SPEAKER_00: I think the people that think AI is going to like revolutionize writing don't actually

SPEAKER_01: understand what writing is. Writing is thinking. Writing is thinking. And every time I've like

SPEAKER_02: tried to introduce an AI into my own thought process, holy, it's such a bad experience.

SPEAKER_03: Could you imagine a world in which you build like a mini medium, like one for India, one for, you know, country X to see if you could replicate the same model in a non-English focused area?

SPEAKER_01: Yeah, a hundred percent. But you're giving me heartburn. And we just barely saved the company

SPEAKER_07: a couple of days ago. I mean, look, this is Twist, not the retirement business show. This is all about

SPEAKER_09: what is next. This Week in Startups is brought to you by Notion. Notion projects combine project

SPEAKER_11: management with your docs, knowledge base, and AI. Try Notion projects for free at notion.com slash twist. Brex, the financial stack founders can bank on. Brex knows cash is king for startups, so they built a banking experience that takes every dollar further. Get the business account trusted by one in three U.S. startups at brex.com slash twist24. And OpenPhone, create business phone numbers for you and your team that work through an app on your smartphone or desktop. Twist listeners can get an extra 20% off any plan for your first six months at openphone.com slash twist.

SPEAKER_15: Welcome back to This Week in Startups. My name is Alex. I am at Alex over on X. And today I have

SPEAKER_16: someone very, very special on the show. Someone that I've been looking forward to talking to for a long time because he is a man of the moment. It is Tony Stubblebine, the CEO of Medium. Tony,

SPEAKER_19: one, hi, and two, how are you? Well, now that I found out I'm man of the moment, I'm feeling really

SPEAKER_16: good. Thank you for that. You are man of the moment because your company is doing well, and we're going to get to some financial stuff later on in the show. But I see you at the intersection of like online content, the AI wave monetization in a kind of fractured ad world. There's a lot of things that I care a lot about that intersect directly with Medium, frankly. So, I mean, you may not like it,

SPEAKER_15: but you are kind of a man of the moment, I think it's fair to say.

SPEAKER_01: Yeah. I mean, we're involved in all those things. I mean, Medium is a blogging platform

SPEAKER_00: that's involved in all of those things, but I feel like we're very contrarian. Like a lot of where the internet is going, we're trying to go in the exact opposite direction. And so, if you think the internet's broken in some way, then you might resonate with me. But the majority of people seem

SPEAKER_22: to be headed down, you know, off that cliff. And I think that's why I'm a little bit, wait, I wish I

SPEAKER_25: was of the moment, you know? Well, I mean, or maybe it's more correct to say that the moment should be a little bit different, but let's start with where you just were, because you and I had a

SPEAKER_27: chat when I was back at TechCrunch on equity. And at the time, you said that Medium today

SPEAKER_16: is a publishing platform that is supported by subscriptions. That money goes to helping people write their very best stuff. It also goes to editors and curators to find the best online, and that you want it to be an alternative to the attention economy. All that's still relevant,

SPEAKER_30: still pertinent? 100% true. That's a great articulation.

SPEAKER_33: Yeah, you should give yourself 10 points. I was prepping for our chat, and I want to start with your point that the internet is, quote, broken and getting worse. This to me is a relatively provocative

SPEAKER_03: statement, but not one that I disagree with. We have your Medium post right here. So talk me through

SPEAKER_27: the argument about why the internet is broken today, and why it's actually becoming more of a problem,

SPEAKER_36: not less. Yeah. I mean, the root cause is the incentives from ad-based models, but the things

SPEAKER_00: that we notice are a lot of misinformation, a lot of anger, a lot of people clamoring for our attention, but not necessarily giving us anything in return. And in a lot of ways, I'm not sure the internet's actually making us smarter. And that was the promise. The promise of this information superhighway is supposed to be really, really transformative in some ways, right? If you dig through it,

SPEAKER_01: but it's such a waste of time. The way the internet is designed is not respectful of your time. At minimum, we can say that, right? And the root cause is ads, because in order to make an ad business, the incentive is merely to get your attention. You don't have to deliver beyond that. In fact, delivering too much beyond that is almost counterproductive. Like as a writer, if I want to show you ads, I just need to get you in the door. And if you're too engrossed in my writing, you're not going to see the ads. It's actually more valuable to have you only partially engaged. And so, I don't know, like people are exploring alternative business models. And like, there's a group of medium employees right now. We should give this to the whole company, but there's a group of us that have I heart K walls t-shirts. And we've just been walking around seeing how people respond to that. But I mean, fundamentally, what we're saying is in order to have an honest business, you have to have a different business model. It has to either be purely volunteer, or it has to be purely donation. I think Wikipedia is a healthy organization, or it has to be direct payment. As soon as you get this indirect payment model where a company is making money off of your time, rather than the value they give you just, it all breaks down. Yeah, that's where we are.

SPEAKER_16: It's funny, because I was I was gonna say Wikipedia to me is the the argument against your argument that Wikipedia actually has become only better over time, broader, more up to date. I'm gonna go ahead and say more trustworthy, though, of course, always verify Wikipedia, but like, it's, it's fantastic. And it has work in this new internet we live in. But it's also not ad based, it's based on direct donations, which is, I mean, I'm going to just guess roughly 10 times harder than subscriptions, but there does seem to be a couple of pockets of good internet out there. But I think our perspective is that most of the internet is a hot mess, and not improving as the dynamics of an advertising based

SPEAKER_01: world don't get better. Right. And then it just got worse with AI, right? Because content, it's cheaper to create high volume content than ever. You know, I think what we see on Medium is something like a tenfold increase in the number of posts, right? Not a tenfold increase in the quality of inbound posts, but just the volume of it. So there's way more work to filter it down to,

SPEAKER_03: to what what's good. That brings us to incentives thing, something you mentioned a little bit ago, I was going through your threads when you were discussing a certain corporate milestone, and

SPEAKER_16: someone asked you, you know, what is driving Medium's recent kind of updraft? And you said, a couple of primary changes, author incentives, recs and curation. I think the net effect of this is that we made a very big shift in which authors people read. So I want to know how you went about balancing the incentives between getting people in the door at Medium, paying writers, taking money, and also supporting the overall kind of Medium landscape, because you're the only company that I can think of right now that has a crowdsourced inbound model, a group subscription, and then payments that

SPEAKER_49: actually works. Yeah, I mean, we're the only company that ever tried to put a bundled subscription on top of a UGC platform. I mean, that's, that's definitely a new model. I think it's part of why it took Medium a while to figure it out. The thing though, that makes that work for us is we knew, basically because

SPEAKER_01: I'm old, that there was a huge market of untapped writers that had just been left out of the current incentives. It's like, what I remember about blogging was in 2003, suddenly you started to hear real stories from people you never would have heard from. Like the deepest knowledge, the deepest wisdom is all locked up in practitioners. It's not hiding out in creators, right? It's like the people actually living a life, you normally don't get to hear from them. I think that's, you know, like, this is why you run the podcast the way you do is like, you bring entrepreneurs on because you want to hear from the source and blogging did that at this massive scale. And then the incentives change in order to get your voice heard in the modern internet, you have to be a growth actor. You have to understand SEO, you have to understand virality. And what we've got are a bunch of winners who are like really good at grabbing your attention, but don't have nearly as much of, you know, the actual knowledge that we want as the, those original bloggers did. So when I said in that post, I redid recommendations and incentives, we redid them to get back to this other huge market of authors. It's like a million times bigger than what people call the creator economy today. And like, they're basically defined by they're too busy living to bother building a mailing list in order to get their, their words heard. Right?

SPEAKER_03: Yeah, but this really depends on medium being able to surface inside of its ecosystem to both non subscribers and current subscribers. By the way, I just re signed up my credit card had expired. So

SPEAKER_16: plus one to you. How do you ensure that you're highlighting content that is written by someone who knows something and is doing quite and has the right knowledge base and ability to convey that information? Because to me, I can do that as a reader on my own, because I've been online for too long, but I can't imagine having that done at scale and getting it right. So that's the thing that I,

SPEAKER_03: that, that almost feels like a secret sauce at medium that I don't fully understand.

SPEAKER_49: Yeah. I mean, that's the, the value of human curation and making that one of the primary

SPEAKER_01: signals and in an algorithm, what we've done is put in, and this is like in the weeds, but I'm proud of it. So I'm going to share it in place a two layer curation system. The first layer are subject matter experts throughout the community. And we've elevated our publication editors who are essentially the community leaders think moderators of subreddits. Like they play that role in the medium ecosystem. And at minimum, they're experts in what's been published already on medium. Like they see, they see everything within their topic. They have so much context and most of them have their own personal life where they're also experts in that subject. So they're out there knowing their area of expertise and saying, Hey, you know what? This is really important. Or this was a best practice 10 years ago, but it's not anymore. It doesn't need to be read. So they're providing that subject matter expertise, but then we're an open platform. You can't just let everyone do what they want because it'll get gamed. I mean, this is the reality of platforms. It's like, they'll always be gamed. So the second layer is the internal group, which is less deep in subject matter expertise, but really, really deep on

SPEAKER_00: ways to uphold a standard of quality. And so those two work in concert so that we're able to reliably, not perfectly, you can never do perfect curation, but you can do it a lot better than an algorithm could. I mean, that's what we found. It's like night and day. If you look at what has been curated

SPEAKER_01: through our system called the boost for kind of obvious reasons that we boost it to more people if it meets the standard, it's just night and day, the substance and the expertise and the knowledge

SPEAKER_64: that's behind what is getting boosted to a medium reader right now. You know, I'm literally reading

SPEAKER_67: this ad read from notion. I've talked about this product thousands of times because I'm addicted to it. I run my personal life. I run my family office. I run launch. I run inside. I run everything on notion. And we're now using notions, incredibly powerful project management pools. These are incredible. I didn't even know they had project management built into notion. I was going to buy another product. Instead, we're using it right now for our sales pipeline, AV checklist for the podcast schedules, events, everything. And guess what notion is about to make project management even better today. I'm excited to share that they've launched officially notion projects, which includes new powerful ways to manage projects and leverage the power of their built in AI features too. Notion projects combines project management with your docs knowledge base and of course AI. So you can stop jumping between tools and stop paying that much for them too. So here is your call to action. Do your most effective work with notion projects. You can try it for free today at notion.com twist. That's all lowercase letters. Notion.com slash TWIST. When you use our link, you're supporting our show.

SPEAKER_16: Go right now to notion.com slash twist. So I want to try to stack this up for everybody. So people arrive at medium, they can write, and then you have human curators from the user side that are very deep on individual topics. And then they notice things that are good. And then there's a layer of medium itself, more humans who are adept at not subject matter, but platform, curation, moderation, whatever. And then you can choose what things to boost bringing those two people who pay. And therefore what kind of bubbles up to them is, uh, both subject matter verified and then also medium quality verified at the same time.

SPEAKER_30: Yeah. Yeah. The head of our curation team just says it so simply. It's like, she just says,

SPEAKER_01: my job is to find the best stuff for you. Right. And it's a super complicated system, but the reader doesn't need to know that. The reader just needs to know that something better was put in front of them. Just think about the service of like how big and messy the internet is. And even just within medium,

SPEAKER_00: like let's say there's 200,000 stories come in a month and we're picking the 3000 best of them and separating them into the topics that you care about and then playing matchmaker, you know, primarily on, on those, that's a huge service to surface you, the things that you wouldn't necessarily find on yourself. And then it has this reverse effect because the two sided marketplaces, the reverse effect is now there's a stronger incentive for a certain type of writing, right? These authors that had been left in the cold by the growth hacker world nature, the internet. Well, now they have a place to come. We're getting more of that. And I think literally stuff is being written that wouldn't have been

SPEAKER_80: published on the internet at all because the authors didn't think they would have, have an audience.

SPEAKER_03: Yes. And then powering this inside the veins of the system that we just outlined is money. People

SPEAKER_16: pay into the system. And then that money is divided up between medium, the company, and then also quite a lot of author payouts. How hard was it to figure out the cascade? Because the system works, the use that set up, but it works because there's an incentive, uh, lines throughout it. And I'm just curious how hard it was balanced to get the money flowing the right way to the right people and also

SPEAKER_03: preserving enough for medium, the corporation. Right. I mean, essentially the money goes to

SPEAKER_01: three people. It goes to the authors, authors, and editors, to the community, to the medium staff, and to Amazon AWS. So a lot of the rebalancing was pulling money away from Amazon so that us and the

SPEAKER_19: authors could share it. As far as getting that balance right, it's just like, we want to, like, we want to reward authors with the things that matter to them. And I think this is where, uh,

SPEAKER_01: medium maybe went wrong in the past is they just didn't realize how many great writers would care more about distribution than money. So we really got lost in this idea that we can't succeed unless we're giving everyone a living. So this might even like shock you to say, like, I don't think we're in the business of like necessarily helping people make a living as a writer. Like we're in the business, we pay, we think it's fair to pay, ends up working at, you know, as an honorarium or part of someone's earnings. But honestly, the people that make the most money on medium are people that, uh, are here

SPEAKER_88: because they want an audience, right? Like, like, yeah, it's, it's the practitioner point. I mean,

SPEAKER_16: people who do something already have an income stream. I mean, let's be honest, if it was super easy to spin up your own publication on medium or sub stack and make a living quote, quote, I would be in my Bentley from my sub stack and I would be cruising them at the top down versus, you know, grinding out the podcast live here at twice. I joke, I love this, but like, you know, it's incredibly hard.

SPEAKER_03: And I don't think if you wanted to replace people's incomes, you would succeed because I think it would

SPEAKER_33: require what prices that were 10 times higher than medium currently charges to have enough money in

SPEAKER_78: the system. It's not even that it's like, it's just this misunderstanding of where the opportunity

SPEAKER_01: is. I think the creator economy is essentially a local maxima that's sitting next to this forgotten expert economy. Like you could, if we use you as an example, someone that's further ahead on a similar career path to you is MG Ziegler, right? Who's one of our investors. How much money do you think he made

SPEAKER_34: as an editor versus how much money do you think he made as a venture capitalist? Right? Like,

SPEAKER_99: Oh, I hope I know the answer to that. Because I used to work with Tucker and Dwight used to work. If he was making more money at TC than he was as a VC, I'm pissed.

SPEAKER_19: Right. Exactly right. So it's like, look, I mean, this is just a fact of life. This is not medium trying to like twist anything for you. It's like a fact of life. Writing is this incredible opportunity to turn yourself into an expert and then advertise it, right? And so if you're into writing for money, like you got the creator economy is this tiny molehill of money, and then you have the expert economy, which is billions of dollars, right? And I think that people forgot that. And so I

SPEAKER_01: just always try to like people that are saying, Oh, I want to make a living just writing every day. Like, of course, I want to make a living writing every day. But if you just actually look at the reality of the market, I would like always like remind people that, you know what, there's this other

SPEAKER_16: thing, other way to approach it, right? Absolutely. And also amongst people who do want to make a living from writing and pursue that hardcore, the number who make it is like seven. Yeah. I mean, I mean, the joke is that every every book author that you know, has a spouse that has a real job

SPEAKER_03: that subsidizes their creative output. And I think that's totally fine. But it's not going to be a system that medium can single handedly fix, especially if you're going after the expert focuses, I wish that there was 1000 times as much money to writing game, you know, but we're not there yet,

SPEAKER_16: maybe we will be in time. I do want to go back though, to the 2000 posts per month 3000 that are the best. And I'm just curious how those numbers are changing as time goes on. Are more people showing up to medium now that the model is showing lots of positive momentum? And how are you guys balancing

SPEAKER_03: getting people in the door versus making sure that the quality level, the bar stays where you want it

SPEAKER_01: to be? For us, we just want more breadth, right? And I'm like, I'm shocked by what we have. I was just thinking about this today, where it's like, we were looking at a paleontology publication that's really popular among paleontologists. And I think, oh, that's good, right? If we're winning paleontologists, we've gotten pretty, pretty broad across the available subjects, but we just always more like, I'm always want to, I want to know when I get interested in a subject, it's definitely going to be available on medium. There's always room for more of that. Just the other day, we opened up our partner program, which is how people get paid to 77 more countries. And I'm someone that just tends to believe in the value of geographic diversity, or it's like our, for whatever reason, there's a backlash against the word diversity these days. But if you would just take it down to lower the diversity, our mission is to deepen people's understanding of the world, like, you know, essentially to learn. And I think by definition, you can't learn anything from someone that's the same as you. You can only learn something from someone who's different, like by definition. So the more that

SPEAKER_22: we get different people into the platform, the more effective it is as a place to understand the world

SPEAKER_67: better. Hey, when founders ask me, what corporate cards should they use? I automatically think of Brex. That's what I use. And when founders ask me where they should store their cash, well, of course, I'm going to think about Brex. Why do I think about Brex so much? Because knowing everything about your money is not negotiable. Nearly 40% of startups that fail, well, they fail because they run out of cash. And 100% of startups can't afford to make financial mistakes. We all know that. So you want to be on top of this stuff. And thankfully, Brex has built a killer banking experience that helps you take your dollars further. And they'll protect your cash while extending your startup's runway. How do they do all that? Well, they offer the best parts of what you need, checking, treasury, FDIC insurance. And they do this all in a single powerhouse account. If you need the best, because you are onto something big, a Brex's banking solution is going to get you 20 times the standard FDIC protection through program banks and industry leading yield on your cash. And the ability to send money worldwide at lightning speed, which given how global everything is now is super important. So Brex is used by one in every three US startups, as you know, because you go out to dinner with other founders and we talk about it. So go get the financial stack that founders trust at Brex.com slash twist 24 Brex, B-R-E-X.com slash TWIST24.

SPEAKER_45: So 77 more countries, that means a lot more languages. And you know, when I think of medium,

SPEAKER_03: it's definitely an English first product. And that's the medium that I see. Is there a side of medium that is much more multilingual than I've experienced that you guys are talking into?

SPEAKER_33: Or is it still going to be kind of an English focus across that new English focus? I think

SPEAKER_118: this expansion to other countries is primarily for the English speakers in those countries,

SPEAKER_01: which, I mean, as an English reader, this is what I want. I want to hear from someone in Nigeria,

SPEAKER_16: what's, you know, their perspective, right? I want to move us gently towards AI by bringing up AI

SPEAKER_03: translation. You know, one thing you can do in the internet today is just use pretty fricking good machine translation to bring something to a different language. I read a lot of Chinese Communist Party output that's in Mandarin, and I read it in English. Well, I mean, I try to keep track of what the CCP is doing from a policy level. And yeah, Tony, I'm a huge dweeb. You know that.

SPEAKER_118: Are you shocked? I'm just debating whether to say anything here. Yes, go on.

SPEAKER_126: No, no, no. Feel free. If you don't roast the host, are you even trying? But, you know,

SPEAKER_27: if you have the ability for people to really make money in more places, and you have excellent machine translation, how important is it to actually get something in a native language versus

SPEAKER_129: something that's brought from a different one? I don't have an answer to this. I'm just kind of

SPEAKER_01: curious how you think about it. Yeah. I think it'll be some definite boon over time, right? That every language has these mega-hits that I think unfortunately never get read outside of

SPEAKER_00: their language. And I mean, my personal experience that I think a lot of people have is the first time

SPEAKER_01: I read The Three-Body Problem. And the thing that stood out to me was how someone who grew up in China feels about the Mao... Oh, the Cultural Revolution. The Cultural Revolution, exactly. How they feel about it from their perspective. I know about it as a historical fact, but in the book, you really get a sense about how disgusted they are with that, right? So disgusted that they would sell the whole planet out to another... I mean, this is sci-fi, but to another alien species, right? That was very visceral for me the first time I read it. And that was... I was reading a translation. I felt really grateful for that perspective. But I think we'll get those for sure. I think that's part of making... pulling as many stories out of people as possible and letting the hottest ones travel as far as possible.

SPEAKER_03: I mean, could you imagine a world in which you build like a mini medium, like one for India, one for country X that had a specific linguistic focus and maybe a separate pool of money to see if you could replicate the same model in a non-English focused area?

SPEAKER_01: Yeah, 100%. But you're giving me heartburn. We just barely saved the company a couple of days ago. So

SPEAKER_49: this idea... Now that we've done that, we can do a lot more.

SPEAKER_07: I mean, look, this is Twist, not the retirement business show. This is all about what is next.

SPEAKER_03: And honestly, this... Okay, let's talk about AI, then business. I have a lot of things I want to get into the business side, but I want to stick to some AI stuff because I know a lot of people tuned

SPEAKER_16: in. That's what they want to hear from us. So you mentioned earlier on AI influx, you're seeing lots more content. I think you said that last year it was like a million spam posts a month, and then it was like 10 million this year. Just two questions. Is that increase still

SPEAKER_27: happening? And then are there the AI spam influx? Is that becoming more sophisticated?

SPEAKER_01: I mean, they're not becoming more sophisticated, which is good. We upped the cost of it a bit. I mean, there's a lot more... Essentially, you're in a war with a spammer. You can't block them completely, but you can make the cost of participation for them higher. But it's hard for me to judge whether they're more active day to day because we've counterbalanced it by going to war with them,

SPEAKER_80: which is what you do with spammers. And essentially AI just represents a new, much higher volume

SPEAKER_03: category of spam. Yeah. And there's two ways that I think about AI as a bugaboo for the world of publishing. One is an influx of spam submissions, spam comments, et cetera. And then the other side is

SPEAKER_16: AI crawling bots. We all know how search engines index the internet. There was always a pretty reasonable value transfer. They would crawl your site. You'd pay for the crawling. And then they would link to you. They drew value. You drew a value. Everyone was happy-ish, I think. Now, there is, of course, AI data companies, model companies that are going to hoover up the whole internet, including in real time. And back in September of 2023, you guys blocked, I think it was OpenAI's GPT bot. We did. One reason why I wanted to have you on was there's been so much conversation about perplexity AI,

SPEAKER_03: anthropics crawling systems. And so I'm curious from the medium side, you guys have so many words that these people might want to take. How hard is it to compete with sneaky AI crawlers that want to essentially yoink the data and then ingest it? Yeah. I love this topic. And this isn't really

SPEAKER_19: even just a medium topic. This is an industry topic. For sure. For sure. I mean, for the 20 years I've

SPEAKER_00: been working on the internet, there's always been this assumed exchange of value, right? This is how society functions. And then OpenAI came out and it was really obvious that they had just rejected that as a concept. Then you can almost predict everything that will happen after that, right? Because if there's no exchange of value, why would we participate in what they're trying to do? We wouldn't. And so it took like a week to decide, oh, we're just going to block them. Because we're a platform and we represent a community, we had to talk to them about it. We can't just make a decision like that and not explain it. And so the way we explained it is for an AI company to train on their

SPEAKER_19: writing, you would expect that they would offer, um, they'd ask for consent and then they would offer

SPEAKER_00: credit or compensation, the three C's. And so that's like the most direct way that OpenAI failed. They didn't ask for consent. They didn't link back. You know, there's no credit. Like you think of Google trains on your, on your data, but then sends you a lot of traffic and gives you credit for it, which the traffic is its own form of compensation. And once our authors realized that they blew their top. And I wondered a little bit as I was explaining it to them as like, should I go try to arm wrestle these companies for money? And so I explained to them, it's like, I think probably if we were

SPEAKER_49: successful, we would get this much money and we could pass it on to you. Is that enough? And the

SPEAKER_16: answer was like, unanimously, no, because they were pause and, and tell me as much as you can about the number you told your community, how much did you think at that time that you might be able to get

SPEAKER_00: from open AI or a similar company for the size of medium? We thought we could wrangle $5 million,

SPEAKER_01: per year or a one-time per year. And we're probably the only platform because of how we're set up.

SPEAKER_00: We're the only platform that could just pass that money directly to the authors. Like Reddit did these deals and just pocketed the money because they have no system for giving the money out, right? Meanwhile, we have a system, right? Yeah. And so that I explained it to the community, like I explained it to the community in the most, this will benefit you way possible. And I said, you know, we think we're negotiating as a service here,

SPEAKER_01: right? We will go have these conversations, take the lawyer fee out of whatever we negotiate,

SPEAKER_00: and the rest can go to you. And they were just so offended by how it had gone down that they said no to that amount of money, which was like a double-digit increase, a double-digit percentage increase in the payments they would have gotten. The real number, by the way, from open AI was about 1 million. That was the opener. But by that time, I knew that our community was just too angry. It's like, what am I going to do? Negotiate a million up to 2 million? I don't care. We just flocked away.

SPEAKER_156: If you use multiple devices and apps to run your business, you need OpenPhone. OpenPhone

Chamath Palihapitiya: simplifies your communications with one simple app. And OpenPhone has rethought what a modern business phone can be. What's magical about OpenPhone is that it works through a simple, elegant app right on your existing phone, or you can even use it on your desktop. I know, because we use it here at launch and our sales team loves it. And you know what? Those phone numbers, those discussions, all those text messages, we need those to be on launch phone numbers. We don't want those on people's personal phones, where if they leave the company, it's distracting. It's annoying to your sales team, as but one example. And we love having a shared number for customer support as well. We do a little round robin where different people can pick up the one phone number. So customers have one number, but we can have that phone call go to multiple people on the team. So we, you know, can just have somebody pick up the phone quicker. It's that simple. Answering the phone quickly is a best practice and it makes people love your product or service. OpenPhone is already super affordable at just $13 a month. My Lord, that's affordable. But Twist listeners can get an extra 20% off for any plan. The first six months, that's incredibly generous. Just head over to openphone.com slash twist. And what if you have an existing number with another service that you hate? Well, open phone, we'll port them over at no extra cost. Easy peasy, lemon squeezy. So head over to openphone.com slash twist and get a free trial and get 20% off.

SPEAKER_16: So Tony, I just want to make sure that I'm, that I'm tracking this correctly. So

SPEAKER_03: currently medium has more than a million subscribers and we'll get to that in a second. Very basic math, about five bucks a month to be a media subscriber. I think I got four bucks because there's a discount

SPEAKER_16: going on. Thank you. That's 5 million a month. So OpenAI was offering 20% of one month's recurring

SPEAKER_167: revenue for access to all of your data per year to ingest it and use it to make more money. Yeah. What were they thinking? Like, did they think you were stupid?

SPEAKER_00: I mean, you know, from Reddit's perspective, it's free money, right? Like they got 60 million, they got less, let's be honest. They got money and services. So the headline of that announcement

SPEAKER_19: was wrong. The headline was they got $60 million, probably let's say 30 million on a company that's making 700 million, but it's just free money at that point. That's all profit, right? I think if we were pocketing it, like, look, if I had an extra million dollars in my budget,

SPEAKER_00: like I'd know what to do with it, right? Like I wouldn't say no to it. It just, because of how we're set up, I needed the, you know, the community needs to go along with it. And I agree with them. It's like, it was offensive. And so if just, if we're talking about the industry, this is what I predicted when we blocked open AI, I predicted that they were creating a competitive situation where the platforms were going to end up locking them out from future training data. And that's exactly what happened. Because if you look at how Google has approached this,

SPEAKER_19: Gemini does give credit. Yeah, right. But for us, we want serious readers. Actually, like the people that just want the answer

SPEAKER_01: that Gemini got, that's not a future subscriber for Medium, right? The person that wants to read the whole thing, I'd actually probably save a server cost if Google could filter out the search traffic

SPEAKER_80: to just people that wanted to read to the end, right?

SPEAKER_19: Yeah. But they launched with that, they launched with a consent, and they have been willing to pay.

SPEAKER_01: They haven't paid us. But I think that's on the table. And so from our perspective, knowing how our community feels, knowing how open AI behaved out of the gate, I think it's incredibly likely that we would do an exclusive deal with Google. And I've seen other

SPEAKER_00: people start doing that. And so the future of AI training, essentially, open AI just ends up getting

SPEAKER_22: locked out because they behaved so poorly in the beginning.

SPEAKER_03: Do you think that was, I know we're doing kind of corporate psychoanalysis here, but does that just,

SPEAKER_61: was it arrogance? Or was it just them not valuing the written word as much as you and I do? I'm trying to figure out if there's bad intent here or not.

SPEAKER_22: Uh, it's cultural, uh, cultural arrogance. I mean, it's blitzscaling, right? It's like,

SPEAKER_00: you see the playbook and they're running, they were running a blitzscaling playbook, right? It's the, you know, they're kind of, the real damage is being done to the creators of the internet.

SPEAKER_01: Meanwhile, they're at Congress saying, Hey, you got to regulate us because our stuff is so powerful. It could become, you know, Be careful! I'll just say, I thought that was the naked marketing stunt, right? Like, what if I went

SPEAKER_00: in Congress and said, listen, you guys need to take a hard look at medium, because if you don't, we're going to help people get so smart. You know, I don't know, like, it's just, please, right?

SPEAKER_187: Um, pipe down with your digital Jesus. We're doing fine. Yeah. So open AI doesn't offer you much.

SPEAKER_16: Google seems to be an attractive possibility. Have you guys actually picked or signed an AI deal? And then,

SPEAKER_27: if so, is it going to be exclusive or is that still a negotiating point?

SPEAKER_00: Uh, I expect that we'll at least get into that conversation, but I'm not far enough along to know how it'll go. I'm just saying, like, as I watch this develop and as I talk to my peers,

SPEAKER_80: that's where it looks like it's heading to me.

SPEAKER_16: Yeah. And then, doesn't that create a world though, in which, like, like the, okay, going back, we've seen a lot of AI models improving quality lately. Putting inside the whole AI training data issue,

SPEAKER_03: it seems that there's four or five models today that are GPT-4 equivalent or better. That's because I think they've mostly had access to the same data, but it seems that in the future you're describing, we are going to see certain models have access to different underlying data

SPEAKER_16: sets that will then yield different levels of quality. So to me, if you picked one, let's just say you picked Anthropic to throw a name out and not Google and not open AI and so forth, doesn't that eventually make you more valuable on the market because they want to have similar access? So then

SPEAKER_03: exclusivity seems to be very expensive from medium because then it precludes other agreements, at least as long as that term is in place.

SPEAKER_78: I just think, I mean, we have to make a business decision in partnership with the community,

SPEAKER_00: but I just think it's bad for the internet, right? And like the dominoes were put in place by other people and that's, you know, here's how it's falling right now. What I had tried to do was get a cohort together of peers. And what happened instead is basically everyone else preferred to negotiate on their own. And I think you're starting to see people that I talked to make their announcements. Like it tried to get a, you know, Reddit in, no. And you saw that it's because they thought if they could do, this is what looked like me, if they could do their own exclusive deal, it would help their IPO.

SPEAKER_197: Right. And it did. That's true, but it's short-sighted in my view.

SPEAKER_01: Right. I think bad for the internet and staffed overflow, uh, has their own AI model, right? So they are opting out for their own business reason there. Wikipedia has different concerns than me. They, I think probably feel like they have to participate. Otherwise, AIs will be less smart and they have a, not a financial goal. Right. A humanitarian goal, if you will.

SPEAKER_00: Right. You know, we couldn't get a cohort together and say, hey, we just need to make

SPEAKER_01: an industry standard here and we need to, um, you know, negotiate as a group. There might, I think there's probably some legal logistical fears about antitrust in there too about us, uh, working as a one cohort, but I just like, I think we could have solved them. I think that would

SPEAKER_00: have been better for the internet. You know, I do like, I think in these training data, this medium is like the 40th, biggest training set. So we don't have any leverage. All we had

SPEAKER_88: is sort of like and like early desire to speak on it.

SPEAKER_03: I think as you grow and as the medium model that we discussed at the top of the show continues to turn and drives value and grows and so forth, then I think you'll become increasingly valuable

SPEAKER_16: and you should have more leverage down the road, which will allow you to do the thing that I care about the most, which is drive more monetary value to both art and the written word in more particular,

SPEAKER_03: because I think that's, that's such a critical and important thing. Now business, we have a couple of things and I am very excited about this. So first of all, you crossed the million subscriber milestone a couple of months back. This is a rude question, but we're here, so let's do it. You haven't slipped back underneath the million subscribers.

SPEAKER_203: Bill above. Bill above. Bill above.

SPEAKER_03: And how has growth been since then? The economy has been a little iffy, media has been a little

SPEAKER_16: wonky, but it's also an election year. So I'm kind of curious how that all adjusts out for mediums.

SPEAKER_00: Yeah. I mean, we're good. I think we probably won't make an, another announcement till we hit a milestone that feels like announcing, you know? And so we're like right now, I guess at some level,

SPEAKER_01: we care where we rank in the media subscriptions of the world. The New York Times is tops above 10 million and, and, uh, probably in the top five, top seven there. And, um, yeah, so I'm dodging your question. That's what I said, what I'm getting at.

SPEAKER_16: No, no, you're, you're, you're dodging it gently. So I'm going to, I'm going to push on it and then

SPEAKER_03: you can dodge one more time. In your post announcing the million subscriber milestone, you talked about how, like, should we even make, make a fuss about it? It's just a number, blah, blah, blah. But then you said people internally were tracking it and it was a good, I think it's good to have goals. I don't think goals have to be anything other than arbitrary, but I mean, like, so when you hit,

SPEAKER_211: is it 2 million that you'll talk about it again, or is it like 1.5?

SPEAKER_136: This is the big goal, but I want to say that about the goals is actually interesting. You know,

SPEAKER_00: having worked in startups for a long time, there's this pendulum swing between people that are mission driven and people that are pretty mercenary. And it was just this great moment for this team, which tends to be fairly mission driven. Yeah. To be given a mercenary goal. I mean, it was essentially like, I had two mandates when I took over as CEO of Medium two years ago, make it profitable and make the quality something that's not embarrassing to the founder, those two things. So first mandate is incredibly practical. And we did both now, but also the quality has changed drastically. But, but you know, it's like, we're just

SPEAKER_01: every day, we're talking about how do we make money, a bunch of mission driven people. And so I think maybe we were like resisting, celebrating the financial milestone, that's not in our nature. But then when it actually hit, everyone just felt in their bodies, how hard it had been to get there. Right? Yeah. Like I was on a plane coming out for meetings and flying from New York to San Francisco. And I'm like,

SPEAKER_00: so grateful that the person next to me was napping, that I was getting all emotional when it hit. And I, you know, I thought I was going to be working on something completely different. And that post, that announcement post is literally written on a whim on that plane as a reaction to how I felt and to how the whole company felt like, like, this is what I like about startups is they might not pay as much as a corporate job, but it gives you a job that you can care about. Right. And, uh, that was,

SPEAKER_01: it was like legit, a good moment for us.

SPEAKER_217: Well, yeah, well, it made me very encouraged. So I know that I I'm, I'm not a medium employee

SPEAKER_03: shareholder. I'm a user now a subscriber, but when you guys hit that milestone to me, it was a signal from the market that even in today's digital era, there are a million people who are willing to pay five bucks a month, or you have a higher paid tier as well for people who want to be more involved in Spotify. Take note. Thank you. Uh, and that meant that, that my team, my, my people, my, the people that I would love to be friends with and sit down with and have coffee with were, were myriad. And then it wasn't just me and six other people who are, who are big, you know,

SPEAKER_16: writing and reading tweets. There's a lot of us. And I, I, to me, it was also an emotional moment in a different way and different level, but I, I, I absolutely hear you on that, but you did mention the

SPEAKER_01: P. Oh, Tony. Well, it's a proof point. I think that's different P than you were about to say, but it's a proof point. If I come on here saying the internet's broken because of ads,

SPEAKER_00: well, who cares unless there's an alternative, right? And so we want to demonstrate like other business models are feasible, like, and we don't think medium is limited to a million subscribers, because as you say, if it works in English, it'll work in other languages too. That's an easy way to explain that there are multiples of a million subscribers out, out in the world. Absolutely. That's a venture scale business. That's not just a healthy lifestyle business. That's possible to do a venture scale business that's not based on ads. And so, yeah, I think it's really important to put that proof point out for other entrepreneurs that are thinking, I don't want to participate in a broken

SPEAKER_07: system. On the venture scale point, I'll just pull in some data for folks who are curious. According to

SPEAKER_03: Crunchbase, Medium has raised date 163 million. Notable investors include Greylock, Andreessen Horowitz, Spark Capital, and Obvious Ventress, which of course is affiliated with the lovely Ed Williams.

SPEAKER_16: So if you're curious what we're talking about here, this is not just between two bald nerds instead of ferns. This is a very- I can't see how bald we are. Oh no, this is going on. People are watching on YouTube. Like this is going to be on. Yeah. So it's, I took my hat off so we could match our hairlines. Yeah. But the profit point,

SPEAKER_03: I want to get back to that because this is to me, the, the follow up, the corollary to reaching a million subscribers. And the fact that you said back in May on threads that quote, medium was month over month cash balance positive, but now you're saying that the company is actually profitable. So is that the same milestone or is there a distinction between the two

SPEAKER_00: accomplishments? This ended up being a funny like milestone to get the company to celebrate because in order to celebrate it, you have to understand accounting, right? Like, are we doing cashflow accounting or are we doing accrual, right? And if you've run a business, you know that it

SPEAKER_01: makes more sense to do accrual, which means like, if I sell an annual subscription, I don't book, I don't say, oh, I made $50 this month. I say I made, cause I have to do a service, you know, for the next 12 months. So I made, you know, one 12th of it and then one 12th the next month and one 12th

SPEAKER_00: the next month. And same with a lot of the bills, you just end up kind of spreading them over the year, ends up being a more rational way to look at a business, but it's not how any of us handle our checking account. We do cashflow. Like, do I have money to pay the bill or not, right? It's just two

SPEAKER_01: different views, two very valid ways to do accounting. And like, we do both because we don't

SPEAKER_00: want to run out of money, which is a cashflow question. And yeah, we literally did have that month in May. First time ever in the history of medium where the bank account went up without investor money coming in, right? And it probably like without literally a check from Ev who participated in every

SPEAKER_01: round, who's our founder and founder of Twitter and founder blogger, who participated in every round, right? I mean, that is a milestone, but the more, the more standard milestone is the accrual accounting.

SPEAKER_00: That's a standard way of doing accounting. And so we are in August projecting, and this is not a controversial projection, like we're on track to be in our first ever profitable month. I feel good about it. Yes. Thank you for the...

SPEAKER_16: You should feel good about it. Also, it's funny because we were talking earlier about being counter narrative, you know, focusing on subscriptions, not ads and quality overreach and kind of contra AI in some ways. But I think getting profitable as a venture backed tech company in 2024 actually isn't super controversial. It's actually relatively popular as a way just to kind of defend the existence of your company. So here, in a nice way, you're actually kind of on track with some themes

SPEAKER_03: we're seeing in the market. But I'm curious what it does to your mindset as the CEO. Because in my notes here, I literally wrote a question that just says, is medium eternal now? Because to me, once you're cash flow positive, and then once you're profitable on an accrual basis to use your words, uh, the business has reached a level of stability that almost unlocks freedom to experiment. And so I'm just curious, you know, like now when you're sitting back having a scotch or a soda or whatever

SPEAKER_56: it is, and you're thinking about what the next six, 12, 18 months are like, do you have more, I don't know, like, like chutzpah, pizzazz, because the company is now like stable.

SPEAKER_01: Yeah, it's, it's, um, I wish I could talk about this more. Actually, it's like, we did, you know, my job is essentially two turnarounds. And I did the first turnaround, but I could do the second one,

SPEAKER_00: right? The first turnaround is like what you picture a mercenary CEO, like come in, find the cost to cut, find a way to grow, et cetera, et cetera. But I was given this extra rope to do it with some bigger future in mind. So like if private equity had taken medium over, they would have cut the team in half. They would have plastered ads everywhere and it would have been profitable sooner, but it would have had no future. Right. And so I tried to do this turnaround, which I guess, I think we can say now that we're profitable bid, check, done to set medium up to be much bigger. So like my Northstar is not a million subscribers or even 10 million subscribers. That's the business goal. My Northstar

SPEAKER_01: is the impact that collecting every person's story on the planet. What's the impact of that on knowledge, right? To me, that was the original promise of blogging, is to pull out people's

SPEAKER_00: stories. And because that's where knowledge and wisdom is hiding and you normally never have access to it. So if we can collect that into, call it like one great library of human stories, you can have the same level of impact on knowledge that Wikipedia has on facts, right?

SPEAKER_240: Well, I feel like wisdom is the better word there versus, because knowledge to me implies

SPEAKER_16: facts. They're almost synonymous, harmonious, but wisdom is the application of facts to life.

SPEAKER_210: Yes. And that, that, this is why we're not particularly threatened by AI. You know,

SPEAKER_01: like so much of living is requires the human wisdom, right? How to make coffee. Sure. That's a set of facts, right? But if you actually love coffee, like if you talk to a coffee aficionado, there's all of this extra information and all this why behind it. And, uh, uh, I, look, I love that, right? I love that.

SPEAKER_00: And I especially, I love how good writing is, how good the written word is for collecting someone's thoughts and sharing it out to other people.

SPEAKER_16: We're in heartfelt agreement about where you guys are going and I'm very excited about it, but

SPEAKER_03: you know, you said 10 million is not the kind of North star gold, but like, I'm just thinking about what you could do with 10X the revenue in terms of more people,

SPEAKER_16: more experts, more wisdom. So to me, like the, the, the result does scale with financial wherewithal.

SPEAKER_03: And I, I just, I just don't want to, I don't want to dismiss that. I think that writers broadly, people who do creative stuff, often back to your company's point about profitability, being a milestone to celebrate or not dismiss the, the, the blood of it, which is just money,

SPEAKER_27: you know? And so I, I almost want to say, dude, North star, that 10 million metric, go crush it and then come back on twist and tell me about how you did it because medium 10X would be awesome.

SPEAKER_00: I want to come back sooner than that. Like how long did it take the New York times to get the 10 million, but maybe we can do it.

SPEAKER_250: I said at the top of the show that you are a man of the moment, ergo, go forth and do the things, you know, quickly in a moment.

SPEAKER_210: You're getting at something that I, I wish people understood more as like the sort of the

SPEAKER_01: archetypes of how to, how to build and run a company. I think the, all, a lot of the ones that we've seen recently are based on a lot of venture capital being available, right? The sort of the zero interest rate phenomenon. I think every CEO is visionary at some level, but some would rather sell that vision to you and others would rather prove it to you. And when there's a lot of money, you get a lot of sales people running the show, right? And they tell a big picture and it's filled with hyperbole. And when there's a, not a lot of money, you get a lot of practical operators like me, who like, I mean, obviously this turnaround is like a victory for practical operations.

SPEAKER_00: And so we're just in this moment where being good at the details is a big part of being successful as a business right now. And that like, that's the, a lot of the medium turnaround is, as you say, like, go out and make money because if we make money, then we can, we have more, more fuel to do the, you know, the things that we want to do.

SPEAKER_03: Absolutely. And if you want to think about the connection between money being available and salesmanship, and then cross that with our earlier points about, you know, open AI going to Congress and demanding regulatory capture, what did they have behind them? Well, it was essentially unlimited money. And I think that there's a pretty direct link between that and what we just talked about. And I think it's good that the AI hype has slowed a little bit so that we can get past the salesmanship and more to the practical elements of, of building real companies and a little bit less on, I don't know how, if I read one more time that every industry will be revolutionized by open by AI, I'm just going to scream and then vomit into a hat. How exactly? Tell me how exactly. Okay. This is, this is off topic, but let's just do this because we're here. Um, day to day, your workflow, CEO of medium, how much AI are you Tony personally using?

SPEAKER_80: I don't touch any AI in a given week. I go back and, um, occasionally test like a certain

SPEAKER_01: summarization tool. That's the one that has like, has my wheels spinning. The first wave of AI is like

SPEAKER_210: too much hype for me. The second wave is going to start having product people that start putting

SPEAKER_01: more interesting ideas out there. And one that I've been sitting on is more hybrid products, right?

SPEAKER_00: So like, what always has my wheels turning is not, can AI summarize medium for people in a useful way? It's more, can AI create the first draft of let's say a topic summary? And can that be then expanded and corrected by humans? And so I'm always testing summarization tools because I'm curious about them. I guess now just being a Google user, like doing a Google search, you'll get, you know, you'll get there for some reason. So I see it that way. But like, I think the people that think AI is going to like revolutionize writing kind of don't actually understand what writing is. Writing is thinking,

SPEAKER_01: writing is thinking. And every time I've like tried to introduce an AI into my own thought process,

SPEAKER_02: holy, it's such a bad experience, right?

SPEAKER_16: All AI writing tastes like chicken. It's always like, Oh, great question. Here's four bullet points about why this is a thing. And I'm like, I mean, if this was an eighth grade essay, I'd fail you because you bored the tears out of me so quickly. And this is why I am personally not afraid of being replaced by, by AI right now, 10 years from now, we'll see where the technology goes. But I, but I agree with you. The summary tool should be cool. I don't use that either. I also don't use AI day to day. I have ChatGPT on my Mac and I occasionally will pull it up and ask it something for the sake of staying

SPEAKER_03: relevant. And I played with the new Grok image thingy, but it's all experimentation and tinkering versus this is a key plank in what I do. And I'm just kind of waiting for that to change. But until it does, to me, we're still in the first cycle of AI and we're not in the second one yet.

SPEAKER_78: Well, my board right now is filled with people that have super early stage companies,

SPEAKER_00: like all four of them just like started something at the same time. And two of them are interesting takes on what is now possible because of AI. Like now I'm a little bit more sensitive to what's happening, but it's not replace human creativity. It's something completely different. They took this in very different, different directions. So I think we're going to get good use cases, but what we're seeing right now is just not explaining it in any, anywhere close to where it's going to end up.

SPEAKER_217: I agree. I want to, if people are listening to this and they're saying, oh my gosh,

SPEAKER_16: here are these two writer guys who think AI is bad to the contrary. Like one thing that I'm most excited about is the, is the, the, the intersection point of increasingly low cost and high quality humanoid robots and LLMs as a way to chain instructions together to do stuff like that.

SPEAKER_03: I've literally sat up and I thinking about how cool that's going to be. So when it comes to AI, huge fan, it's just in my workflow and also in my first love, which is writing, it hasn't yet crept into my, my flow, but at the same time, I do have friends that use it, uh, you know, day in, day out. I'm just, I don't know. I, I'm not ready to get ready yet to, to export part of my brain to someone else

SPEAKER_138: or something else. I just, cause to me, then why am I here? What am I good for?

SPEAKER_01: I mean, this is a little controversial. This is maybe not the official medium talking point, but I sometimes wonder if some people we're hearing from don't like themselves.

SPEAKER_275: Uh, in terms of like the people who are raving about AI don't like themselves.

SPEAKER_78: Some of them. Yeah. Like they wish they were something different, not just someone different,

SPEAKER_01: but something different. And like, there's something about being a human that they are rejecting. And I mean, I don't know, maybe they're gonna end up being right, but my impulse is to think we'd all be better off, you know, this is coaching and therapy coming out

SPEAKER_00: here, but we'd all be the faster path would be to find some, some way to accept and embrace who we

SPEAKER_03: are. I have so much to say about that, but I'm not going to get into it. Cause we would stay here for another half an hour. I want to bring up one more point about, about medium and its feature, because when I was just pulling my notes together for this, it struck me that there is

SPEAKER_16: two things that medium pays for essentially. One is new stuff from great folks. The, the, the practitioners, the experts we talked about earlier, but then it also builds a catalog over time that gets larger and therefore more valuable and it probably cures or ages well in some way. So is there going to come a tipping point in like the medium model in which the back catalog of, of,

SPEAKER_03: of words is taking up like the majority of the available revenue bucket. And could that contrast or conflict with having enough money in the medium pool, if you will, to keep incentivizing increased creation? Like, is there a historical weight that eventually tips the scales in a weird way and makes the medium model a little bit wonkier than it is today?

SPEAKER_78: It's interesting to have it framed that way. Cause if I could just put that aside,

SPEAKER_00: I would love to get into innovative ways to organize, right? Like we've seen blogging platforms before we've seen publishing platforms, right? Like Tumblr, Twitter threads, right? Like all very TikTok, even very similar models, but none of them really took the organizational layer seriously. I mean, we have tags, but you could go so much further than that. And yeah, I think that's a huge opportunity to do that. I don't think, you know, at that point, I don't think there's pressure for all the money to flow that direction because at that point it's already written. And also the people that tend to write evergreen content are people that I think they care about the visibility more than they care about the money. But no, I don't think it's going to disrupt our business model to answer your question directly. But yes, super hyped for, you know, talking about things we can do in the future now that we're profitable, like sure. Medium in other languages, absolutely. If it works in English,

SPEAKER_01: it's going to work in every single language. But that's a thing that we've never seen on these UGC platforms is a real librarian sense of like, I'm looking for this topic and the answer is published. The best thing on it was published 17 years ago and here it is.

SPEAKER_03: Yeah. And then here are the four things that have come out since that change. You almost need to have, well, I wonder if your curators become essentially librarians in time and then maybe,

SPEAKER_281: maybe we do a decimal system medium.

SPEAKER_78: That a hundred percent. This is what I came in to medium thinking about. Like me,

SPEAKER_01: I was a publisher on medium and me and all my publisher friends were all thinking about our back catalogs and how to turn them into encyclopedias. So my very first day on the job, I reached out to the Wikipedia guy, no person, Wikipedia person who, um, uh, manages our page.

SPEAKER_19: They happen to live in San Francisco and, uh, I was in San Francisco, so it made it work and took them out to coffee. So like all of the things I'm doing to take over this company. I was like,

SPEAKER_00: I want to be your friend. I want to, you know, like have your input. Um, and now they work as one of our curators. Um, and they, I think have been curator of the year, uh, at Wikipedia. So like, I just a hundred percent that impulse is there and people don't understand how active the curation

SPEAKER_01: impulse is on medium. Like I think there's something like 3000 community publications and something like 9,000 active editors who are working with authors, you know? So it's like you,

SPEAKER_00: you would tend to think of medium as individual people's blogs, but it actually works a lot more like Reddit and subreddits in that way where that curation part of the community is a million times more active than people outside of medium understand. So, and I, you know, far be it for me to be your,

SPEAKER_99: um, wildcat product guy, but you know, Reddit is a place. We'll do it. You won't. No, no, no, no,

SPEAKER_135: no, no, no, no. I aren't. I'm having a second baby and I have too much work already, but you know,

SPEAKER_27: it strikes me that you already have the curators, you already have the writers, you already have the readers, you have the subscribers and they're already on the platform, interacting, writing,

SPEAKER_16: working together and having this, this version. Why not throw up some, uh, forums and, um, do a

SPEAKER_138: different form of UGC inside of the medium world and, um, pick Reddit's ass. Okay. Great. That's

SPEAKER_01: that's interesting. More and easier community. Definitely on the road. Okay. Because you guys are

SPEAKER_27: the ant antisocial media in some way. Like it's, you're the, you're the opposite of an endless scroll

SPEAKER_03: or a, a short form video. Cause it's words. You have to read them. And here's a lot of them. And that must have a different core audience than, than tech talk for short being the, the other end

SPEAKER_16: of the medium spectrum, but Reddit's somewhere in the middle and they're doing quite well and you

SPEAKER_00: should go take all their money. I'm actually like a big, a big fan of Reddit's potential. Like I would put money into it in a, if I was, I don't, I do mutual funds, but if I had individual, uh, uh, uh, stocks to pick, I would pick them. I think their long-term potential is great. People, uh, outside of social media, I'm not like picking on you here because I think you're just floating this as, as an idea. I think they really underestimate how the format like defines everything. Right. So it's even something like you, it's hard to port a newsletter to a blog. The voice is different. Right. It's just like, it's jarring to read a newsletter inside of your blog feed. It just is weird. Right. And those are things that are the same length, often very similar goals. Right. And, uh, we've played with short form and it's like short form and medium form are weird together. We bought a book company and we're like days away from letting you read books inside a medium. And then we didn't like it. And we put that whole project to rest. Right. So this, like, I think there's a, I have a ton of respect for Reddit. I, I think it it's amazing that it exists on the internet, such an important resource. I just think there's room for like, we're always, we're so different, um, in ways that are subtle that, uh, it'll like, no, we're not going to compete with that. We're, we're going to be

SPEAKER_03: friends with that. Well, I mean, so much for the, uh, eventually becoming the mercenary CEO. Does that mean as well though, that on, on the book point that a serialized fiction wouldn't sit well inside the medium model, because it would be a series of pieces of content that were directly linked in a sequential format versus discrete articles around one topic pillar.

SPEAKER_00: Yeah. It, it ends up feeling weird. There's something about the, like, it's jarring to switch,

SPEAKER_88: switch modes of reading, uh, serialized posts that never done well, um, on medium.

SPEAKER_45: Ah, cause it, cause, cause there's a world in which medium has cracked this code of, of, you know,

SPEAKER_03: collectivized payments on top of UGC for one type of written word. And then I see platforms like Royal

SPEAKER_16: Road that have cracked a different content side of things, but not the monetization element of it. And I just want to stick them all underneath your umbrella so the money can, can flow out. But it's, it sounds like, um, the way to profitability and the way to growth is by not doing everything and

SPEAKER_297: not trying to compete with both Reddit, Royal Road, and everyone else.

SPEAKER_01: To me, one of the biggest things that changed since when I got started in startups in like 2005, the internet's so much bigger, right? Like you can have a massive company inside of a tiny niche, right? It's like Grammarly is a huge company. Maybe it's affected by AI, but still it's like, this is spell check, right? And it's like, that's a huge company on its own, right? And, uh, so I just like, this is the thing that I often tell other entrepreneurs, like niches get huge now because they can be so big and medium isn't even a niche, you know, blogging only, but, you know, with aspiration of every language for every person in the world across every topic, that's,

SPEAKER_300: you know, that's already pretty broad. Yeah. We don't then have to do short form and long form on

SPEAKER_138: top of that. So last thing is just on, on, on mediums continued independence, because you mentioned

SPEAKER_16: private equity earlier, and I was going to bring this up anyways, but I think that's the greatest example of what I'm afraid of because you guys are now something that if someone purchases, there's a clear way to ring cash out of it, which means that you are now a target for a different form

SPEAKER_03: of potential buyout, AKA private equity, which to be clear, as a fan of your, of your company and your work, I really hope doesn't happen. So how happy is... This is one of the first questions I

SPEAKER_46: got, you know, by the way, it's like that someone was like, oh, they just hired this guy to clean

SPEAKER_78: the company up to sell it. And I was like, have you seen my resume? I'm the least qualified person

SPEAKER_45: to do that, right? Let's go through it. Better humans was a medium publication and coach.me is

SPEAKER_78: not a private equity firm, you idiots. Right. But it, I mean, especially like my qualification was knowing a ton about medium, but I've never sold a company, right? And so you don't

SPEAKER_00: pick me to be the CEO if the goal is to flip it, right? Worst possible person. And you, we would have flipped it a year ago, right? If we were trying to flip it and we would have taken a totally different

SPEAKER_78: approach. So you picked me to take a swing and you know, this is the controlling investor is someone

SPEAKER_00: that takes big swings, the controlling investor is still Evan Williams. And he's like, his career is defined by trying to do something like by the big wins, not the little ones.

SPEAKER_225: Yeah. You don't, you don't make a billion dollars by collecting quarters. You just don't, you know?

SPEAKER_310: You don't.

SPEAKER_56: Okay. And so it sounds like that Ev is happy with things.

SPEAKER_16: That means he is because you wouldn't talk smack if he was mad at you. I don't think that would be

SPEAKER_03: a weird dynamic, but I don't need to be worried about medium. Normally people come on and I harass them. Like, when are you going public blah, blah, blah, blah, blah. But I just want you guys to be

SPEAKER_01: okay. You know, because we're okay. Okay. We're okay. Now. I mean, it was, look, medium was failing

SPEAKER_00: when I took over, but I, I think I understood how to execute Ev's vision, you know, and that way we're complimentary. He saw something in the value of a subscription that I never would have seen. And, but when, as soon as he told me that what he saw, I understood immediately how to execute it. And so I've like gotten this opportunity. Um, this is just perfectly matched my, you know, my skills to try to execute his vision. And then over time, you know, more of my vision comes, but yeah, the two of us are in good shape and have a lot of, uh, respect for each other. And I think over time,

SPEAKER_80: we've understood, you know, where each of us has, has gaps, right? Like no one, no one has the

SPEAKER_25: complete package. And I certainly am not. Yeah. There's no full stack developer for life,

SPEAKER_16: if you will. No. All right. Well, Tony, I gotta let you go. As always an absolute treat. We'll have

SPEAKER_03: you back on before you reach 10 million subscribers. I'll make that commitment. Great. Thank you, Alex. We will ring one more financial data point out of you the next time you come on. So think about that in advance, but congrats on the profitability and, uh, surpassing the million subscriber milestone and, uh, putting medium back on the right track. I'm Alex. This is this week in startups. Of course,

SPEAKER_27: we are on every podcast platform. We're on YouTube, hit subscribe, hit the little bell thingy. I'm supposed to say, and we adore you all. We'll see you later. Bye. Bye. All right, everybody,

SPEAKER_69: we're going to do another jam with J Cal session right now. This is where I talk to founders about their company, about their startup, their product, their customers, their team. And, uh, we jam out. They tell me, they show me what they're doing. They tell me their challenges. I give them some

SPEAKER_66: advice based on having invested in 400 companies and then 2000 episodes visiting startups. Uh, this is brought to you by our friends at dot tech domain names. If you don't have a dot tech, what are you doing? I mean, there's so many great dot tech domains available in that namespace. Grab one now.

David Friedberg: And today I'm talking to the founder of core pod, uh, which you can check out at C-O-R-P-O-D dot T-E-C-H. Welcome to the pod. You will on deck. You'll on deck. Did I pronounce your name?

SPEAKER_328: Oh, you could call me. Oh, long. Like you. Oh, Lama. But and at the end.

David Friedberg: Okay. Yeah. Oh, long. How are you doing? Uh, why don't you show me your product? Tell me about it. And then, uh, tell me what your biggest challenge is since people are listening. You

SPEAKER_334: want to describe what they're going to see on screen. Corport, your personal sleep companion. We all have our smartphones nearby for me is a great tool to connect to my family and friends. Uh, keep up to date with all the latest news, but also is a huge source of anxiety and distraction. I always get messages, notifications. I do scrolling at night and I have insomnia. I wish I had a digital companion. You can, I can talk to, can voice my feelings and help me decompress before I go to sleep

SPEAKER_338: at night. Something like this. Hi, Milo. Woo. Woo. Hi there. How can I help you today?

SPEAKER_340: What's been bugging you lately? I'm having some trouble sleeping. Uh, I just can seem to fall asleep at night. Takes me more than two hours to fall asleep. Woo. Woo. Oh no. Sorry to hear that.

SPEAKER_347: That sounds really frustrating. Can you tell me more about. You remember beacon, the dog at the Olympics

SPEAKER_334: that helped us athletes, uh, with their mental health problems. So it's a kind of Gen AI dog right inside your speaker. We've talked to many therapists, uh, about mobile apps and they say the mobile app solution, it doesn't work. It should be a completely separate device. And we've made it as one speaker powered by Gen AI. And you can talk to it like if it was your friend. Uh, now let's talk about value. It may seem a quite expensive at first glance, but it's actually a bargain. Uh, one month is cheaper than one single hour of therapy session with a human expert. We can reach 10 million with just 17,000 customers with about 200 million with one in 70,000 users. Who are these users? People with high pressure jobs, uh, medium to high income and ready to invest in their mental health. Our team is passionate about building this product. And we also have advisors. Uh, we've made five prototypes recently. I finished founder university cohort eight. And last year I attended to team drapers hero training program. We got funding from a high technology park of Kyrgyz Republic and European bank for reconstruction development. And now we are rising. We want to connect to smart watches like Garmin to fetch all the health data, build product, ready product, uh, and start pre-orders. And this is my context. What is your biggest

SPEAKER_66: challenge? What are you, what's your challenge with this speaker with the screen that has an avatar or a virtual dog who is going to help you sleep better? What is the big challenge with core pod? Uh, you know,

SPEAKER_334: Jay Cal, uh, we are B2C and also it's, it's very hard to build B2C. And also we are hardware. It's even high, even two times, uh, harder to build this kind of soda. So my question is, what is, uh, um, what is the best go to market strategy for this sort of product? And especially that, uh, given the fact that we are

SPEAKER_357: related to mental health. Okay. So you're in the tinkering phase, you're building some hardware, you're

David Friedberg: building some software and your idea is you want people to sleep better. Interestingly, I have two investments in this space. Last night, I hit an 85 on my sleep score with my eight sleep. They're on their fourth generation and they study your sleep, your REM, your deep sleep, all of these issues. And it's quite nice.

SPEAKER_66: And it, it really has helped me, you know, sort of get into a better pattern. And I, uh, we were, we were investors in com.com and I used sleep stories with myself and my daughters to help go to bed. And so those two things have become good investments for us, great investments, in fact, and they do tackle sleep. Other people like whoop, uh, watch does it. So you are not only doing hardware and software and doing a startup, you're also going up against a pretty competitive set. So this is not going to be easy and doing hardware as the eight sleep team will tell you, or the cafe X team will tell you really, really increases the degree of

David Friedberg: difficulty. You have to be able to finance all of, uh, you know, this hardware. So I think what you have to do is maybe step back and ask yourself, are you making a virtual companion in a speaker format that's different than, you know, say, um, an app because you don't have to take your phone out and you have all the other distractions there. And is that the innovation here? Or do you want to, from first principle say, I want to help people get to bed at night, and this is the best way to do it. And I

SPEAKER_66: think that's what you have to ask yourself because clearly you're creative and you have technical ability and you've tried these different things, but which is it? Are you, are you passionate about hardware

SPEAKER_364: and creating this companion or are you passionate about sleep? Which one? Uh, no, I would say that

SPEAKER_334: the core product is the software. So it's personalized, uh, like your friend, which knows you and he knows how to help you sleep better. And the hardware is just the way you

SPEAKER_66: interact with this software. Yeah. You would think that you can use AI to be a sleep coach and that that sleep coach requires hardware. I wonder if your sleep coach requires the hardware. I would see if you could build a sleep coach with your iPhone that actually, uh, you know, lets you put your iPhone face down and interacts with you and put you to sleep, right? So I think a better model might be, Hey, eliminate the hardware business for now. You can always do that later, but see if you can make this first part, which is an interactive companion who I talked to and they put me to sleep, which was kind of like your dad or your mom telling you a story at night, putting you to bed and tucking you in. I kind of feel like maybe now that I've heard you respond to my question that you believe the interactive AI assistant is kind of like a sleep coach or a parent tucking you in. Am I correct?

David Friedberg: Am I correct? Yeah. Yeah, exactly. So I think you want to first prove that that actually works and you could do that through an app. You could get a hundred beta users and just see if the interactive model actually puts people to sleep better than listening to a sleep story on com because that's your competition for 50 bucks a year. Or, uh, if it puts you to sleep, uh, better than say, listening to an audio book or putting on a sound machine, I think what you're up against is, um, not the therapist who charges $500. I had that was in the presentation. Like, I think, um, like, um, that didn't ring true to me. I think what you're up against is the sleep stories on com and com is up against, you know, free sleep stories

SPEAKER_66: and sleep meditations on YouTube. So they have to be that much better. The packaging, you know, and the, uh, tracking and, and all that. And then the, and the beautiful interface, they have to be better than the free stuff. And then eight sleeps gotta be better than just using a sleep story to go to bed. And they're going to quantify itself. So you're gonna have to find that market in there. And I think prove that first piece and then hardware is so difficult. Um, but you seem good at it. So I would just wait on the hardware. I would just make that first piece work. Um, and how close to you are you to proving that out? How many nights, how many users have been put to sleep by your interactive person?

SPEAKER_334: Yeah, we have waiting list and up to 14, uh, thousand users. Okay. So nobody has been put to

SPEAKER_66: sleep by this yet, except for you. Yeah. Okay. So, you know, there's some demand cause you got the waitlist. What I would do is I would get 10 people, you know, and I would look for people in the subreddit on Reddit for insomnia. So there's a Reddit for insomnia. And I would go to the insomnia one, say, I'm a software developer. I'm trying to build software to help people fall asleep. And I think you can do it with an interactive person. And I'm looking for a couple of test users. Here's what I think is gonna work. What do y'all think? Would anybody want to try this DM me and see what happens

SPEAKER_375: if they allow you to post that? And if not, you post it on hacker news or other places or in a reply.

David Friedberg: Well, you could put it on Craigslist and just on Craigslist say, Hey, if you have insomnia, you're reading this at 1 AM, I'm building some software. I'm paying people a $10 gift card to try it for three nights. If you do it as a thank you and fill out a survey and tell me what you like and don't like about the software. And I would get to, you know, let's say a hundred people trying it and filling out the survey and see what you learn. Cause right now it feels like you're building, building, building something very complex without answering a first principle question. Can a language model, um, put somebody to bed like mom and dad did when they tucked them in

SPEAKER_347: and read them a story. Yeah, actually it's all based on the research. I don't know if that works,

SPEAKER_379: but, uh, some research, the proof that LLMs even text-based can really help in this regard.

David Friedberg: Great. So now take what you learned in that study, take what you know about, you know, hardware and tinkering and just prove it to the investment community and prove it to yourself and see what you learn. Maybe it turns out, you know, asking, you know, but three questions of the user puts them to bed. Maybe some people want to have, you know, a back and forth 20 times until they fall asleep. Who knows what the language model will come up with the technique, but I think you got to get closer to the customer and do a little more testing before you spend a bunch of money and raise a bunch of money. I don't think you're going to successfully raise money with a hardware company in 2024 going after this, because it's very competitive and hardware is really hard, but I think you could get into an accelerator or get some seed investors. If you did a small hundred person study yourself

SPEAKER_66: and then had that information and you could literally record it, you know, you could have all the interactions from the language model and from your little URL. You don't even have to send them to build a full app or the hardware to test this out. You just send them to a URL with a little script. They open the web browser browser, talk to it and see what happens. So I wish you great luck with it. And, um, it's a, a really great idea and clearly you're passionate about it. So good luck. Yeah. Thank you, Jake. Awesome. Thank you. All right, everybody. If you want to jam with J Cal, you can do that. We got one more slot left. It's brought to you by our friends at dot tech. What a great partner. Everybody's using dot tech, including me go to jam with J Cal dot tech to learn more and apply. We've seen so many creative ideas and I want to hear about your idea. Go to jam with J Cal dot tech. You just have to have a dot tech domain name. And there's a lot of great names left in that domain space. So get in there now. Maybe you want to buy a couple of them. So you have a couple of options for your dot tech and you have to have under 2 million in funding. We want this to be for early stage startups. Thanks to our friends at dot tech and we will see you next time. Bye. Bye.