# Taste Is The Only Moat Left | NYU Professor Ruby Thelot Explains MTS · 2026-07-28 In this episode, NYU professor and cyber ethnographer Ruby Justice Thelot unpacks how internet culture has balkanized into distinct digital communities, each with its own vernacular and norms, and argues that taste—historically a mechanism for virtuous consumption in an age of abundance—is the only moat left in an AI-driven world. Thelot explains that 50% of internet traffic is now bots, introducing 'machinic taste' where algorithms, audiences, and agents shape aesthetics and behavior. He reveals that while Americans broadly distrust 'AI' due to fears of job loss—as shown by his analysis of 5,000 videos on the topic—they enthusiastically use chatbots for practical tasks like fixing a faucet, a sentiment captured in Anthropic's survey of 85,000 users. Thelot also discusses how trends like 'heteropessimism' are often misidentified because journalists' algorithms amplify loud minorities, and he predicts techno-optimism will rise only when AI's benefits directly align with people's desire to remain employed and provide for their families. ## Questions this episode answers ### What does Ruby Thelot mean by 'taste' and its historical significance? Ruby Thelot explains that taste emerged in 17th-century England as a guide for the new wealthy to consume properly, with thinkers like Shaftesbury defining it as harmony with nature. Today, amidst AI-driven abundance, taste reclaims its function of helping us consume virtuously to avoid societal decay. [18:23](https://mts.podhood.com/b299a583-a527-417c-a299-96df1015711d?t=1103000) ### Why do Americans dislike the term 'AI' but still use AI chatbots? Ruby Thelot notes that while Americans associate 'AI' with job loss fears, they readily embrace specific tools like ChatGPT. For instance, mothers in Bryan, Texas use it to fix household leaks by uploading photos, viewing it as helpful like Google, not threatening employment. This reveals a language disconnect between the technology's perceived and actual impact. [25:59](https://mts.podhood.com/b299a583-a527-417c-a299-96df1015711d?t=1559000) ### What is the Balkanization and Babblification of the internet? Coined by Ruby Thelot and Ruyi, the Balkanization refers to the internet splitting into insular digital enclaves. Within these isolated communities, speciation occurs: they develop unique vernaculars, leading to babblification, where the same word means different things across groups, making communication dangerous. [8:46](https://mts.podhood.com/b299a583-a527-417c-a299-96df1015711d?t=526000) ### What is machinic taste according to Ruby Thelot? With bots making up 50% of internet traffic, Thelot argues that as creators post for both human and bot audiences, an aesthetic feedback loop emerges. This 'machinic taste' represents the aesthetic preferences shaped by agent-to-agent interactions, distinct from human or algorithmic influence, and will become more pronounced as bot prevalence increases. [11:44](https://mts.podhood.com/b299a583-a527-417c-a299-96df1015711d?t=704000) ## Key moments - **[0:00] Intro** - **[1:07] Cyber Ethnography** - [1:13] Q: What is a cyber ethnographer? A: Ruby Thelot defines it as someone who studies groups of people online and how they behave and what they do. - [1:46] Ruby Thelot categorized Instagram story likes into content appreciation, romantic affiliation, and friendship continuation. - [2:01] Ruby Thelot studied Instagram's close friends feature to understand how users decide who gets access to digital spaces. - [2:55] Ruby Thelot cites Dana Boyd's MySpace Top 8 research as inspiration for his study of Instagram's close friends. - [3:21] Ruby Thelot: 'Every novel technology leads to changes in being, communicating, and belief.' - **[3:48] Trend Detection** - [4:20] Ruby Thelot on spotting trends: 'Is this a real thing or is this three TikToks in a trench coat?' - [4:36] The 6-7 trend is real and has seeped into real-world behavior, Ruby Thelot confirmed through teacher interviews. - [5:09] Ruby Thelot's analysis of 2,000 TikTok dating videos found that heteropessimism has decreased and most dating content is positive. - [6:13] Journalists' personal algorithms lead to overreporting of micro-trends as widespread phenomena, says Ruby Thelot. - **[7:11] Culture Origins** - [7:51] Ruby Thelot describes the current media landscape as a pluriculture with many active cultural epicenters rather than a shared monoculture. - [8:28] The internet has balkanized into small digital enclaves that develop their own vernaculars, a process Ruby Thelot calls 'babbilifikation'. - [9:31] In isolated digital communities, the same word can have vastly different meanings, making cross-communication difficult, Ruby Thelot warns. - [10:37] H&M sold Dime Square shirts based on Ruby Thelot's research mapping small digital communities for marketing. - **[11:08] Machinic Taste** - [11:30] Cloudflare reports 50% of internet traffic is bots, complicating content creation, notes Ruby Thelot. - [11:59] Ruby Thelot explains how creators face audience capture, algorithmic capture, and now machinic capture as bots become half the audience. - [13:02] Ruby Thelot introduces 'machinic taste': the aesthetic preferences that emerge from agent-to-agent connections in a bot-dominated internet. - [13:33] A diverse LLM ecosystem could create machine subcultures with distinct aesthetic preferences, Ruby Thelot speculates. - **[13:57] New Aesthetics** - [14:03] New mass aesthetics are rare because culture has been subsumed by finance, leading to a proliferation of sequels, says Ruby Thelot. - [14:45] Ruby Thelot: new aesthetics emerge from synthesis, like a kid making Brazilian funk mixed with J-pop while playing Counter-Strike. - [15:35] Critics no longer uplift truly novel works because ad-based platforms reward popularity over quality, Ruby Thelot argues. - **[17:39] History of Taste** - [18:23] In his book 'A Few S's on Taste,' Ruby Thelot argues that taste is not just aesthetic preference but a way to consume virtuously. - [18:57] Ruby Thelot explains that the concept of taste emerged in the 17th century after England's Glorious Revolution to guide the new elite on virtuous consumption. - [20:02] We are in a similar transformative era as the 17th-18th century, facing AI-generated abundance and the same question of virtuous consumption, says Ruby Thelot. - [20:43] According to Shaftesbury and Pope, taste is harmony with nature, and tastelessness results from excessive luxury and lack of proportion, Ruby Thelot recounts. - [22:27] Ruby Thelot: taste is fundamentally about how to deal with abundance so that it doesn't destroy society, and now we face an abundance of intelligence. - **[22:43] Cultural Contrasts** - [23:39] Ruby Thelot was banned from Less Wrong for an essay that became a talk 'The Devil You Know' at Dataworks VC fund. - [24:48] Ruby Thelot observes that different US regions have distinct relationships with technology, language, and software use. - [25:28] Ruby Thelot recounts a San Francisco teen debate: is 14 texts from a boy obsessive or endearing? Novel communication modes need new social rules. - **[25:51] AI Sentiment** - [25:59] Ruby Thelot finds Americans distinguish between 'AI' (negative, job-threatening) and 'chat' (positive, like a helpful tool), even though they're the same technology. - [26:47] A mother in Bryan, Texas used AI to fix a faucet leak by taking a photo, illustrating practical AI use, Ruby Thelot shares. - [28:00] Ruby Thelot predicts techno-optimism will rise when AI's benefits clearly align with Americans' basic needs like employment and feeding families. - [28:51] Ruby Thelot: 'People want to be employed. It’s quite simple.' - [29:32] Ruby Thelot's firm is publishing a paper analyzing sentiment of 5,000 AI-related videos to track positivity over time. - **[29:47] Social Cues** - [30:15] Ruby Thelot: 'Everyone should be charm maxing.' - [30:36] Ruby Thelot notes the 'Summer of Lud' movement in New York: Gen Z and Gen Alpha rejecting phones to dance and socialize in parks. - [31:14] Ruby Thelot highlights conflicting text response norms: some expect 24 hours, others three days, leading to social friction. - [31:40] Ruby Thelot argues online culture is based on platform, not ethnicity: there's a Tumblr culture, an X culture, etc., each with unique norms. - **[32:52] Closing** - [33:08] NYU design students are enthusiastically using AI to broaden their thesis projects, Ruby Thelot reports, enabling engineers to fabricate and increasing project scope. ## Speakers - **Sofia** (host) - **Ruby Thelot** (guest) ## Topics Taste ## Mentioned Instagram (product), TikTok (product) ## Transcript ### Intro **Sofia** [0:00] Americans don't like AI. **Host** [0:02] Mm-hm. **Sofia** [0:03] AI is a very broad term. They all use chat. And like, I love it, I love— I love chat, but they don't like AI. There's the boogeyman of, like, "I'm going to lose my job." **Host** [0:12] Mm-hm. **Sofia** [0:13] But when I talk to mothers in Bryan Tech, they're like, "Oh yeah, like, I had a leak in my— in my faucet, whatever." I took a photo, and in about an hour or two I was able to fix that thing. I love that. There's a bit of a language issue here, around what does this technology do for me, versus what is the negative impact that's coming in a somewhat near potential future. **Host** [0:37] Today I'm joined by my co-host Sofia, as well as Ruby Justice Thelot, who is a designer, artist, cyber ethnographer, and professor at the Integrated Design and Media at NYU. So he's written about everything from internet folklore to digital memory to AI companions, creativity, and the future of culture. And I think he has one of the most interesting perspectives on what it really means to live alongside this technology. So, Ruby, I am so excited to have you on today. ### Cyber Ethnography **Sofia** [1:07] The question on everyone's mind: what is a cyber ethnographer? And also, how has the field evolved through the years? **Ruby Thelot** [1:13] So a cyber ethnographer is someone who studies groups of people online and how they behave and what they do. So my research ranges from studying furries all the way to now studying how people are using Instagram. **Host** [1:25] Wow. **Ruby Thelot** [1:26] Um, I had a recent, uh, a piece for Protein where I asked people how they— what they understood, what an Instagram story like meant to them. And so around 500 people, uh, 20, uh, 20 sort of closer interviews to build a lexicon, a glossary of what does this digital interaction mean to people. Is it content appreciation? Is it romantic affiliation? Or is it just, you know, friendship continuation? Which are the three sort of— the three-partite meaning structure that I built for Instagram story likes. Uh, with the Mozilla Foundation recently, I did research on close friends. Okay, so we have this new thing now called close friends on Instagram. If you're in my close friends, are we acquaintances? Are we, like, families? And how do we choose who gets to be in these different sort of digital spaces? These are novel behaviors that did not exist two years or three years ago, and we need to build understanding and language around them. **Host** [2:20] Yeah. **Ruby Thelot** [2:20] Yeah. And that's a big part of my research. **Host** [2:22] Yeah. What's interesting here is that cyber ethnography is not something that's necessarily new. This is not a new field. However, um, my hunch is that this field has probably been changing a lot in the last few years, or at least in the past decade, with just the progress of technology and the introduction of new types of social medias, new types of online behaviors. Even just Gen Alpha, just a generation underneath, has a whole different interaction with their technology than we do as Gen Z, or than millennials above us. And so I'm curious to get your perspective on how this whole space has changed. **Ruby Thelot** [2:55] I mean, there's the— there's the goats like Dana Boyd. I mean, she's a big inspiration for my own research. Um, a big part of me trying to understand what close friends meant was looking at her research on MySpace Top 8 when she was— she was at Microsoft for a long time. Now she's at Cornell as a professor, um, and she was leading a team on Microsoft research. And so people have been doing these things for a long time, but of course, the cycles are accelerating. **Host** [3:20] Yeah. **Ruby Thelot** [3:21] And the features are getting shipped faster and faster. And we have these changes in, you know, I like to say that every novel technology leads to changes in being, uh, communicating, and belief. And the faster these things happen, the less time we have as researchers or academics to do the research and do the interviews. But I think it still needs to be done because millions, if not billions, of people are using these tools and these features. **Host** [3:48] Yeah. **Sofia** [3:48] Yeah, totally. So one of my questions was, like, basically how long does it take to identify a new social phenomenon online? And then how many of them are actually perceptible versus imperceptible? Because I think, for example, like, a lot of teachers, if they have young kids and they're teaching like Gen Alpha, they will, like, be very confused for a long time about, like, what is this new 6-7 thing? ### Trend Detection **Sofia** [4:11] Um, and so they— like, there's a whole layer that they are just, like, not privy to. So yeah, what do you see? **Ruby Thelot** [4:17] So this is one of the hardest things to do on the internet. **Sofia** [4:20] Yeah. **Ruby Thelot** [4:20] I like to say, "Is this a real thing or is this three TikToks in a trench coat?" Right? Because oftentimes what feels like a trend, what feels like a novel social behavior, can just be super micro, super small. **Sofia** [4:35] Right. **Ruby Thelot** [4:36] And so what my firm, R2S, what we do at 131, is we do a lot of, like, quantitative analysis of social media to see how prevalent is this thing really. 6-7 was definitely real. Not only do we see an increase in the mentions of that on social media, we also see the behavior seep through the IRL. **Host** [4:53] Yeah. **Ruby Thelot** [4:54] And when you're doing interviews with teachers, when I'm traveling the country asking people how they're using technology, I'm also seeing that sort of leap over into the physical world. Um, a good example of this is, uh, there's been a few articles about this thing called heteropessimism. This is the sort of aversion that certain straight women have about dating men. And I was like, "Okay, is this a real thing?" So we went— we did— we took a look at the top posts about dating on TikTok. Um, anything over a million views, take a look at 2,000 videos. And then, um, you know, using the scripts, identifying, "Did heteropessimism grow since 2020 on TikTok, Reels, uh, and across different platforms?" And we found that dating content is mostly positive, but our beloved journalists, I think because of their algorithms, I think, are getting hit with a lot of that content. **Host** [5:44] Right. **Ruby Thelot** [5:44] And then they think it's a trend. But when I take a look at the data of the top 2,000 videos across five years, I can say that, like, it's actually kind of decreased over time. And most dating content is, for the most part, positive, about 40 or 50 percent. **Sofia** [5:58] Yeah, that's fascinating. So there is, like, a sort of loud minority phenomenon happening. **Ruby Thelot** [6:03] Exactly. **Sofia** [6:04] With content. How many of the trends do you see are, like, entirely sort of, like, the proportion of them is just, like, wrong, like erroneous? **Ruby Thelot** [6:13] Well, and it's not only the loud minority, it's also the people who write about them have specific algorithms. **Sofia** [6:18] Yeah. **Ruby Thelot** [6:19] You know, they have specific lives that push certain things to them. And so that also has an influence to what gets reported. **Sofia** [6:25] Yeah. **Ruby Thelot** [6:26] So I'm a critic, I'm a— I write for magazines, and sometimes I'm looking for a piece and I'm going to go on social media and I'm going to try to find something interesting. That does not mean, whether it's a fashion trend, whether it's a behavior like 6-7, that that actually has any prevalence in the rest of America, you know? **Host** [6:42] So how do you determine that? How do you actually determine what has prevalence? How do you determine what might just be a biasy that's on your algorithm or on a different journalist's algorithm? **Ruby Thelot** [6:50] Well, you do research, but, you know, sometimes when you want to write a piece, you don't have time to do full research. So you actually will interview the TikToker, the person who's made that first post. **Host** [7:01] Yeah. **Ruby Thelot** [7:01] So there still is something happening, but the question to me is always, is this only in Bushwick or is this actually happening outside? **Host** [7:10] Yeah. That's so true. ### Culture Origins **Sofia** [7:11] So what have you observed about where culture starts? Because, like, there's been a lot of, like, culture theorists that talk about this, of how, like, you know, they're very contained environments that sort of breach containment. And it's all, like, very, you know, small subcultures that are honestly pretty weird. Like, for example, the, like, incel subculture ended up, you know, trickling out. And, like, now everyone uses terms like looksmaxxing, for example. And there's been a million examples of this sort of thing happening online. You mentioned furries. I feel like this is another one where it's like a very core group of people and then, like, things just start, like, coming out of there. Like, these start emanating out of there. **Ruby Thelot** [7:51] Yeah, we're definitely in a period of what I would call a pluriculture,right? Where before we had 10, 5 channels, we all watched the same things. Now, um, we are in a pluriculture where there's very strong, very generative, um, very active cultural epicenters that are kind of sometimes touching each other, sometimes far away from each other. Something that feels as big as, like, a show like White Lotus or Succession, if you take a look at what the viewership for that is compared to a show like Mash, compared to even the finale for Lost, not that many people have seen it. But if you're on X, it feels like everyone's watching Succession. **Sofia** [8:27] Right. **Ruby Thelot** [8:28] And yet the numbers kind of, when you compare them to even a show like Big, more people watching Big Bang Theory than watching Session, which kind of sounds crazy when you think about it. And so I think what's very important for me, and, you know, my friend Ruyi and I worked on this, uh, on this paper called The Balkanization and Babblifikation of the Internet. **Host** [8:45] Mm-hm. **Ruby Thelot** [8:46] So the Balkanization is this splitting of our monoculture into these small digital enclaves, you know, they've been balkanized. And on our little insular digital islands, what do we do? We have speciation. We develop our own languages, much like the turtles on the Galapagos Islands,right? Our own specific vernacular, our own specific language. And the problem with containment, with, uh, being so insular, is that when we have something like looksmaxxing or anything, uh, that breaks, well, because now we've been sort of abelified, we've developed our own unique idiosyncratic, um, micro-languages, we no longer know what things mean when we speak to each other online, you know? We're using the same word, but they have, like, completely different meaning if you're in the furry community, if you're in the, uh, incel community, if you're in the, you know, high fashion Twitter, whatever it is, and you're using the same word and it means completely different things because you've been in that digital bubble. And that's the kind of that bablifikation means is very dangerous. **Host** [9:47] And in those digital bubbles too, there's some external incentives to group people into specific individual bubbles because it's easier to market to them. And, you know, even on, like, Spotify, it's like, "Oh, you are a pink pony, princess, uh, pop girl." And then everyone's like, "Oh, like, this is the my culture. These are the people I identify with." And it's almost people want to feel like they're part of a tribe. And at the same time, uh, companies want us to also feel like a part of the tribe. Are you seeing that just happen more and more and more and more, that there's just such tiny sub-niches of our new culture? **Ruby Thelot** [10:23] Yeah, I mean, a big part of my job when I do consulting, you know, anywhere from H&M to Pepsi is to kind of go in these small digital communities and, like, map them. **Host** [10:33] Yeah. **Ruby Thelot** [10:34] So that H&M knows what to sell, you know? **Host** [10:36] Interesting. **Ruby Thelot** [10:37] H&M recently had a Dimesquare shirt last year. And so that is based on research that was done on different digital communities. **Host** [10:44] As someone who was in New York for two weeks over a few weeks ago, I'm not sure if you know Dimesquare, it's the new micro-neighborhood. **Ruby Thelot** [10:50] Yeah, exactly. **Host** [10:51] Oh, I know. **Ruby Thelot** [10:52] Um, but so, so people want to market,right? They want to be able to say, "Okay, I have this specific group." And it also makes the job of the algorithm easier because the title, the name, the nomenclature makes it more legible to, um, the algorithm. ### Machinic Taste **Host** [11:08] How is this now playing into our current modern-day world where you don't just have people who are deciding these algorithms, but rather it's artificial intelligence that are deciding these algorithms? Is this accelerating the development of these micro-communities, of these labelings, of almost these decisions that get made by algorithms, less humans? **Ruby Thelot** [11:30] So, um, I have an essay that came out recently called Machinic Taste for Lorandum Art, um, where I kind of talk about this. So we might have seen the numbers from Cloudflare coming out this year, about 50 percent of internet traffic, as they see it, is bots. What does this do? So you had a period of time where people were posting to their friends. Then they started posting to their followers, and then they started posting to an audience. When a creator or a poster develops an audience, what we see is that their content changes. They become captured by their audiences. Meaning, when they see that something works, they tend to do more of it. The concept of audience capture I get from my dear friend Matt Kline, who's the head of Insights at Reddit. The thing that's happening now is that not only is there audience capture, but to get to your audience, there's also a gatekeeper, which is the algorithm. **Host** [12:17] Yeah. **Ruby Thelot** [12:17] That's why MrBeast looks like that on every single video with the big text,right? And even our beloved tech podcasters also kind of do the same kind of MrBeast face in order to get views. And so you have audience capture, you have algorithmic capture. And now when your audience is now 50 percent bot, 50 percent human, we suddenly see machinic or agentic capture happening too. Because who is responding to your post? We see this on X. It's not just human, it's also bots. **Host** [12:44] Yeah. **Ruby Thelot** [12:44] And so there's this waltz, this three-legged waltz between machine, algorithm, and your human audience. And if we predict, as I do, that the number of bots is going to increase, suddenly it may just be that we will only have bots making things,right? Um, we'll only have, you know, AI-generated content, an AI audience. And suddenly what we'll see is what I call machinic taste. We'll actually get to see what are the aesthetic proclivities, what are the aesthetic preferences of an agent-to-agent connection. **Host** [13:19] Yes. Yeah. That is absolutely fascinating to me. Do you think that, like, agents will have diversified preferences? Or, like, do you think that there will be, like, some sort of convergent, like, you know? **Ruby Thelot** [13:31] Well. **Host** [13:31] Will agents have subcultures is basically the question. **Ruby Thelot** [13:33] Kind of what you guys were talking about before. If we have a rich LLM ecosystem with different kind of models and different kind of weights, then we can one can surmise that they'll have their own specific aesthetic preferences. **Host** [13:44] That's awesome. So speaking of aesthetics, um, you wrote a piece about, or you participated in an event about, like, why are there no new aesthetics? So I wanted to get your take on this because this is something that I think about all the time. Even people like Tyler Cowan have, like, created fellowships for, like, "Somebody please make a new aesthetic." So why do you think this is? ### New Aesthetics **Ruby Thelot** [14:03] So in the lecture, I think the first thing is that there are many new aesthetics. They're just not mass. And it's very difficult to make a new mass aesthetic because, for the most part, mass culture has been subsumed by finance. I give the example of architecture. I give examples in movies,right? If you take a look at the graph of sequels over time, there's way more sequels than there were before. But I think we also have, as an audience, a responsibility to seek out the things that are novel. And, you know, I say new aesthetics require search. New aesthetics require a novel philosophy as well,right? Um, we need to start to think about what the world is going to be. New aesthetics require a novel vision of the future. **Host** [14:45] Yeah. **Ruby Thelot** [14:45] I'll say that new aesthetics emerge from synthesis. Like, it's the kid who's making this amazing kind of, like, Brazilian funk that's, like, mixed with, like, Japanese popright now that's being made on YouTube. And I'm, I think it's a kid with three screens. He's, like, playing Counter-Strike. He's listening, like, he's listening to Brazilian funk and, like, also listening to anime at the same time. And that's how the music sort of, like. **Host** [15:04] Yes. Yeah, yeah. That's become the new aesthetic. **Ruby Thelot** [15:06] Yeah. **Host** [15:06] Brazilian funk anime. **Ruby Thelot** [15:07] Yeah. So, but it's happening. And, like, I think it's not at all, like, the number one. I played on, um, Ella Langley's Chasing Texas because I was like, "That's the number one song in America." **Host** [15:16] Yeah, yeah. **Ruby Thelot** [15:17] And so there's a difference between the revealed preference of, like, "That's the number one song." People say, "We want new things," but are we actually engaging with new things? Are we actually giving that kid a shot? Um, it's very difficult for our current culture industry to do so because all their decisions have been financialized. They will make a movie if they think that, like, it's going to place well, do well. They'll pick a person who already has a following on Instagram if they think the song's going to do well. And the last thing that I deplore, and this is, like, me talking about my own job as a critic, is that critics also are no longer doing their job,right? Because we publish on platforms that are advertising-based. And so we can't be as mean and as critical as we used to be or uplift something that, like, very little people are going to care about,right? There's, like, an incentive to, I think, we know the pop-timism, to be more popular because the platform lives on clicks. **Host** [16:11] Yeah. **Sofia** [16:12] We'll continue monitoringright after this message from our sponsors. Scale your startup on Neon, the Postgres backend for apps and agents, built by Databricks. neon.com/mts. ElevenLabs, AI that communicates at human level across every channel and modality. ElevenLabs.io/mts. Special thanks to our sponsor, Kong. Kong, the AI connectivity platform. Connect APIs, LLMs, agents, and systems with serious security and governance. Try it at konghq.com. Thanks to Blitzy, autonomous software development for enterprise codebases. Ship 5 times faster at blitzy.com. **Sofia** [17:03] Let me tell you about Merge. OpenAI, Dropbox, and Ramp use Merge to get AI to production faster. Merge.dev/mts. Quick word from AdQuick. Make your brand a billboard. Out-of-home advertising as easy to scale as digital. adquick.com. Thanks to our sponsor, Macroscope. AI codebase understanding for engineering leaders. Try it out at macroscope.com/mts. **Host** [17:39] So you're almost designing your own human judgment, your own human taste based off of what these algorithms are probably going to be promoting. I am also curious to get your, uh, takes on what taste even means. You write a lot about it. ### History of Taste **Host** [17:53] And I think this is a very misunderstood term, even I personally. It's like, "Okay," everyone's like, "Oh, taste is the only moat in the AI age," and all of these conversations about it. And I was like, "Okay, personal preference, like having good style." Okay, interesting. Reading your works, you really go through the actual historical meaning behind this and how it's almost connected to morality and all of these other aspects of it. And I think this is really important information that maybe isn't talked about quite enough. I'm curious to get your perspective here. **Ruby Thelot** [18:23] So I wrote a book called A Few S's on Taste, where, um, the first essay is called A Brief History of Taste. And I make the, I make the case that taste is not just aesthetic preference. Taste is a way for people to consume virtuously. Um, the question of taste emerged in the 17th century at a time where people needed guidance as to how to consume properly. So in 1688, the Bill of Rights, uh, happens in England after the Glorious Revolution, and the nobility's powers are restrained. And as they lose political power, they also lose cultural power. At the same time, you have the growth of the British colonies. You have immense amounts of wealth that are being generated in, uh, in the United Kingdom. And suddenly you have this nouvelle bourgeoisie, you have this sort of new elite that emerges, and there's a belief that this emergence of excessive luxury will lead to the decay of society,right? There's even a Masekola writes in Luxury 1977 says, you know, that they believe that, uh, uh, excess would lead to effeminacy, that, like, men would lose their masculinity because things were too good. And so it is in the vacuum, in the reduction of the cultural power of the previous nobility that the question of taste emerges. And so you have people like the Earl of Shaftesbury, people, um, like Alexander Pope who write about this in the, um, 17th and 18th century. What's interesting to me is that we're in a similar position now. We're entering a similar industrial revolution. We are faced with, or at the eve or maybe in the midst of immense amounts of wealth being created, and we're asking ourselves, "How do we consume virtuously?" And that is why the question of taste is emerging. It mirrors 17th and 18th century England in that sense. **Host** [20:17] It's like a self-governing, like, sort of structure that emerges so that people, like, in a way, it's kind of like a survival instinct to, you know, like, so much content being out there and so much slop being out there. **Ruby Thelot** [20:28] Yeah. And so what's notable about this is that what Pope and Shaftesbury tells us, Shaftesbury is the first aesthetician of the English, uh, tradition. He says, "Taste is harmony with nature,"right? Because he's a, it's kind of a deus. He says, "Well, if nature is God, and so nature is beauty, taste, the decisions must be in line with what, what, you know, unity, harmony." Um, and that is his perspective on taste. And so the nouveau riche, the person who's tasteless, you know. I think the way Pope says, like, "His house looks like a, it, it, no proportions, like a big, big gate, and then he looks like an insect next to the gate. The lake looks like an ocean,"right? Everything's too big because there's too much wealth. **Host** [21:12] Right. **Ruby Thelot** [21:12] And so it pushes for a natural harmonious restraint, you know, tasteful decision because it is virtuous to make good aesthetic decisions. **Host** [21:22] Yeah, it's almost, we become ungovernable. Good taste means I can't be fit into this little box that the algorithm is designed for this. And it's almost when people say the importance of taste, it's not just about having a preference or it's not just about the aesthetics of it, but it's this bigger mission of how do you actually become ungovernable? How can you see past these algorithms and all of that? **Ruby Thelot** [21:45] Yeah. **Host** [21:46] Yeah. That is so beautiful. Like, the virtuous point, I just got to keep ruminating on it. Um, how do you think, how else do you think taste manifests beyond just, like, aesthetic preference in the digital age? **Ruby Thelot** [21:58] Um, I mean, what's interesting to me is that I think it could be effectuated around so many different facets of one's life,right? If it is truly virtue, then virtue is about, like, daily actions. **Host** [22:07] Right. **Ruby Thelot** [22:08] Um, not to be, like, guru and self-helpy, but, like, one can make tasteful decisions in many, um, different spheres of the life. I do like, uh, in, in, uh, in SF when we people talk about, like, taste and also problems to find,right? Um, and there are virtuous problems and some problems that are a bit less virtuous, perhaps. **Host** [22:26] Yeah. **Ruby Thelot** [22:27] Um, but fundamentally it is a question of how to deal with abundance,right? Right now we have an abundance of intelligence, abundance of capabilities, and it is a question of how to deal with that so that this abundance does not wreck and destroy society. That is the purpose and the function of taste. **Host** [22:43] In SF speak, we call it, like, signal to noise ratio. ### Cultural Contrasts **Ruby Thelot** [22:46] Signal to noise ratio. **Host** [22:47] That's like the very SF way of putting it. Um, how would you, like, compare and contrast sort of the, like, um, you know, we have, like, the Dime Square phenomenon in New York. It's a very different cultural landscape. And then we have SF. And SF is, like, the complete opposite of all of that. Um, how do you, do you see these cultures sort of, like, merging or talking to each other? Like, how do you see the, how do you see them evolving? **Ruby Thelot** [23:13] Well, I'm not American, so I feel, I feel very much like, um, Alexandre Tocqueville. Like, I'm, I'm here and I'm, I'm just in America and I'm, I'm seeing how things are. **Host** [23:20] Yeah, yeah. **Ruby Thelot** [23:21] Um. **Host** [23:22] That's a good term. **Ruby Thelot** [23:23] Uh, so I think, I think it's, it's all very fascinating to me. Um, I'm learning as I'm going what the different parts of the culture are. Um, you know, uh, I'm met quite a bit of, I mean, I'm quite a bit of rationalist here and I'm kind of learning about that culture. I'm famously banned from Less Wrong. Um. **Host** [23:42] Why? Why? Why? How did that happen? **Ruby Thelot** [23:45] You say certain things and suddenly you're banned from there. **Host** [23:48] Oh, it happens. **Ruby Thelot** [23:49] No, I mean, I, I was trying to have, like, I think what was a, a, a healthy, productive debate. The essay that got me banned actually ended up being a talk that I gave at Dataworks, the VC fund in New York. It's called The Devil You Know. So it's still out there. I think it's cogent and sensible. I don't know why exactly I got banned, but sometimes you speak truth to power and things go bad for you. Um, but, you know, you learn and then, and, and I'm, I'm, I'm, there's the language, there's a vernacular here that's very specific, signal to noise, agency, all that stuff. **Host** [24:17] Right,right. **Ruby Thelot** [24:18] Um, and I mean, New York is, like, highly diverse. It's, there's everything from Dime Square to, you know, like, specific enclaves in the Bronx that are completely different. **Host** [24:26] Yeah. **Ruby Thelot** [24:26] Um, but I think what I'm learning more as I'm traveling the United States and actually asking people these questions about technology is that every single region has its own particularities, frankly. Like, I'm going to LA tonight and LA has its own culture, its own thing going on. Um, and the way that they use their tech is very different. And the way that they talk about it is very different. **Host** [24:47] Yeah. **Ruby Thelot** [24:48] Um, and same thing in New York and same thing when I go to Miami or I'm spending a lot of time in Texas as well. So all these places have different relationships with their language, their tech, and how they use their software. Like, when I'm on a plane and I'm sitting next to a young person, I'm always looking at their phone. And, like, I ask them a question. Okay, I was on the, I was at the San Francisco Art Book Fair yesterday. It was two, two young girls talking about how she broke up with a guy because he was texting her too much. And I was like, interesting. What is the proper amount of time that he should have texted you? Mind you, the guy texted him, texted her 14 times. And so I was like, okay, is there an acceptable amount of time,right? This is a novel mode of communication, which comes with its new sets of rules. Um, and maybe she has a set of rules. And the girl was like, oh no, I kind of like it. He's obsessed,right? And, like, we have to define these things societally because otherwise, you know, you grow up, you're 16 and you don't know what it means to be texted 14 times,right? **Host** [25:44] Yeah. **Ruby Thelot** [25:44] So yeah, that's kind of like the ever, ever ongoing research as I, like, move around the United States and try to understand how people use their technology. ### AI Sentiment **Host** [25:51] How people use their technology or is the, is the conversation more recently how people are interacting with AI or even the sentiment on AI itself? **Ruby Thelot** [25:59] That's very interesting. I mean, so this is, I think, the biggest difference that I've seen for the most part. And this might be, you know, a stated versus revealed preference. I find that Americans don't like AI. AI is a very broad term, a very broad umbrella here because they all use chat and, like, I love it. I love, I love chat, but they don't like AI. Does that make sense? There's like a bit of a divergence there. There's a, there's a, the, the, the boogeyman of the data center. There's the boogeyman of, like, I'm going to lose my job. But, you know, if you take a look at Anthropic's survey, they surveyed 85,000 people to ask how they were using their AI for career advancement, so like at work or life management. And so when I talk to mothers in Bryan, Texas, like, oh yeah, like, I had a leak in my, in my, my faucet, whatever. I took a photo and in about an hour or two, I was able to fix that thing. I love that. **Host** [26:52] Yeah. **Ruby Thelot** [26:53] But to them, it's kind of like Google. And it's, that's not the thing that replaces their work and it's going to, you know, put them on unemployment. So there's a bit of, um, once again, a language issue here, uh, around what does this, what does this technology do for me versus what is the negative impact that's coming in a somewhat near potential future. **Host** [27:12] Yeah. **Ruby Thelot** [27:13] Um, and that hasn't been really sort of split. The good thing, I guess, uh, from an AI perspective is that people for the most part still like essentially free chat. **Host** [27:24] Yeah. **Ruby Thelot** [27:24] That's like quite useful. **Host** [27:26] Right,right,right. Yeah, I think that's absolutely fascinating. I was literally at a mass yesterday because my cousin was in town and, like, our aunt made us go to mass. And the pope, or not the pope, the priest there was, like, anti-AI. He was, like, giving, like, an anti and he was, like, an ex-software engineer. So it was really fascinating. Do you think, like, we're going to trend towards a more, like, maybe a more rigorous, like, categorization of what AI is so that people can actually be like, okay, I disagree with this part, but I do agree with this part. Do you think, like, techno-optimism will be on the rise? Like, where do you think the current, uh, will it stay very, like, you know, segmented? **Ruby Thelot** [28:00] Tough question. Not a seer, but my sense is techno-optimism will be on the rise once the claimed benefits of the technology are aligned with the desires and interests of American constituents. What do Americans want? A wife, some food, some clothes to feed their children,right? The world is called the Gilscot Heron. It's quite, it's simple,right? Uh, we all know kind of like the picture of the American dream, which, you know, in many ways. When you tell, okay, put it very simply. Even Dave Chappelle says this one of the specialist says, you never get between a man and his bread. So I don't know who thought it was a good idea to go on national television and tell people they're going to lose their job, but if I'm an American, which I'm not, of course, but I wouldn't want to lose my job because that's how I feed my family. It's kind of, it's very simple, frankly. **Host** [28:50] Yeah. **Ruby Thelot** [28:51] And I think the thing that I love with doing all my research is that I get to talk to people who put in very blunt terms what, you know, I could have written like a five-page report on this, you know, make a paper out of all this, but it's quite simple. You know, people want to be employed. **Host** [29:08] You'reright. So it's more of the threat to employment you think that is just like. **Ruby Thelot** [29:12] Yeah, I don't think, I don't think your average person cares about exceptional risk, frankly. **Host** [29:16] Right. No, totally. I agree. I wonder too if maybe the people who are talking about this, the journalists who are talking about this, their algorithms themselves are also extra biased to find the things that are going to get people to be, uh, super hyper aware or scared or fear hunger. **Ruby Thelot** [29:32] At R2S and 131, we have a paper coming out specifically on that. Same thing with the hero pessimism. We're going to be analyzing about 5,000 videos on AI to sort of determine what is the current sort of positive negative and see how it tracks over time. It's going to be one of our indexes that we publish, um, in the next month. **Host** [29:47] Yeah. Um, another question because we touched on this briefly. Do you think we're getting better or worse at social cues? Because one could make the argument that there are more social cues to keep track of now that we're online. ### Social Cues **Host** [29:58] And, like, now a social cue is like the story like, you know, like, like we are getting literally more and we have to keep track of more. But at the same time, I feel like IRL, there's like a sentiment that everyone's just getting more like awkward, you know? Like people are less like well-socialized in the real world. So what do you think? **Ruby Thelot** [30:15] I think everyone should be charm maxing, you know? I think we should lean charm maxing. **Host** [30:19] Charm maxing, yeah. **Ruby Thelot** [30:20] You know, the little, the little wave, little finger, little wink. Because for the most part, I think there's a lot of beauty in these sort of human-to-human interactions. Um, we've all heard about the Gen Z stare. You know, you, you ask for your coffee and you get the. **Host** [30:34] Now imagine the Gen Alpha stare. It's going to be gone. **Ruby Thelot** [30:36] So, but the funny thing is there was this big event happening in New York called the Summer of Lud,right? And it's this sort of grassroots Gen Z, Gen Alpha movement. Uh, people try to reject their phone and just going to the park together, dancing, having all these events. I mean, you know, Washington Post and Wired, I think, reported about this. Uh, so I think you're seeing a bit of a pushback against the, uh, infringement of technology across all facets of life. But I think it is very difficult,right? Think about your book of etiquette that your grandma might have had on her, on her table. This is what you're supposed to do at a restaurant. This, uh, when you meet someone, shake their hands. And now you have this times like all the different apps you're on. **Host** [31:14] Right. **Ruby Thelot** [31:14] I can ask the studio, like, what's an acceptable amount of time to respond to someone via text? Some might say 24 hours, some might say three days. And these, and people feel very differently about these behaviors,right? Someone might be insulted that you take 24 hours, but you're like, no, I'm a 72, I'm text is like email, I'm a 72 hour kind of person. **Host** [31:33] Yeah, yeah. **Ruby Thelot** [31:34] Um, and you have to hold these rules not only just for yourself, but for all different people in, people in your life, which is highly difficult. **Host** [31:40] You have to calibrate to somebody's like online etiquette, like whenever you meet them. It's like a different rule for what you do online versus in person, and no one has the guidebook for that yet. **Ruby Thelot** [31:50] And whereas like culture usually is somewhat ethnicity-based or nation-based. Like I'm French-Canadian. When I meet someone, I give them that visa. I kiss them on the cheek. The culture online is not specific to what a person looks like, but more which platforms they're on. I would argue that there's a Tumblr culture,right? **Host** [32:08] Yeah, there's definitely a Twitter culture. **Ruby Thelot** [32:11] There's an Instagram culture. **Host** [32:12] It's interesting even for MTS because we're trying to figure out the algorithms for the different, uh, you know, for X has a very specific culture that they want. Instagram has a very specific culture that they want. And you have to try to figure out how to blend all of these together. And it's almost like, like, even in real life, when you meet someone, you can tell which platform they're on. Yeah, yeah, exactly. You just know how they talk. Yeah, exactly. Based off of like what words they use, like, yeah, like sense of humor, I think, is very platform-specific. **Ruby Thelot** [32:38] 100%. **Host** [32:39] My friends say I have Twitter lingo in my everyday vernacular. I'm like, oh my gosh, I don't know how that. **Ruby Thelot** [32:44] Yeah. **Host** [32:44] Like I remember when I got my first job, my manager was like, you talk internet. Like. **Ruby Thelot** [32:49] You talk internet. **Host** [32:50] I was like, wow. **Ruby Thelot** [32:51] You talk internet. **Host** [32:52] Yeah. ### Closing **Ruby Thelot** [32:52] Yeah. **Host** [32:53] I, I do know we're at time, but Ruby, I did want to thank you for joining us today. And maybe one last thing for you is, I, you are a professor at NYU, specifically in the Integrated Design and Media Lab. What do you think is the most interesting thing that your students are curious about for this next decade? **Ruby Thelot** [33:08] I mean, my students are sort of gung-ho about using AI. Like, I think they quite love it, especially because we're interdisciplinary. So you might be an engineer, like one of my students is an engineer, but they might not know how to fabricate. And so the ability to be like, okay, I want to make a robot. And then having access to that information, it changes your thesis quite drastically. Because before maybe you'd find the information, we have a lot of TAs and a lot of help overall, but I've actually seen an increase in like the size and the scope of the projects. People feel like they are able to tackle them now. **Host** [33:38] Yeah. **Ruby Thelot** [33:39] Um, so that's always super, super exciting. And I, I specifically, I teach thesis. So that's, that's a final grads, uh, grad, uh, grad program project. Um, so that's very, very exciting. And then as always, you know, the goal is not to just make something cool and big, but to be critical and to think about it. Um, so we always try to push them to be, you know, intentional and very conscientious in what they build. **Host** [34:00] Definitely. **Ruby Thelot** [34:00] Yeah, totally. **Host** [34:01] That's a great place to wrap. **Ruby Thelot** [34:03] Yeah. **Host** [34:03] Well, thank you so much for joining us. **Ruby Thelot** [34:05] This is so awesome. **Host** [34:05] Thank you so much. **Sofia** [34:09] MTS is an X native live streaming news and interview show covering technology, business, politics, and culture as it happens. Catch us every weekday live on X, YouTube, or wherever you get your podcasts. --- This library is powered by PodHood (https://podhood.com), the podcast website platform.