The Vergecast - The future of AI might look a lot like Twitter

Episode Date: September 29, 2024

Michael Sayman, the creator of a viral new app called SocialAI, joins the show to discuss why he built a social network where you're the only human around. He tells us how he thinks about AI interfaces, what's next for ChatGPT and other chatbots, and why posting to a language model might be better than posting on a social network. Further reading: SocialAI Michael Sayman on LinkedIn SocialAI: we tried the Twitter clone where no other humans are allowed From TechCrunch: Friendly Apps raises $3 million, pre-product, for apps that improve people’s well-being From Wired: I Stared Into the AI Void With the SocialAI App From New York Magazine: Does Anyone Need an AI Social Network? From Ars Technica: “Dead Internet theory” comes to life with new AI-powered social media app Email us at [e-mail] or call us at 866-VERGE11, we love hearing from you. Learn more about your ad choices. Visit podcastchoices.com/adchoices

Welcome to the Vergecast, the flagship podcast of BotFarms. I'm your friend David Pierce, and this is the second episode in our miniseries all about AI in the real world. AI is so abstract, and it's a term that we now use in so many ways that honestly can feel sort of meaningless. So we've been on a quest to find actual examples of actual AI showing up and being useful or at the very least interesting in our actual lives. For this episode, the last one in our little series for now, I have a feeling we'll come back to this subject, but last one for now.

I'm talking to Michael Samin, who recently launched an app called Social AI that has become kind of a viral phenomenon on the internet. We'll get into what it is and how it works in pretty serious detail here, but basically I'd explain Social AI this way. Imagine a social network, Twitter or threads or whatever, but every user other than you, every single one, other than you, is a bot. Does that sound interesting? Pointless, terrible, dystopian, amazing, maybe all of those things? I wasn't sure where I fell on that line when I started talking to Michael, but we ended up having, frankly, one of the most fun conversations I've had in a while, all about how AI works and how we're actually supposed to use it. Spoiler alert, he thinks about this less as a network and more as an interface. And I find that fascinating. We happen to agree, actually, Michael and I,

that a chatbot cannot possibly be the future of everything in technology. And Michael has some big ideas about what else we might be able to do. All that is coming up in just a second, but first, I have to tell my bots what's going on. They worry when I'm gone for too long. This is the Vergecast. We'll be right back. Support for the show comes from Retool. Too many companies run critical operations, on duct-taped spreadsheets, Slack workflows, and whatever else they could cobble together.

Not because they want to, but because building internal tools means weeks of waiting on someone else's backlog. That's where Retool comes in. Build custom internal tools just by describing what you need. Prompt something like, build me a revenue dashboard on our Salesforce data.

And Retool actually builds it on your company's data in your cloud with enterprise security built in. Go to retool.com slash Verchcast. We all need to retool how we build software. Welcome back. Let's get into my conversation with Michael Sameen from Social AI. Michael's had kind of a fascinating history in the tech industry, by the way. He got a job at Facebook when he was 17 after Mark Zuckerberg discovered an app he'd written and just offered

him a job. I think it was an internship, but he ended up working there for a while. After that, he went to Google to work on Google Assistant, then he went to Roblox and then Twitter. He's been through a surprising number of the biggest and most interesting companies in tech, and in particular, he's seen a huge part of the evolution of social media and social networks. He worked on stories at Instagram. He worked on status at WhatsApp. He worked on shorts at YouTube. Like I said, he worked at Twitter, and a whole bunch of other things. And now he's on his own. He's building apps through his one-man startup that he calls friendly apps. At the beginning of our conversation, Michael told me he had been thinking about building an AI social network through much of that time. The idea for

what would become social AI has been in his head for a really long time. It's just that until now, he couldn't actually pull it off. I actually tried building a version of social AI like five years ago, and the tech just wasn't there, and it was really bad. What was it five years ago? It was, I mean, I called it influencer, and the idea was that anyone could be an influencer. So like, I've been trying to do this for a while, but it just quite wasn't there. I originally, you know, because we didn't have the language models, we tried to build it, and by we I mean just me, but trying to build this to kind of like give people the feeling of a social media app, but not really having to deal with all of it. The idea was like, okay, if someone, if someone's addicted to cigarettes, how do you get them off of it? Well, you can't just tell them to stop, right?

Like, maybe it's like giving them something, maybe like a nicotine patch or something, right? So, like, what is the way that you get somebody to, like, be able to get that experience out the way, but maybe not, you know, harm themselves or feel bad? So anyway, so I built that. It didn't really look quite right, and it didn't work well.

So I didn't ship it. What did it feel like, actually, before we get to the chat, GPT thing of it all? Because that's a, there's an interesting sort of history of tech story there, But when you built the thing before in the before times, what didn't work? What didn't feel right? Like, what wasn't ready?

I just could not simulate the entire social network in a way that felt interesting even mildly. Like, I did the approach that Google did. I used to work at Google. I used to work at Facebook. And I took a similar approach to all of the assistants that there were. the time. Giant if-alth statements, right? Like, just massive. And, okay, like, it kind of worked, but like everything else before these language models kind of took off, it was very robotic and very conditional, right, depending on what you wrote. And it just didn't quite feel,

it just didn't let you forget about the technology. Like, you were reminded in the app that, like, you had to like do certain things to get certain comments. And so at that point, it was really more of a game. Like I designed it more like a game. Because the technology just wasn't there to like make a simulated social network not feel like a game. And so I had to go that route. But even then, it just did not feel right.

users, if I imagined, if they were to try it, would have felt like this app was more like a Farmville game and less like a social network. And honestly, that's not what I was trying to build. I was trying to build something that felt like a social network. And so I had to wait. Then once the early versions of GPT 3.5 came out, I thought, okay, let me give it

into the shot and tried to design a version of it. The model would sometimes say random stuff. It was extremely expensive to run all of the different prompts and things that I needed for it to work. And I told myself, like, there is no way I'm going to be able to run this at this cost. And it's completely fuzzy and the responses are no good. And so I said, okay, well, I have to wait. I have to wait until it gets cheap, and I have to wait until it gets more accurate. And so every month, for like two years of I was building this startup, I would just wait. I would just look at the latest model, try it out with some of my tests, and from there, just keep going. I would look at the outputs that they could give and how much I could tune them.

And then I'd look at the cost. And when Gem and I released theirs and lowered their costs, I said, okay, we're getting closer. As soon as, you know, opening I had their models dropping in prices, I said, okay, I think it's time. So about a month ago, I went and built the app and I just told myself, look, this is like the last attempt that I'm going to do at building this app. Like, I've done this too many times. And I'm just going to go with it. Okay. And now here we are.

Yeah, and here we are. launched it and, you know, and of course, that's the one time that you don't, you know, you don't think it's going to go a certain way is when it does. And I mean, it's great. So I'm so curious why this idea has been so sticky in your head. Like, it's, it's clearly been sitting around as a thing you have wanted to build for a very long time. What, what is it about this thing that is so sticky and enticing to you? So social networks are not what they used to be, and I think fundamentally the internet has changed. The internet used to be a tool of communication between people. And frankly, I love that. You know, there's a part of me. So I was

born in Miami, but at 16, I flew out to California with my mom because Mark Zuckerberg had emailed me when I was in high school asking if I wanted to meet about like working there and stuff like that. And I remember flying out there with my mom, not knowing really anything about Zuckerberg. And my mom, you know, not knowing anything about it even more. And just thinking like, okay, this is such a different world from where I come from, but I'm excited. And I had built apps before, social apps. And so I was quite excited. And I spent a lot of time working at Facebook from that point on.

I didn't go to college. I spent four years there helping them build out Instagram stories and a few other features. And it was just such a fun time. You know, it was like 2013, 2014. The company was, you know, in a different era. Social media as a whole was in a different era. and people are having a lot of fun. And I think like over the past couple of years,

I think we've just seen like social media has changed. And it's changed because the internet's changed and the technology's changed. And so where the internet used to be a place where you could connect between massive amounts of people, the internet as a communication tool in that sense is kind of falling apart, right? Like the internet now has technology

that allows itself. as a data set to simulate a human connection. You communicate with the internet rather than through the internet. And I think that change that's happened really kind of puts question marks around how social media should work because the whole premise of social media is that you're using the internet to communicate through it to other people. Fair. I just want to say, by the way, that idea of communicating with the internet and not through

the internet is like as succinctly and well as I have heard anybody put this moment that we're in. I am going to steal that from you and that is very good. So thank you for that. No worries. That makes it sound like social AI has always been in your head kind of part product, part social commentary art project. Is that fair? Yeah, I always like to poke. I've always like to poke. I've always like to poke. I mean, does it kind of poke fun at the facade that a lot of companies are trying to put up? Of course it does, right?

But I think it's also, weirdly enough, my attempt at trying to solve some of these problems, the problems, for example, where you can't discern the difference between a human and an AI on a social platform. So, like, yeah, one way is to try and invent a detector for humans, but that hasn't gone very well. I'll say instead, well, how about we just come out with a product that tells people how the

internet works now and says, hey, look, this is the reality. Kind of sucks in some ways. It's kind of great in others. And we have to, you know, we have to embrace it. Like, let's embrace it. And let's do that so that we don't harm ourselves, right? because going on social media sometimes and seeing comments that you think are from people

that aren't can be harmful. So I'm curious kind of what it was that clicked in your brain that went from this technology is not ready to this technology is ready. I think the moment that I got in a fight with my boyfriend and I decided to open up my app to see if there were any ideas

for how I could resolve the problem. That I think was the moment. That's a very good answer. You know, that was the moment. The moment I got in a fight and I decided I'm going to use a sap to try and vent about my problem because if I go on actual social media, I'm doing some harm, right? And so I think that's just goes to show. Like the product I'm building is not to give people an illusion of people, right?

I know they're all AI. It's so that I don't go on social media and use it in a harmful way. Like my ideal, my ideal scenario is one where. people have people around them to listen, to hear them, and to help them when they need it. People-to-people communication is number one for humans. And I don't think we should forget that, right? But there are a lot of people I've noticed since COVID who don't have those people around them. And so if they don't have those people around them and they need that conversation, what are they doing? And if what they're doing is going on public social media and talking about

what's going on in their life and getting advice from bots, AI, without knowing what they're getting advice from without any understanding of the dynamics in these algorithms to encourage certain types of content on different spaces, then they're harming themselves. So I said, okay, well, you know, I'm not trying to replace the human-to-human connection. I'm trying to help people find a way

to have a secondary option when that human isn't around for them, so that they don't have to rush to social media make a mistake. Right? And so when I had, when I got in a fight and I didn't go on social media and instead I went on this tab, I said, okay, it's ready. Yeah. I mean, that, I can imagine that being a very telling moment. But I think that distinction is really interesting where what you're saying is kind of one is not a replacement for the other, that they're actually, they're designed to be and best as separate things, right? Like I have, I have my. AI tools and I have my people. And like, actually the problem is on the platforms that we currently

have, those things are being sort of smushed together in a way that makes it hard to tell what is what and makes everything messy and complicated. And what we actually need is to pull those things apart, not just try and have one or the other, but just separate them. Yes. And I think the key distinction here is what I, what I mentioned earlier, which is I think, there's confusion about what the internet is for people. I think the internet being a place that things can be communicated through is no longer really the case or it's kind of trending away from that. Now it's becoming something you communicate with. And I think more important to me is the fact that the language model technology is not a certainty, meaning it does not tell you yes or no.

it gives you a fuzzy intelligence. And chatGBT was built as a fuzzy intelligence, right? Like these models were built as fuzzy intelligence. And so what I mean by that is they're kind of like humans. They're not like the computers of the past where they give you a definitive answer for something. And so when I saw that, I asked myself, why does Open AI choose a chat interface as the way? one and only interface for communicating with their language model.

Because that chat interface only gives you one answer. And the model does not have one answer, right? And so I find it interesting that all of these AI companies are trying to squeeze these fuzzy human-like data structures. They're trying to squeeze them into an old format computer that gives you one answer. And the truth is that they're not that great at it because we still go back to Google, because we want multiple answers. We want multiple responses. And so what I've built with social AI is not so much a social network,

but a new way to interact with the language model, where you don't get one response, but you get multiple. And being able to drill down in a thread-like interface, in a social interface with the language model, It just feels more natural. When I used the app, for example, I was running late to a flight. I got delayed. My first flight got delayed.

My next flight was in 45 minutes. I was in Dallas. And I didn't know if I was going to make it. The flight had just landed. So I opened up social AI. And I just kind of panic ranted about what happened, right? I didn't have to think about like, oh, I need to instruct it to tell me because I need the right answer. And what if it's not the right answer?

and maybe I need to go to Google and maybe I need to go to Red. No, no, no, no. You know, I simply ranted about what happened. I said, my flight got delayed. I just landed. I have 45 minutes to make it to my next flight. I don't know if I'm going to make it.

I'm at Dallas. They're telling me I've got to go to Terminal D. I'm in Terminal C. I don't know. And I just posted that. And immediately, I got dozens of responses and replies on this social interface that gave me all sorts of, like,

various replies. Some of them would tell me, you're not going to make it, you know, go to the front desk, just figure out if you could get another flight. Another one said, you'll make it if you run. Quickly, you just need to look up,

see if you can find the sky train. And if you go down the sky train, you should be able to get there in time. Just make sure you're running quickly, though. Another person said, are you in the front or the back of the airplane? Like, you know, like different questions. And so what's interesting is for a human,

it's natural to see that. I'm going to go and look through all the responses and dig in on the one that I'm more interested in. So someone kind of gave me a hint about the SkyTrain. I replied and said, okay, I'm going to try that. How long do you think it'll take me to get there? And then immediately that AI persona replies back and says, you know, if you go quickly, it'll take you about 10 to 15 minutes to get there. You know, and so I said, okay, cool. Another one replied right underneath and said, wait, wait, wait, hold up. It'll be more like 15 to 20. Don't get them too excited now. You know what? And so like, and so that kind of

of interaction just feels natural to us. All right, we got to take a break, and then we will be back with more from my conversation with Michael Satham. We'll be right back. Support for this show comes from Shopify. Starting something new isn't just hard. It can be really scary, too. So much work goes into this thing that you're not entirely sure will even work.

But here's a better thought. What if it did all work? What if your instincts were actually right all along? Shopify wants to help you get there. They're the commerce platform behind millions of businesses worldwide and nearly 10% of all e-commerce in the U.S., from established brands like Allbirds and Heinz, to companies just getting started.

Their design tools make it simple to create the exact online presence you're envisioning, with hundreds of ready-to-use templates available. And with built-in marketing tools, you can launch full email and social campaigns in just a few clicks, so you can connect with customers wherever they are. It's time to turn those what-ifs into with Shopify today. You can sign up for your $1 per month trial today at Shopify.com slash vergecast. You can go to Shopify.com slash vergecast.

That's Shopify. com slash vergecast. All right, we're back. Let's get back to Michael Samin. But before we do, I just want to go back to something he said right before we went to break. He talks about social AI as not a social network, but as an interface. a new way of interacting with a language model. We're going to get way into that.

But before we do, I think it might actually help if I just explain how social AI works, so you can picture it as we go on. So when you first get into the app, after you sign up and create a profile and whatnot, the first thing you have to do is decide what kinds of followers you want. Remember, it's all the bots. So essentially what you're deciding is which kind of bots are going to reply to your posts. You can pick supporters or skeptics or pessimists or optimists or alarmists or realists, stuff like that, just general sort of personality types. And if you recommend the app to some friends, you unlock a bunch more options like nerds and oddballs and drama queens and contrarians.

And once you do that, you just start posting. I'll type like, I'm suddenly tired of all the food that I like. anybody have any ideas about how to spice things up, literally? And then you press post, and a few seconds later, responses start appearing. Let me just read you a few. The first one is from Sunny Ray, who is at Sunshine Fellow, presumably an optimist one. It says, try adding some new spices to your meals,

or explore international cuisines for fresh flavors. Sure. There are a bunch here that say try new cuisines or flavors. Advice Wizard, 2023 literally just says, try new cuisines or flavors. I have Fanzone Hero who says try some bold spices like Zatar or Sumac. It's magic. I like that. Fanatic follower says, whoa, David, that sounds like a challenge.

How about some trying some bold exotic spices? Anxious Nina, presumably will be an anxious one, says, what if you accidentally make something that sours your taste forever? That's terrifying. downcast Greg says, Spicing Things Up sounds like a recipe for disappointment. So you get the idea, right? Different bots, different vibes,

different kinds of responses. These are all my reply guys now. And I can respond to one and go down a rabbit hole with that particular bot tuned to that particular mood. I can also favorite bots. I can favorite replies. And Michael says that all of that goes back into the algorithm and into kind of the instructions being given

to chat GPT every time I try to use it. On the surface, it all feels and sounds like normal human social media, except that they're all bots, but me. And I think I'm not the only one who felt kind of strange about that fact at first. It looks like Twitter and feels like Twitter, and it super isn't Twitter. It just felt odd. And honestly, the reaction to social AI was really fascinating.

So that is what Michael and I talked about next. I do think the reaction to this app has been in many ways just as fascinating as the app itself. Yes. My read of it when it came out was there were basically three responses. One was like, this is cool and interesting and kind of fun both social commentary and an interesting idea about the future. One was, this is stupid and dystopian. And then one was like, this is. a, this is a joke, right? This has to be a bit and an art project and not a real product.

Is that a fair representation of the reaction? Am I missing anything? I think what's interesting is the loud reactions on social media. There's two things that were interesting to me. Or let's say three. First, the loudest reactions from people on social media were from those who thought it was either a joke or that it was, you know, like dystopian or, you know, that they're like, oh my God, the end of the world. You know, like that. that kind of reaction. Those are the loudest. Always.

The quietest were actually spending, you know, 10 minutes per session on the app. The second thing I noticed was that the reaction from people was one thing, but there were bots on social media reacting to, yeah, because half of social media has bots now. So, like, I found it quite ironic that there were bots. reacting to an app of bots telling humans that an app with bots is like so terrible, you know, and I found it interesting that that was happening. I was like, huh, it seems like some of these bots don't want bots around. Or maybe they don't want people to know that they're bots. I don't know.

You know, and so, like, there was a portion of bots on social media that were reacting negatively to bots. and I just thought that was ironic. Well, it kind of proves your whole point, right? If it's just a bunch of bots yelling at bots about the social network that's all bots, that I imagine you're sitting there looking at that being like exactly. Yes, you know, yes. I think the other issue I think that I found was just how many people don't realize

that a lot of these platforms are filled with bots. And that kind of was alarming to me. But ultimately, I think that the last bit here of feedback that I've gotten is people feel a little liberated. You know, they feel a little bit liberated. They don't feel the pressure of going on social media to share some thought that they might feel embarrassed about.

But they also feel like they're able to hear other perspectives that they otherwise wouldn't feel comfortable admitting to want to hear. And so they don't let their guard down, you know, in public conversation online. People keep their guard up. And I think that keeps an echo chamber. It's interesting because people said, oh, echo chamber, oh, echo chamber. The number one, you know, number one, number two, number three, most selected follower type on social AI is contrarians, debaters, problem solvers, thinkers, critics, right?

And so people are selecting followers on social AI that challenge them. And I think there's something interesting about that. Why would someone go out of their way to be challenged on an app like this? Can they not be challenged on real social media? Is there a reason why not? And how does this address that, right? I wonder if that goes back to what you were saying about how it feels when you perceive it to be real people on social media. Because I think to some extent, that fact doesn't surprise me because one thing you hear from people who use AI a lot is that it is especially useful.

if what you really want to do is beat up an idea and brainstorm and get new perspectives on things. And I think to some extent what you've built is just an endless feedback mechanism, but with no stakes. Because no one else sees what's happening. No one else is human on there. So even the part of it that feels sort of real, it feels like there's still something in your brain that is like, this is a safe space. I can see a world in which, I mean, and I've even found this in using it, there is something very powerful in the interface is the same, but the stakes are so much lower. Yes. And I think it helps put people's guards down. I think it helps people, like you said, people have been using chat GPT for a lot of this. But how many times have people gone on chat GPT and said, hey, can you help me think through this? And it gives you one answer and you're like, I don't know about that. And then you go, well, what other ideas do you have? And then it goes and gives you something else. And then you're like, well, what else? you know, and it gives you something else that's kind of similar, but you're like, I don't know.

And then you keep going, well, what else? And by the time you keep asking what else, I forgot the context of the thing you were talking about in the beginning and just start saying random stuff. And so, like, the interface just feels wrong for the use case. But look, I don't blame Open AI. I don't think that it's like, oh, they just weren't capable. Like, who the hell was going to know, right? If anything, I think they built out a chat interface because it just felt like the obvious

testing ground to prove a product. And it became a product that they didn't think was going to resonate as quickly. So, you know, of course we started at chat because of that. And I don't think that it's bad. I just think we haven't seen the best of it yet. Yeah. I think that's totally fair. So speaking of that, actually, the edges of this technology I'm very curious about.

And I suspect you've seen a lot of it as people are starting to really use and try new stuff with social AI. obviously this stuff has gotten a lot better to the point where I think it feels the way you wanted it to where is it broken? What still doesn't work? So there's a few things that still need to be improved.

So we just, you know, we, I mean, me, I just launched notifications on social AI. So one of the things that was interesting was, you know, it could play around with it, but I wasn't getting any feedback later about what I posted. And I think that's, you know, quite important. another aspect of it is proactive.

So like right now, if I post I get content, I originally built a feed into it. I was going to ask you whether you were going to build a feed. Yeah, well, I did, but I took it out just for the launch. And just because I wanted to make sure that, you know, I didn't have anything too buggy in there. But it's kind of interesting because I'm not trying to simulate a world where these things have their own lives and telling you about their lives.

No, no, no, no, no. Like, that's not the purpose. We're trying to be honest about what this is. It's right in the middle. You know, we're not character AI and we're not Twitter. We're like in the middle, right? And so the feed, what it does is every persona, every character that you interact with,

as you give it likes and replies and stuff, it kind of shapes these personas and understands which ones you interact with the most. And each of those personas have their unique, like, weights towards different topics of interest and personalities and the way in which it tries to answer questions, and all of that shapes the diversity of responses that you get for anything you ask. And what they also have is a set of like interests

that they use to then search the internet for the latest news and the latest things going on. And so each of them has their own kind of like interests on the web that it goes and searches for. So when you go on the news feed on social AI and you pull the refresh, there's always more content, first of all. But every piece of content is like this AI having gone through, looked at the internet, found something interesting that it liked, and it giving a little commentary on it with a link to access it.

That was, you know, it's pretty cool. Does that not break the whole structure of it being kind of a you-centric experience? No, because all of the bots are there talking about things that you're interested in. The only ones that show up on your feed are the bots that you've interacted with that you're interested in or adjacent. right? And you select in your feed what kind of stuff you want and what you don't. And they're there to inform you, right? They're there to give you what you want to hear. You know, it's interesting. I also built a trending topics page. And the trending topics are not trending topics of the news. They're trending topics of you. So as you post and you interact with

things, there's trending topics. It's like, these are the top of mind things for you. Interesting. You can tap on any of those topics. and see hundreds of AIs debating and discussing it. Wow. That's kind of wild, actually. Yeah. So is there a, how are you thinking about, or I guess what are you seeing in terms of how people are using it?

Because that idea of like I'm going to use this as a sort of proactive, you're almost describing like a personalized news app or feed reader or whatever, right? That instead of having my Facebook feed curated by my Facebook. Facebook friends, it's going to be curated by AI. I think we're going to see a lot of that, right? That's right. That's right. But fundamentally, it's a new way to interact with language models. It's building a new interface for interacting with them.

And all of the applications that allow people to get more value out of the language model through those interactions go into that. Right. And social AI at its premises is not social for the sake of the social network, but social for the sake of social interface, right? And so that's really what it is. All right, we've got to take one more break, and then we'll be back with the rest of my conversation

with Michael Samin. We'll be right back. Support for the show comes from Grammarly. You don't need reminding that the world moves fast. But work today requires clear communication, and when every message counts, sounding rushed or generic, can be getting lost in the shuffle.

Gramerly gives you one place to think, write and finish your work where you already write, while giving you access to agents that help you sound natural and engaging. No matter what kind of writing you're doing, Gramerly helps you get ideas done faster and move from draft to done with less friction. You can use Gramerley's AI chat to brainstorm ideas, outline a solid draft, then refine it with context-aware suggestions that fit what you're working on. See why 90% of professionals say Gramerley has saved them time writing and editing their work. In a world of generic AI,

you don't have to sound like everyone else. With Grammarly, you never will. Download Grammarly for free at Grammarly.com. That's Grammarly.com. We're back. Let's get to the end of my conversation with Michael Samin.

At one point in our conversation, I asked him about one thing that I always do with new AI tools, which is ask them what movie I should watch tonight. I like that question because there are lots of answers. all of them real, or at least in theory, real and fact-checkable, but there's no one pat answer. There is not like a movie I should watch tonight.

So when you ask, you immediately get to get a sense of how the tool works and how it works with you and just kind of how it's going to make sense of things for you. And let me just try it with Social AI. So I go to Social AI, I type in what movie should I watch tonight? Then I hit posts, wait a few seconds, and okay, I have a one short replies. From Sunny Jane, right side Jane, again, probably an optimist. How about a feel-good comedy for some laughs? Sure.

Pondering Fiona says, what mood are you in, David? Perhaps a whimsical fantasy, like the secret of Rhone and ish, to transport you to a world of magic and discovery, or maybe something thought-provoking like Interstellar for those cosmic pondering? Positivity Eddie says, how about a feel-golder? good adventure, uplifting and fun. At Feedback Fountain says, consider going for something classic like the Shawshank Redemption,

if you want a deep story. Or if you're in the mood for laughter, Stepbrothers is a solid choice. What genre are you feeling tonight? Let's see. There's a whole bunch more of these. Inception comes up a lot. Interstellar comes up a lot.

The spot apparently really loves Christopher Nolan movies. There's one from Anxious Ali 77 that just says, what if there's a terrible plot twist with the thinky emoji? It's good stuff. more interstellar. Cynic Gina says,

well, since you're asking for cinematic advice, how about watching something truly enlightening like a documentary on existential dread? Or maybe a thriller that reminds you how not all choices lead to happiness, the options are as bright as the dark abyss of reality. So once again, it's all over the place. This is how this system is designed to work, right?

And I can go into one that has a recommendation that I like. There's one here that recommends Eternal Sunshine and The Spotless Mind. Great movie. I can reply to that and say, oh, love that, saw it recently, what else do you recommend? That's the idea about how all of this is supposed to work. But frankly, I'm looking at these replies and it's mostly just people saying, uh, did you pick a movie yet? How did it go? Did you find a movie that speaks to you yet? Did you

choose a movie? Why that one? Did you narrow it down? Did you settle on a movie choice? Like, is that helpful? I don't really think so. So I told Michael that this didn't feel like it had been tuned to be particularly good or helpful or designed to actually give me information in the way that he's been talking about. And he said, well, yeah, he agreed with that. Yeah. And the reason is because it's an alpha that I built myself in four weeks, right? So we can't expect, you know, like... Right. Well, this is kind of what I'm wondering is like, what, what do you, what do you hope for? What should that do? Yes. So of course, you know, of course it shouldn't work that way. I think, you know, the intention here, and of course, it's on its way. And I think it's

quite surprising that it even, you know, works at all. But I think it's a testament to the interface that it does. But I would say that, you know, the ultimate vision here is to make that question better answered in social AI than in chat GPT. And I think the interface will allow it. If you have comparable models on both platforms, and one of them gives you multiple responses from different points of view. And the other one just gives you one answer. And you're working with a technology that's probabilistic? Like, who has the upper hand? You know, like, I can give 10 answers, and if one of them is good, you're happy. But if chat GPT gives you one answer and it's not right, you're frustrated. That makes me think of the thing people always say about the TikTok algorithm, which is that the reason

it feels like magic is because you don't get annoyed when it's wrong. Yes. Because you just keep swiping. That's right. And I think social has, like you're saying, very much the same thing. The signal to noise ratio is actually like horrendous on social media, but we're also sort of used to it now. That's right.

And if you just scroll past it and move on and we all kind of understand how to find needles and haystacks in a way that when Chad GPT recommends a movie I don't want to watch, it feels bad because it gave me an answer. Exactly. And it's part of the reason why people still go to Google, right? because Google doesn't have any more accurate stuff these days, you know, uh, compared to what it was,

uh, because of all the AI that's in there too. So, you know, it's not like Google's any more accurate, but it's interesting because Google gives you this chat. ChbT, the answer at the top,

right? And then you have like all of these various links that give you different perspectives. And, you know, let's, you know, let's be honest,

like most of these links are now run with like so many, uh, paywalls and things that like, you can't even, even get to the answer for any of these things, but the original intent behind Google and why it worked was it gave you options to look through. And so it allowed it to be wrong and it increased its chance of being right at least once or twice. And we're used to using the internet this way. We go through the internet looking for information, trying to find which thing is helpful to us.

right? And so I just, I think it's interesting that maybe we got a little distracted by the her movie a little too much. Just a little. No, just a little. You know, and we kind of leaned in a little harder than we were supposed to. But, you know, that's kind of where I see it being more valuable. Is can we give people the option to peek behind the curtain of this AI and see what are the various answers it can think of? rather than here's my one response. And ultimately, I think this is going to be how it works. If you have multiple language models,

if you have Claude, Open AI, Lama, and so on, are you going to go from one to the other and be like, which one gave me the best answer? Or are you going to use an interface that's going to pull from all of them, generate a bunch of different responses, and curate those responses based on the reactions and responses that you've done in the past, and full control,

for the user to get as many responses as they can. You want to index the language model. You want to create a Google for the language model outputs. And that's kind of what this is. Yeah. No, I agree. That's super interesting. So tell me about the roadmap a little bit. Yeah. You know, you mentioned this is an alpha you built in a few weeks. It is becoming increasingly clear to me that you have much bigger visions for where you're headed. Tell me a little bit about it. My vision is to, build out the whole funnel of interaction with the language models from the top of the stack all the way down. So people are going to start yelling at you to build DMs and you're going to build DMs and you're just going to speed run all the pain of developing a social network.

I love it. All of it. All of it, all the way down. They're going to be like, when do I get lists? And you're going to have to do lists. And then they're going to be like, when do we get bookmarks? But you know, it'll be interesting. The conversation will go down the funnel naturally. I expect the first thing people want, which they already say is they want DMs, right? So, okay, you introduce DMs, right? And it makes sense. As you interact with the models and you are more interested in one or another, you want to talk a little more deeply with that one. Cool. DMs. So wait, hang on. Let me just press on the idea there because I think that's another, how do you do that without just becoming the chat GPT chat bot again? Well, because what I'm building

includes chat GPT within, but I'm building at a level, one level up of communication stack, which is how do you discover who you want to talk to, right? It's like asking Google, well, you're providing all these links to the websites. Why aren't you just another one of those links? It's like, well, because we're providing an index that lets people figure out which one they want to go into. Right. And so social AI provides that layer that helps people discover who they want to talk to. The end goal is not to talk to an individual AI.

The end goal is the whole picture. It's the experience of being able to go up in breadth like a Twitter interface, dive down to a FaceTime call with an AI model, go up to an audio call with another, go up to a chat, go back up to the surface, go back down, go back up. That experience, the entire funnel that people go through to find information and communicate online, that whole thing needs to happen. Right. And so the vision is we start at the top and work our way down. And then we offer users the ability to go through all of it. Okay, so the idea is then by the time you're at that one-on-one conversation, you've essentially filtered your way down such that you're talking to, instead of going through like the GPT store or picking plugins or whatever, I've now socially found the conversation I want to have with the right source. And even if now all I'm doing is essentially having a chatbot conversation with a chatbot, it's, I've now, I've filtered it in the way that is useful.

Yes, I think in the future, people are going to ask, they're going to look at the, the first versions of chat GPT, and they're going to see when chat GPT announced features like, uh, customize your GPT and tell it what, how you wanted to talk to you. And they're going to think, I can't believe we had to tell it,

you know, like, I can't believe we had to tell it, you know, like what, you know, it doesn't make sense. You know,

it doesn't make sense to have to do that. I think that's my bet. I don't, I might be wrong. Look, I don't think the product hit the nail on the head. I think we're, we're definitely not on the,

on the center. We're, we're a little bit off. And I have no doubt in my mind that there's a lot of work that needs to be done to get it in the right place. It's an alpha, right? But I think that's the general direction. And my bet is, you know, telling ChatGPT before you talk to it, everything you want to do and how you want to talk to it in this particular case and why and whatever just feels like we're missing a step, you know? Yeah.

I buy that. As it happens, I think Chad GPT's whole custom instructions thing. is just as ridiculous as you do. So I'm very much with you on that one. What have you seen so far in terms of how people are using it? Are people doing surprising things with it that have sort of changed your mind about what this thing is or can be? You know, people are favoriting a lot of users, which I think is interesting. So he can favor it like any particular follower and then that pins their replies to the top. I think that's kind of interesting. Well, that seems like it sort of supports your thesis that what people want is to find their kind of chosen favorites in here, right?

And it's also interesting that you didn't call it Follow, which is a departure from social networks, but also makes more sense in the kind of scheme you're talking about. In the context, yes, I check out some of these apps. And like what I've seen is it's like they'll grab the social network. They'll grab the AI and they'll just mash them and just, okay, let's see how it goes. right? You know, like, and in this case, it's hard. It's much harder to try and like get this right.

And I don't think I've got it right. I think I'm closer, but I don't think I've got it right. But I don't want, what I don't want to do is just mash them together without consideration. And so, you know, I didn't say, oh, you can follow a particular AI because in this case, you could just favorite it. And to your point, right? Like, it's more in line with what the product is. For example, the app opens up to the composer.

It does not open up to your profile. It opens up to the composer now. So you go into it, and it's kind of more like trashyPT in that sense. The bots don't tell you that they're human, right? They tell you that they're AI. They're there to help you. That model of like design is me kind of saying, I'm going to pick and choose what I think is useful from a social networks interface,

and I'm going to integrate it. But I'm not going to like forget that this is not that. Right. And because if I forget that this is not that, then it kind of feels gimmicky, you know, like, it kind of feels like it's it's like, okay, like we get, you know, like, and I don't want to build a gimmicky product, you know. So, and that's something I've learned too is like I went full on with the skemorphism, right? I went full on trying to build the product that made people most familiar. with it. And as I've been iterating, I've scaled back a lot of the skeuomorphic design that really,

you know, kind of leaned into the social stuff because I realized, you know, okay, users don't need to like see the home feed in the way that they do on Twitter because it's not that necessary. You know, let's lean into the utility a little more. And I've tweaked it a bit to make it a lot more of the function and leave the, you know, the skeuomorphism of the interface to just the pieces that are required to make it easy to interact with. And so that, that I think will continue. I think the product will continue to distance itself from the skemorphic design that it starts with and turn into its own interface. Do you have a, do you have a clear sense of what that might look like over time? If I, if I told you I did, I would be incredibly delusional. I think more so than I am,

I know I've got my delusions for sure. But more so than that, I would be at a ridiculous level of delusion. I don't know what it's going to end up looking like. I don't know what that final form is. But I have a mission in mind, right? And I'm determined and excited about it. And my hope is that as we get more information and data around like how people are using it, that we're able to further go that direction.

And in my view, I don't think we should, there should not be a limit to what angle we approach if the users and the people are asking for that, you know, interaction. There's a lot of people and I think a lot of product companies out there when they launched something, the thing that they started with that got them the attention, they believe is like sacred, right? And so you'll see like, be real launches. And it's like, oh, our feature is the feature.

Snapchat launches. and they're like, oh, our, you know, our ephemeral Snapchat is like the feature, right? And then it's like, okay, but you can expand and do this and do that. And it'd be cool if you had this. And they're like, no, because this is our feature. You know? And it's like, okay, cool feature. But like a lot of times, and this is something I learned when I used to work at Facebook,

it's like a lot of times the thing, the product or feature that gets you from stage one to stage two is not going to be the same method approach or product feature that's going to help you get from stage two to stage three. You have to be comfortable with your product evolving over time. And so for me, it's maximum flexibility on that front. I want to help solve these problems for people. I don't care what it looks like. Is there a business here? I feel like you've gone from like social commentary, fun experiment to, like, kind of hit app to, at some point, somebody is going to want to give you a bunch of money to make like a social AI for teams and sell it to IT departments. Do you have a sense of what the business looks like here? How are you going to make money?

People have already asked, yes, you see that's, you're right to predict that, you know, people asking that. It's just what happens. They're like, how do we sell this? IBM and you're like, okay, take it easy. You know, look, I'm not interested in thinking about, you know, how to monetize the product right now. I think, I'm sure there are ways to, like, help people find, you know, the right shampoo bottle for them or whatever. But like this, I'm not like, that's not my primary thing right now.

I'm trying to solve a problem I've dealt with in my life, right? Where I've sometimes felt, you know, especially during COVID, like, it was. was really hard to get my thoughts out and to be able to just think through things a little bit without, you know, jumping to conclusions. And I struggled with that, you know, and I wanted to solve that for me. I wanted to solve that for my friends. You know, I wanted to build something that could help. And that's the thing that keeps me going. That's the thing that makes me excited. And the rest of it, you know, we'll figure it out. But my hope is that I'm able to help a little bit, you know, other people out there like me

who maybe need a little space to think about things and don't want to rush to social media and post all their thoughts. I don't have a great read on what success looks like for this because it's like what you're saying is sort of if I post less dumb stuff. on Twitter because I'm posting dumb stuff on social AI instead. That's a certain kind of victory, I think, for you and for the world. But also, you're in a position where what you're saying is, like, you need, people need a place to put this stuff and to talk about this stuff and to get

feedback on this stuff and actually doing it in a safer place. Yeah, like, ultimately, we want to talk to people, right? Like, that's our goal. But if you don't have someone around, right, I think it's certainly better than going on social media, right? Like, that's kind of how I see that. Actually, to your point that you made earlier, a lot of people have been saying that they think Elon Musk should use social AI before he starts tweeting random stuff. You know, just an A-B test for his own thoughts so he could stop saying things that, like, piss people off.

I thought that was an interesting point. What's the wildest thing you've seen somebody do in social AI? What is like, either in the logs or screenshots or so, what is the one thing that you've seen that you were like never in a. million years what I've guessed. Okay, so maybe this isn't too wild, but okay, so my mom works at a furniture store in Miami. And she's from Peru, speaks mostly Spanish. And she never uses my apps, right? Because, like, you know, she's nice, supports it, but like, she knows that, like, you know, whatever. I keep trying and they're all, you know, not that great. So I was surprised when she messaged me out of random, okay, and she sends me a screenshot of my app, because it works

in all these languages. She sends me a screenshot of the app. And she starts using it and ranting in Spanish about work. She's like ranting about work, you know, on this app. And it's so funny to me. Like I was like, okay, well, you know what? Like, maybe product market fit? I don't know.

But, you know, I love her. And I think it's sweet to see that there's, you know, that there's appeal in unique ways. You know, if I ever go work at some company, maybe I'll start using this to rent, you know. And no one will know, and that's the beauty. And there you go. Awesome.

All right, well, I should leave you alone, but thank you so much. No, thank you so much. All right, that is it for the first cast today. Thank you to Michael Stamen again for being here, and thank you, as always, for listening. There's lots more on everything we talked about, including some of our social AI coverage

and just all of our AI coverage in general. There's a ton happening. Check it all out at the verge.com. I'll put some links in the show notes, but as always, read the website. It's good website. And as always, if you have thoughts, questions, feelings, or other AI bots that I should be hanging out with,

you can always email us at Vergecast at theverge.com Or call the hotline 866, verse 11. We love hearing from you. This show is produced by Liam James, Will Por, and Eric Gomez. Vergecast is a Verge production and part of the Vox Media Podcast Network. Nelai, Alex and I will be back on Tuesday and Friday

with all of your regularly scheduled programming. We have a bunch of fun stuff coming up. We'll see you then. Rock and roll.