and everything else. And<\/p>\n
Alexis Madrigal (02:00):
That would be Annalee Newitz, of course, a nonfiction writer who really in their nonfiction works on this kind of civilizational scale. In fiction too, books like Automatic Noodle is creating dream machines. It\u2019s creating these kind of other minds. And I do think Annalee is a figure that is just so San Francisco in this mix of being tech forward, being politics forward, being wild and free along dimensions other people in other regions of country maybe have never even imagined.<\/p>\n
Robin Sloan (02:31):
Thinking creatively and critically about the future, right? It\u2019s those two things together.<\/p>\n
Alexis Madrigal (02:35):
Without further ado, here is our interview with Annalee Newitz, the first on dream machines. One of the questions that I think we\u2019ve been asking ourselves is why San Francisco over and over for these booms? I mean, the Bay Area broadly, but also San Francisco quite specifically in this case and certainly over the last 20 years, what do you think? What do you make of it?<\/p>\n
Robin Sloan (02:57):
Because it\u2019s easy to say, you want to say, oh, it\u2019s a boom town. That\u2019s the nature of it. But you\u2019re like, okay, you\u2019ve told me nothing. Over all these years, what\u2019s your read of that tendency, that power?<\/p>\n
Annalee Newitz (03:09):
Yeah, it is true that this is a town of boom and bust, but more than that, it\u2019s a town of booms motivated by engineering breakthroughs. And we\u2019ve always had in the Bay Area, if you include things like Silicon Valley in the kind of greater Bay Area, we\u2019ve always had innovators and industrialists coming together. And one of the things I think that\u2019s so interesting about San Francisco is this weird alchemy of oftentimes very conservative proto-oligarchs of industry. Whether it\u2019s the railway tycoons of the late 19th century or people like Peter Thiel now or Sam Altman now, you have these people who are just dumping money into the economy and they\u2019re not interested in the public good really. I mean, or they have sort of a thin scrim of care for the public good. They just want to make money, but they\u2019re dumping that money into a city that\u2019s full of radicals and subversives, people who\u2019ve been kicked out of every decent city in the country.<\/p>\n
(04:15):
People who are too queer for their families, too leftist for their political parties. And they come together, they meet that money, and then something fricking weird happens. That\u2019s San Francisco. And I think that\u2019s why San Francisco has such weird politics too. You\u2019d think we\u2019d be a city that would be like Arcada up north, that we\u2019d have an all green city council. We\u2019d be run by socialists. But no, we\u2019re a city where the socialists somehow are hanging out with these industrialists.<\/p>\n
Alexis Madrigal (04:48):
Right, right, right. So Annalee, I know your origin story and I have read a lot of your wired.com articles and other things and following in your trail once I arrived here. I am curious, as you have watched this city change so much and through these different eras, but particularly this kind of AI boom, which is like trillion dollar companies suddenly implanting into San Francisco, how has that changed the city for you? And do you think that\u2019s changed the way that you write about the city or write about technology?<\/p>\n
Annalee Newitz (05:25):
A hundred percent it has. I think one of the big differences between the AI boom and say the dot-com boom, which was the turn of the century, is that there\u2019s a real nihilism to this tech boom. I felt like during the dot-com boom, yes, it was bonkers and there was way too much money, but there was a kind of exuberance and delight and silliness to it. Whereas I feel like the AI boom is associated with politics like the network state or very regressive sort of Peter Thiel style, let\u2019s return to monarchy type politics. And so the politics are very reactionary and very anti-democratic. And that has really shifted my understanding of San Francisco, which I always though of as a very anarchic, freewheeling, happy town that maybe was a little too rich, always a little too rich, but also kind of spent money freely on sex and drugs.<\/p>\n
(06:29):
And now I feel like what I want to reflect in my writing about the city is this dark turn toward a kind of self-denial and anti-hedonism and pro-authoritarianism. And so the city is a darker place, but ironically, it\u2019s made me want to write happier, cozier, more friendly stories about it to encourage people to remember, you can have some fun and you don\u2019t need a lot of money to do it and the city could be beautiful.<\/p>\n
Robin Sloan (07:03):
We\u2019ve talked about this quite a bit that among your many virtues and great strengths as a writer, you are perhaps the foremost chronicler or imaginer of other minds, including machine minds.<\/p>\n
Annalee Newitz (07:17):
It\u2019s really both.<\/p>\n
(07:18):
I definitely got the inspiration. I get the inspiration from my non-human characters from the world around me and from the machines I interact with. And I am a ruthless anthropomorphizer and I do make fun of myself for this. And I realize that it\u2019s kind of a character flaw, especially if you write about science. But I can\u2019t help but see the personhood in everything, including machines. Although not so much in ChatGPT, maybe we\u2019ll talk about that later. We need to get to that. Interestingly. But this microphone, I could easily have compassion for this microphone. How many times has it been breathed on? And what\u2019s it thinking about that? Does it have opinions about different smells that it encounters? And what\u2019s it going to do about it? So I think that\u2019s part of what happens to me is that I\u2019ll be writing a story that\u2019s fact-based and evidence-based.<\/p>\n
(08:12):
And a part of my brain is like, \u201cOh, but what if this was alive? What would it be thinking right now?\u201d Because what is human consciousness, but this thin little layer around a bunch of just survival imperative.<\/p>\n
Alexis Madrigal (08:26):
Why don\u2019t you briefly set up the world of Automatic Noodle and we\u2019ll use that to kind of think about some of these robots that you\u2019ve created.<\/p>\n
Annalee Newitz (08:33):
Yeah. It\u2019s the first book I\u2019ve ever written that was set in San Francisco. And it\u2019s about what happens in California after the state has won a war of independence against the United States. So it\u2019s set five years after the war has been won. So things are being rebuilt and resettled. And one of the things that\u2019s happened in the wake of the war is that some of the robots who were used in the fighting have been given limited civil rights in the nation of California. And that\u2019s partly because the nation of California thinks of itself as very futuristic and wants to kind of be the first nation that joins the UN having these kinds of rights for sentient AI. But also it\u2019s because it\u2019s kind of undeniable that some of these robots are people. And so my main characters are four robots who have these limited rights.<\/p>\n
(09:25):
They\u2019re free to work, but not to own property. They are free to associate with each other, but not form unions. And so they\u2019re struggling with a lot of the same things that many people in the states have struggled with and now in the nation of California. So they decide through kind of skirting the laws that they will open a noodle restaurant to survive. And the book is really about them learning to get along with their human neighbors, cope with robophobia, and kind of find a home together in the new San Francisco. I just started writing the sequel.<\/p>\n
Alexis Madrigal (10:03):
Oh.<\/p>\n
Annalee Newitz (10:03):
Yeah. So you\u2019re going to find out a lot about California politics in the sequel. It\u2019s going<\/p>\n
Robin Sloan (10:08):
To be another sort of slim.<\/p>\n
Annalee Newitz (10:09):
Yeah, it\u2019s going to be another novella. So it\u2019s called Noodle Union. So that gives you a hint of what I might be thinking.<\/p>\n
Robin Sloan (10:18):
I mean, there\u2019s so many ways we can get into this, but I would say primarily first as a writer, as someone who\u2019s thinking about minds and voices and characters, do you look at this new creature before us with curiosity, with interest? Can you imagine a language model on the pages of an Annalee novel or just how does it all strike you?<\/p>\n
Annalee Newitz (10:40):
I am super curious about it. I\u2019m very curious about humans\u2019 reactions to it as well and seeing how many other anthropomorphizers are out there. And people are feeling anxious about whether these are people that we\u2019re building, whether they\u2019re just tools that will destroy our lives as people. What do we do with them ethically? I think all of that stuff is super interesting. And I\u2019m glad we\u2019re thinking about it, even though I am not convinced that the LLMs we have now are people. We absolutely should be thinking about what the heck happens if it turns out that they are. Because as I\u2019ve said many times, the business model of most of these LLM companies is to build people and then buy and sell them. So we\u2019re trying to reinvent slavery. Well, we are not, but captains of industry are. They love slavery. Oh, those guys are so crazy about it.<\/p>\n
(11:34):
But what I would say is that when you\u2019re thinking about LLMs or you\u2019re thinking about AI, what we\u2019re calling AI more broadly, they do have bodies. We think of them as disembodied, but here we are in the middle of huge debates over data centers. Data centers are the bodies of these creatures, of these programs and algorithms and models. So it\u2019s very late capitalism to have something whose body is remote and can\u2019t be seen. And then you\u2019re interacting with it as something that appears to be out of thin air.<\/p>\n
Alexis Madrigal (12:11):
A bright, friendly, chirpy presence.<\/p>\n
Annalee Newitz (12:13):
Right. Right. And all of the secondary effects that that thing is having on people\u2019s lives and the environment is somewhere else, usually somewhere that is a have not state or a have not region. And so it\u2019s very colonial. And many people have written about this, about the kind of colonial model of AI. And so I\u2019m always interested in reminding people all of this stuff is embodied. There\u2019s no such thing as an AI that\u2019s like here in the air floating around as my first boss at the Bay Guardian said like, oh, the internet, it\u2019s in the air. No, it\u2019s actually not, dude. It\u2019s in a server form.<\/p>\n
Robin Sloan (12:51):
Right though. Except that the thing itself, the voice, whatever it is, in a sense, I would argue it doesn\u2019t care that it\u2019s in the data center. It matters a lot that it is, but it doesn\u2019t care. What it cares is that it\u2019s in language. I mean, it\u2019s terrain, it\u2019s landscape. If you think of the robots of your novel, they\u2019re hiking up hills and skating down the other side. The landscape, the literal environment of all these language models is language itself. And of course<\/p>\n
Annalee Newitz (13:22):
The \u2013 Is<\/p>\n
Robin Sloan (13:22):
It though? It absolutely is. And the \u2013<\/p>\n
Annalee Newitz (13:25):
Don\u2019t you think its landscape is circuitry and the wires it travels through and the ether that it travels through that it\u2019s part of a radio wave and it\u2019s part of an electrical current. It has a physical embodiment. Stop denying that it has a body<\/p>\n
Robin Sloan (13:40):
Robin. I shouldn\u2019t have better than try to outplay Analynus. It\u2019s just a normal physical body like the vast micorrhisal mat that connects all trees in the forest. Thank<\/p>\n
Annalee Newitz (13:52):
You. See, you understand.<\/p>\n
Robin Sloan (13:53):
Outplay it again.<\/p>\n
Annalee Newitz (13:54):
Listen, you\u2019re the one that wrote about sentient sourdough bread, so you started it.<\/p>\n
Robin Sloan (14:00):
Fair enough. Do you think that this boom that\u2019s been driven by this technology, do you think it\u2019s different in scale or kinder character from the San Francisco booms before? Do you think it\u2019s kind of just of a piece or how does it strike you as a longtime<\/p>\n
Annalee Newitz (14:16):
Chronicler? Yeah, that\u2019s a really good question. I think it\u2019s the same. So if it\u2019s gold, if it\u2019s microchips, if it\u2019s models, LLMs, it\u2019s all kind of the same thing because what ends up happening here in San Francisco is we get just crazy investment and huge levels of investment. I think $200 billion have been spent investing in AI in San Francisco alone over the last six years. It\u2019s<\/p>\n
Alexis Madrigal (14:44):
Like half the money that\u2019s gone into it in the world.<\/p>\n
Annalee Newitz (14:47):
Yeah. Imagine if we put even part of that money into say, I don\u2019t know, public transit or a smart grid, which actually is AI. A smart grid is AI. Why don\u2019t we invest in that? So I think that is the part that\u2019s very San Francisco. We\u2019re just going to put billions and billions of dollars into this thing that we don\u2019t really know what it is. It\u2019s like a tulip or maybe it\u2019s gold or maybe it\u2019s some other thing. Yeah, it\u2019s definitely in character for this city.<\/p>\n
Alexis Madrigal (15:16):
I have to say I find it a little more thrilling than the couple previous booms. The kind of Uber for everything boom was really kind of sad in a labor exploitation hustle kind of situation. And then before that you had social media, which Rob and I have talked about as being maybe one of the great oxbow lakes in technological history. Eventually people will move on in some crucial ways. In fact, they kind of already have, at least from that kind of Mark Zuckerberg social network kind of sense of social media.That\u2019s gone for most people now. And yet San Francisco was the heart and soul of that and the Bay Area.<\/p>\n
(15:55):
This one, I guess I find it interesting because I am really interested both in language and consciousness. And it seems to be just driving stakes through kind of cherished ideas that maybe I had about these things, that you couldn\u2019t get this kind of linguistic performance without a conscious thing there. And it just turns out actually that does work. And I think the slippage between what you might expect and what actually happened is way more interesting than if you say, \u201cWhat if we got rid of taxi medallions and anyone could drive around and pick you up?\u201d That you could kind of imagine. It\u2019s just kind of like a ground game logistics operations thing. And this has some mystery at the center of it where even the people working on it don\u2019t totally get it.<\/p>\n
Robin Sloan (16:46):
I mean, you\u2019ve put your finger on an important distinction there and that is a difference. It\u2019s organic. It\u2019s a technology that I don\u2019t know if it\u2019s fair to say for the first time is kind of as much organic as engineered or technical.<\/p>\n
Alexis Madrigal (17:01):
But sort of like an organism, it\u2019s how those things on the multi-scale things work to generate the thing is still, that part is kind of mysterious. So that part of it I find very interesting. And at the same time, it\u2019s also occurring against this socioeconomic and political backdrop that is so much more grim than when you first arrived in San Francisco in the 1990s, which was the survey data bears it out. It\u2019s not just our vibes. People basically thought the world was getting better. And that is not across the political spectrum how people feel in 2026 as we\u2019re staring down whatever is happening inside these models.<\/p>\n
Robin Sloan (17:42):
Is the mystery compelling to you, Annalee? Or are you kind of like, \u201cNo, I\u2019m okay.\u201d<\/p>\n
Annalee Newitz (17:51):
I know what you mean by the mystery. It\u2019s that we have this technology where you\u2019ll talk to a developer and they\u2019re like, \u201cActually, we put in these inputs. We get some outputs. We don\u2019t fully understand what\u2019s happening in between.\u201d I mean, people say that about Microsoft products all the time too. You know what I mean? They\u2019re like, \u201cOh, this product is so big. Thousands of engineers contributed to it. No one person understands Windows.That ship has sailed. No one person understands a video game, like a premier video game because again, thousands of people have contributed to it. So calling it a mystery, sure, it\u2019s a mystery. But the other thing about that is, as you guys were talking, and I think we all fall into this because we\u2019re writers and we deal in language and we\u2019re like, but language is the coolest, most awesome thing that makes us human.<\/p>\n
(18:42):
And that\u2019s what we have over all of the other animals. We\u2019re an animal, but we\u2019re special because we have language. What if it turns out language is just kind of nothing big? What if it\u2019s just like, oh, the reason why LLMs can do this so easily is because it turns out language is no big deal. And it\u2019s actually not what makes us cool. What makes us cool is something else. That<\/p>\n
Robin Sloan (19:05):
Would be one of the great kind of recalibrations of all of history, actually. I welcome it.<\/p>\n
Alexis Madrigal (19:11):
Yeah.<\/p>\n
Annalee Newitz (19:12):
Especially written language.Because think about how bullshit written language really is. As somebody who I write a lot \u2013<\/p>\n
Robin Sloan (19:19):
Hold on. McMillan, we have a new strap line for to print across the front of all of Annalee Newitz\u2019s novels.<\/p>\n
Annalee Newitz (19:26):
Just get the audiobook people. But we think of written language as being this. It was this amazing technology. It has a mystery at the heart of it because oftentimes you can come across language without knowing where it came from. And you\u2019re like, who wrote this? What even is this? Linear A, still trying to figure it out. And I write a lot about archeology in my science journalism life. And it really is the case that ancient civilizations that left writing behind are often harder to understand than ones that did not. And the reason why is because we get so tricked by the language and we\u2019re like, \u201cOh, well, this culture was just all about King Jaya Varman.\u201d Yeah. That\u2019s all they though about all the time. And it\u2019s like, no.<\/p>\n
Alexis Madrigal (20:08):
The one segment they found in the trb. Woodscriribe.<\/p>\n
Annalee Newitz (20:11):
Yeah. It was like the king had the money and the power to make people write stuff. But then there\u2019s all these other people that we\u2019re not even thinking about because we get so enchanted by this one piece of writing that we have. Or in Rome where it\u2019s like the only people writing are the guys with power. And they\u2019re saying all kinds of nasty stuff about freed slaves and sex workers. And it\u2019s like, but we don\u2019t know what the truth was. So the writing often hides the truth is what I\u2019m saying. So writing is cool. I love it. I\u2019m so grateful for it. It\u2019s something I am good at. But maybe it\u2019s just kind of yet another \u2013 The fact that it can<\/p>\n
Alexis Madrigal (20:50):
Reproduce those patterns of our language to you does not impress you as much as it might some other people.<\/p>\n
Annalee Newitz (20:58):
Well, I think, like I said, maybe that\u2019s the recalibration is maybe we need to say, maybe it\u2019s not impressive. Maybe it shows us that actually writing wasn\u2019t really that cool of an invention in the first place. And we love writing, keep writing, but let\u2019s keep going. Yeah, but<\/p>\n
Robin Sloan (21:12):
Let\u2019s talk about some other better inventions like the noodle.<\/p>\n
Annalee Newitz (21:17):
Like cooking. It<\/p>\n
Robin Sloan (21:18):
Turns out<\/p>\n
Annalee Newitz (21:19):
That cooking might actually be a better<\/p>\n
Robin Sloan (21:21):
Invention than writing. I have to say, sometimes I feel like our conversations about this, and it comes back to humanness and the things that we love and sort of are proud of in almost a species level. And they feel like these bullworks, and then you set them up and they do kind of get challenged or even knocked down. In my novel Sourdough, I make a big deal about the egg test and how these amazing agile robot arms and they can do all these great things, but they cannot crack an egg with the grace of a human cook just knocking it with one hand. There is a robot arm company down in Palo Alto. They got these new robot arms. Now I don\u2019t \u2013<\/p>\n
Alexis Madrigal (22:01):
CEO emails<\/p>\n
Robin Sloan (22:01):
You check this<\/p>\n
Alexis Madrigal (22:02):
Out.<\/p>\n
Robin Sloan (22:03):
I don\u2019t know if they read sourdough. I don\u2019t know if they read sourdough, but they\u2019ve got some really slick new demos and they are cracking some eggs. And I will tell you sincerely, I found it kind of both delightful and disturbing because I\u2019ve thought about that so much. And I did. I sort of believed in it. And I think that\u2019s one of the challenges of this time.This has been happening for the last 10 years with these technologies. There have been these really easy ways to say, yeah, yeah, nice, but it can\u2019t X. And then it X\u2019s. And how do we think about that?<\/p>\n
Alexis Madrigal (22:35):
I mean, I\u2019m also curious about this, Annalee. I would say in your fiction, you come across to me as pro robot. You would like to have robots here in this room with us and you could interact and hang out with them and<\/p>\n
Annalee Newitz (22:48):
Run on the<\/p>\n
Alexis Madrigal (22:48):
Streets of San Francisco. But how do we get from here to there minus, I don\u2019t know, some trillions of dollars worth of investment into LLMs and other things? Is there a preferred route of AI development for you that would somehow go around some of these kind of economic and political issues that you\u2019re thinking about? That\u2019s such a good question.<\/p>\n
Annalee Newitz (23:11):
It is a really good question. I think that one possibility is that we think about this as an economic problem. There\u2019s way too much money going into the development of LLMs. And we\u2019ve seen the results, which is that it\u2019s getting stuffed into every product, into every service, places where we don\u2019t want it, where we don\u2019t need it, where it\u2019s actually hindering productivity rather than bolstering it. And so that\u2019s a very bad side effect of this huge amount of investment. I don\u2019t think we should invest no money in it. I\u2019ve heard lots of accounts that Anthropics products really do help with the development process and software. I heard a lot of people crying about that like, oh, all the things I was trained to do, I don\u2019t have to do anymore. But at the same time, it\u2019s exciting. It is a tool. It\u2019s like a calculator would\u2019ve been back in the 19th century.<\/p>\n
(24:04):
It\u2019s like, oh, this can help us get to a certain point and then now we can develop on top of it. I would love to see more involvement of the humanities and social sciences in the development of LLMs. What if we poured some of that AI money into universities where we study the human impacts of this? Because all of these companies are claiming we are going to build humans, but we\u2019re not going to consult the humanities, which is literally the study of humans and culture. It\u2019s like, how do you make humans? How do you make people without understanding society? And I\u2019m sorry, but software development, love it. It\u2019s great. It\u2019s not the study of culture and society.<\/p>\n
Alexis Madrigal (24:49):
I do think that a lot of the people who work in this field who now are scattered throughout the streets of San Francisco, more or less think like humans suck at figuring stuff out. That the way to do it is to just plow all the information into the machine and have it find the patterns of what is human? Well, this thing is going to figure it out. We\u2019re just going to keep feeding it every word ever put out by humans, every video, every uploaded by human. Every podcast episode. Every podcast. And then out at the other end, they\u2019re just going to say, what is a human? And something will emerge. Some pattern will be there. And that\u2019s to me, the sort of bounce on effect of this for the city of San Francisco is we\u2019re now like, there\u2019s tons of people here who now think in the terms of these large language models and think in the terms of AI.<\/p>\n
(25:44):
And I\u2019m not sure what that\u2019s going to do.<\/p>\n
Robin Sloan (25:46):
The term of art for that in the industry and in academia is the bitter lesson. It\u2019s literally called the bitter lesson of data, which also sounds like the title of a good novel that I would read. Yeah, bitter<\/p>\n
Alexis Madrigal (25:57):
Lesson. I would<\/p>\n
Robin Sloan (25:57):
Read that. Yeah,<\/p>\n
Alexis Madrigal (25:59):
Yeah,<\/p>\n
Robin Sloan (25:59):
Yeah. Totally. So yeah, people are walking around basically bitter lesson pilled. And it\u2019s like, we\u2019re not going to reason our way \u2013 We\u2019re not going to figure anything out. Or puzzle our way out. We\u2019re just going to. Yeah. Totally. Yeah.<\/p>\n
Annalee Newitz (26:09):
It\u2019s very nihilistic and it is very anti-human. I think that it\u2019s as much a reflection of contemporary politics as it is of any kind of technology. And again, all of the money going into this technology is a political choice. Saying that this is the primary technology that\u2019s going to drive our understanding of ourselves. That\u2019s again, a political choice. The technology is just going along doing sentence completion or whatever. It really is, as Emily and Bender has said, a stochastic parrot. It says horoscope or something. But the idea that we\u2019ll shovel all of this data into a machine and get answers about people or about society, I think is a symptom of a culture where we\u2019re moving toward authoritarianism, where the idea is that some single centralized entity will figure everything out and make decisions for us. And then we don\u2019t have to worry about what actual people say that they want because the machine says, \u201cWell, what you need is X.\u201d Asking the machine what humans are is a way of avoiding democracy, avoiding voting, asking the actual humans who they are.<\/p>\n
(27:23):
We don\u2019t want to hear from them. They don\u2019t know. And it\u2019s like, actually, it turns out we know best what\u2019s good for us. And our voices should be heard, not filtered through a machine, through a model that distorts what we mean. Actually just talk to us. Ask us what we want. I just want trans people to be free. And I just want to have public transit that works. And you\u2019re not going to get an AI to explain that to anyone. You have to do it at the voting booth. You have to do it in the streets.<\/p>\n
Robin Sloan (27:56):
Hearing you describe that, it kind of creates a paradox. It\u2019s a very vivid paradox of this Titanic effort. I mean, sort of a construction effort, an infrastructural effort, a financial effort driven at least in part by exhaustion and sort of a community, a society saying, \u201cI give up. We can\u2019t figure this out. We\u2019re not going to try anymore. Let\u2019s just build a machine God that can<\/p>\n
Alexis Madrigal (28:21):
Do it for<\/p>\n
Robin Sloan (28:22):
Us.\u201d So what\u2019s the next five years look like right here in San Francisco? Forget the world. Just think about this. We\u2019re in the hot crucible of this weird new<\/p>\n
Alexis Madrigal (28:31):
Thing. Just crash out.<\/p>\n
Robin Sloan (28:31):
Yeah. Yeah. What do you think it looks like, Annalee? That\u2019s the first question. And then we\u2019ll finally close out with what should we do? But what do you think? What do you think is just ahead of us now?<\/p>\n
Annalee Newitz (28:42):
I mean, there\u2019s a couple of pathways. I mean, there\u2019s many pathways. This is a city that loves to burn itself down. We are right near a massive earthquake fault. Bad things could happen. We could be on fire. We could be recovering from all kinds of \u2013 Are<\/p>\n
Alexis Madrigal (28:58):
These the hopeful scenarios or are they<\/p>\n
Annalee Newitz (28:59):
Not helpful? I\u2019m just saying \u2013 Did you ask for hopeful scenarios?<\/p>\n
Alexis Madrigal (29:02):
No, no, no, I wasn\u2019t. I was just<\/p>\n
Annalee Newitz (29:03):
Curious.<\/p>\n
(29:03):
I think we\u2019re going to see a backlash to that investment model. We\u2019re also going to see, and Robin, you and I were talking about this a little bit before we started recording, and a backlash against just the internet, information and language that we receive from places that we think of as kind of polluted by AI slop. I hear a lot of people talking about cozy technology. How do you create the feeling of coziness that was torn away from us when Twitter turned into X? So I think we\u2019re going to see new kinds of community spaces, for example, like local economy, which Alexis, you\u2019ve worked on so hard in Oakland, and I just think is absolutely a model of the future. And we\u2019re going to see more and more of that. We\u2019re going to see more new types of community gatherings in older spaces like bars or libraries or schools.<\/p>\n
(29:59):
And I\u2019m more focused on authenticity being associated with something being physical. But I think that we are associating people just talking to each other and writing things down with human connection versus if you do it on the internet, you never know if an AI wrote it. Did a person write it? But if I give you a zine that\u2019s all scribbled and weird and I\u2019ve cut things up, you\u2019re like, \u201cThat\u2019s definitely Annalee.\u201d<\/p>\n
Alexis Madrigal (30:29):
I mean, I do think you\u2019re right though, Annalee. I think that there\u2019s a bonus that\u2019s now accruing to the physical world and anything that is. I mean, the problem is the words now can be printed from an AI, but just real actual three-dimensional IRL, real-time interaction suddenly has a bonus that it didn\u2019t have before, which is you know you\u2019re not interacting with AI slop of any kind. Wait, before we let you go, do you want to tell us about your new book just before you<\/p>\n
Annalee Newitz (30:58):
Want to do that? Yeah. So I\u2019m working on two things right now. One is I have a new book coming out in October, which is called A Wall Is Also a Road, and it is about a cellular slime mold who is a grad student. Classic. And the other thing I\u2019m working on is I\u2019m helping to reboot the magazine Star Log as a print publication.<\/p>\n
Robin Sloan (31:18):
As discussed.<\/p>\n
Annalee Newitz (31:19):
No AI slop, nothing. It will be only print and that\u2019s coming in November. So I\u2019m very excited.<\/p>\n
Robin Sloan (31:27):
Amazing.<\/p>\n
Alexis Madrigal (31:27):
So much to look forward to.<\/p>\n
Robin Sloan (31:29):
Yeah. Annalee Newitz, the great chronicler of other minds. Thanks for bringing your real mind here with us<\/p>\n
Alexis Madrigal (31:35):
Of San Francisco.<\/p>\n
Robin Sloan (31:36):
Yeah,<\/p>\n
Alexis Madrigal (31:36):
What a pleasure.<\/p>\n
Annalee Newitz (31:37):
Thank you so much for having me. Yeah, the fandom is very mutual.<\/p>\n
Alexis Madrigal (31:40):
Annalee Newitz, thank you so much. Thanks for the book and the future books and everything else.<\/p>\n
Annalee Newitz (31:46):
Thanks so much for having me.<\/p>\n","protected":false},"excerpt":{"rendered":"San Francisco has been at the center of successive waves of technological and cultural change \u2014\u00a0 from the…\n","protected":false},"author":2,"featured_media":428710,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[30],"tags":[101,103,102,104,106,105],"class_list":["post-428709","post","type-post","status-publish","format-standard","has-post-thumbnail","category-san-francisco","tag-san-francisco","tag-san-francisco-headlines","tag-san-francisco-news","tag-sf","tag-sf-headlines","tag-sf-news"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/us-ca\/wp-json\/wp\/v2\/posts\/428709","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/us-ca\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/us-ca\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-ca\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-ca\/wp-json\/wp\/v2\/comments?post=428709"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/us-ca\/wp-json\/wp\/v2\/posts\/428709\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-ca\/wp-json\/wp\/v2\/media\/428710"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/us-ca\/wp-json\/wp\/v2\/media?parent=428709"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-ca\/wp-json\/wp\/v2\/categories?post=428709"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-ca\/wp-json\/wp\/v2\/tags?post=428709"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}