Airbnb’s Co-founder is Redesigning Government Services | Joe Gebbia

0:00 I don't think people deserve mediocre. Why would you put something mediocre into the world, especially at the nation level? Is that how we respect Americans? The Airbnb could not have launched years earlier. The internet had to grow up to prepare for Airbnb. I remember we were in the apartment when we were 10 employees. And the early engineer gets a phone call. He goes, "Oh, hey. Okay, go to the top right." Yeah. Click where it says profile. No, no, no. Go back to page. No, no, mom. Mom, go back. We all looked at each other and we go, "Guys, if our moms can't figure out how to use our website, this will never become a mainstream company." Now in the government it's different because it's not just moms actually upgraded and we operate the team that has to be grandma friendly. We've come up with not another website, not another way of bureaucracy, but an entirely new front door to the United States. It's called America.gov. Wo taking 29,000 entry points and saying let's just make one. It's the front door. You just tell us what you need and we meaning the government go get it done for you. To take that literally like is it like a Google or chache style text box?

0:56 >> Yes. One, people don't trust the government, and two, people don't trust AI. How do you increase the likelihood of people really getting the upside of what you're describing? The first part of your question sounds very familiar to the early days of Airbnb. How are we going to trust that? I don't trust that company, Airbnb. I don't trust that host that's advertising their home on it, their platform. In some ways, there's parallel of having to navigate through a gap in trust. Good design is when a user actively thinks or says out loud, "Wow, somebody really put a lot of care into this to make my experience better.

1:27 Welcome to Dialectic episode 59 with Joe Jebia. Joe is the first chief design officer of the United States and leads the National Design Studio which was created via executive order by President Trump in August 2025. Joe is about a year in and he and his team are leading a modernization of so many of the interfaces and systems that Americans use to access government services. And that includes the recently announced America.gov gov. A central entry point for accessing information from the government. America.gov is a chatbot style interface with LLMs under the hood, including Rampart, a privacy preserving openweight model that the NDS developed earlier this year. We talk about that as well as many of the other things the NDS has worked on from Trump RX to Trump accounts to real food.gov or the new food pyramid as well as the early work Joe did to modernize and speed up the federal retirement application process. Joe is also the co-founder of Airbnb and Samara. We had a conversation about why design matters, how he thinks about designing for trust, and the scale and stakes of a role that serves every American and that has a ticking time clock associated with it. I hope you enjoy the conversation and are inspired to think boldly about how you might serve others or your country. If you enjoy the conversation, please share it with a friend. I'll also link to America.gov, the transcript, and everything else we discuss in the description. Before we get into things, I would like to thank Notion, Dialectics's presenting partner. Notion has become something like a Switzerland in the crazy era of AI that we are in.

3:05 As new models release constantly and new capabilities and agents are always available, Notion acts as a context and consolidation layer for all of the ways you and your team work with AI and agents by sharing and storing context skills via the skills API and all the other ways that you work with AI. Notion can be a command center, a control hub, or simply a resilience layer so that even if you change the AI model or product you're using, your context can come along with you. You can also work with agents right from within Notion, whether that be custom agents that you create there, workers via Notion's developer platform, or tagging other agents you work with across various tools. OpenAI also just announced that you can sign in with OpenAI right within Notion to use your credits there, and I suspect that will follow for many other services. I don't think the capability ramp on what these tools and agents can do is going to slow down anytime soon.

3:59 So, it's great to have a single place in my workflow that doesn't change. Notion is that hub where all of my context lives. You can learn more at notion.com/dialectic. Now, here is my conversation with Joe Jebia. Joe Jebia, thank you for joining me. Jackson, thanks for having me. >> This is an honor. we're here in DC and feel like we're we're not quite in the middle, but we're close to the middle of this crazy new endeavor of yours. So, I'm excited to talk about it.

4:26 >> We're very much in the middle. >> Yes. Okay. Okay. I guess we'll get into that. I want to start maybe going a little further back with a quote that I is exemplary of one thing I want to talk about which is the storytelling side of your work and your design and your career. This is a quote I think from a conversation with Tim Ferrris in 2018. This is you design is more than just a product. It's more than just a digital experience. That cushion is 50% foam and 50% story.

4:56 What do you mean? >> Wow. Well, it it was a a great lesson early on in my design entrepreneur career. you know, I studied design at at Rhode Island School of Design and look, in design school, it's it's pretty theoretical. You know, you you work on an assignment, you make a prototype, and then you move on to the next assignment. And I was always curious, well, what happens next? What's the gap between that prototype you make and the shelf of a store?

5:25 What is that mysterious black box? So when I graduated Risy, I decided to demystify it and decode it. A lot of my friends went off to get design jobs and they moved to LA or Brooklyn and I sat in my apartment and spent the next year bringing a a product to market so that I could decode this mysterious gap. and I used a a design that I had created during my senior year, for a seat cushion to sit on during art critiques.

5:55 I also called crits where you put your art up on the wall, pin it up, and you spend all day in the art studio going piece by piece critiquing the art. You have to defend yourself as the artist and your classmates and the teacher are all saying, "Well, why'd you do that? Why does it look that way? I'm confused about that." that and you have to really stand in front of your art and defend it, which is a good skill for later in life. and so in these critiques, they're really long.

6:20 It's uncomfortable. And, you know, early on in my art school career, it was I'm sitting on the hardwood floors and it's just so it's so uncomfortable. you know, by hour six, your butt really hurts. And, by the end of the day, everyone stood up and walked out of the classroom. And I remember seeing the, you know, charcoal dust and the ink that's on the hardwood floors had rubbed off on people's pants and it left a bun print on people and I'm seeing everyone walk out with a bun print. I look at my own pants and sure enough, I had one, too. And I was walking back to my dorm room thinking, I've got four more years of these crits. There's got to be a better way to to participate in these that's comfortable and keeps you clean. Got back to my dorm room. I sketched that bun print shape on the back of people's pants and I called it crit buns. It' be this foam cushion with a handle that you know would be affordable for college students and something to sit on during these crits.

7:16 And so that was the product. I'd prototyped it in school and then by time I graduated it had won a competition to be given out to the entire graduating class of mine >> as a gift from the school. And so I had to figure out how to make it you know 800 of these cushions in a very short amount of time. It was almost impossible in about four weeks. And the day after graduation the school gave me the metal mold from the project was like $5,000. So I figured my biggest capital expense is out of the way.

7:50 >> Let's bring this product to market and figure out how you get a product on the shelf of a store. >> And so I learned so much from this. It was it's kind of like my own grad school in a way where you know I was doing everything the manufacturing the partnership the website the promotion the sales the web design the graphic design the branding the packaging it was all of these things >> customer service yeah >> all from my bedroom in Providence and what I learned is you know after that summer, I went back to Risie and said, "Hey, that was great for the graduating class. What if we did this for the incoming freshman class?"

8:34 >> A pipeline. >> 400, you know, fresh students that are going to look for would love a seat cushion. School loved the idea. Placed my first order. I tacked on my own inventory on top of that and they were distributed to all 400 freshmen. People loved it. Everyone was running on campus with these brightly colored, strangely shaped seat cushions and every be like, "What is that?" say they they really are bun. I don't know what I was envisioning, but I was kind of envisioning a I don't know those typical cushions you baseball. They are bunshaped.

9:05 I'll link a picture, but it's amazing. >> Anatomically correct. Yes. >> As we say. and so that was an exciting fall of this is going back to the year 2005. And so I had tacked on a few hundred cushions and did the order for myself. And I remember they're in my basement in Providence. And one day I went down to the basement and I'm looking at these boxes of these sea cushions and I realize nobody's coming to this basement to buy them. I have to go out into the world and actually sell these things.

9:35 And that was an incredible learning experience of, you know, walking cold into stores, finding the store manager and figuring out what is the right line, how do you frame this right way to get people excited about it. The first store I walked into was the Brown University bookstore. And I met with the the buyer, sat down, I had a sales sheet I designed, my best shoes, finest dress shirt, and in about 20 seconds of my pitch, she goes, "I'm not interested." And got up and walked out of the room. And I was like, "Oh shit." this is going to be a lot harder than I thought. Mhm.

10:10 >> but I remember walking back to my my house just a few blocks away and there was this equation that I was reciting in my head that I had heard a previous weekend at an entrepreneurship gathering also at Brown University. I snuck into as a risy guy and the person talking was about the sales process and he was sharing how rejection is a normal part of sales and some will buy your product some won't who cares move on and so the equation was SW 2 plus WC equals MO some will some won't cares move on >> and so I was like walking back to my my apartment and thinking oh man all right some are going to love it some aren't who cares I got to move on and the next store I went to they didn't buy it next store. They didn't buy it. But the fourth store, they did buy the cushion and it made it to the shelf of the store. I went back later that night after they bought it. I dropped it off and there through the window I could see this cushion on the shelf of the store and I thought, "Okay, now it's the next biggest story I can get." And the whole adventure continued from there. Another big lesson from that was that I would sell these this cushion at these fairs for designers where you'd get a small table and like in New York City at the international contemporary furniture fair every May it's like design week in New York >> and so this website called design boom would host these tables for young upcoming designers to sell whatever they they create and so I had a table full of crit buns and I got to interact with people and learn how they thought about the decision-m process. And so I literally over the course of a couple years did a few of these. I saw thousands of people. I observed them of how they came into the product, made a decision about it, and the most fascinating thing I was able to break into five steps. There's the what the hell is that phase, >> especially for this product.

12:14 >> Yeah. People would see it from across the room and they go, you see the face, they go, "What is that?" But the design was cool enough to draw them in. >> They had to know what it was. So they come up to the product. Next is the touch and feel phase. They squeeze it and they see that soft. That's when the first smile comes in their face. >> They go, "Oh." The next third is the story. They flip it over and on the back of the label, it tells the story of how the product came to be. And you see there, they go like this. very discerning look as they're reading and then they go, "Oh." And the second smile comes up as the dots connect about the origin of this product.

12:52 >> Fourth step is the demo phase. >> They got to see if it actually delivers. >> Yeah. >> So, I always had a demo chair ready to go. >> Oh, we should had some ready. Dang. >> And it's like clockwork. People go look around. They go, "Oh, demo chair." They sit the cushion down, they sit on it, and the third smile would come on their face. And so then I'd be sitting there next to them and I learned the most important question you could possibly ask in that moment, which is simply, "How do you see this product fitting into your life?"

13:22 >> And people would go through, "Oh my gosh, well, my kids soccer games, this would be great on the bleachers, or I go camping all the time. I'd love to take this with me. Oh, this would be great for my pregnant wife. Oh, my uncle has sciatica. This would take pressure off the tailbone." I heard dozens of things I never even imagined, >> right, >> it could be used for. And here people are selling it to themselves. They convince themselves why they needed this cushion.

13:47 the next question they'd ask is how much is it? That's it. 1999. And they go, " that's it. I get a sale." Now, the sixth step is the the answer to your question. I would see people sit up from the demo chair. They've got the cushion in their hand. The next person in line would go, "What's that?" And I would literally just sit there and I'd watch the person tell the story to a stranger that I had just talked to them about or they read on the back and the stranger would go, "Oh, that's cool." And they'd grab one and they'd sit down and the whole cycle would repeat.

14:23 >> Wow. >> And it was in this moment I realized that a product is 50% foam, this krypton, and 50% story. People walked away with owning the physical object, but they also own the story. And it was so it was such a illuminating experience to see that any object you have is just foam. It's just a glass. It's just a pair of shoes. It's just a phone. It's just a >> pixels. Pixels.

14:54 The other half of that product is the story that you wrap around it. M >> and the more humorous or enlightening or memorable it is, I think it's as equally as important as the battery life of something or the technical features of something, the story has to match the product, the look and feel and the materials and the capabilities of it. They're they're not different. >> Yeah. >> And to me, the best products in the world find a way to marry those two things.

15:24 >> What a what an opening answer. I have a couple of observations. One is you have spent your career since then since critons working on things with a little bit more scale. both a little company called Airbnb and now the US government. or I I shouldn't say the US government. products that probably something to the order of most to every American will use. And so what that introduces to me is two kind of interesting wrinkles on the on this idea. one is your ability to give that first person the story right up close like to hear it from you is obviously a lot more difficult and two the idea that some will like it some won't changes a lot certainly yeah there are people who don't like Airbnb and prefer hotels especially your job now >> >> I would also add that Silicon Valley in general tends to have a view that like maybe it's changing but historically like storytelling doesn't matter that much. Products, great products sell themselves. I don't think Airbnb is necessarily that type of it sort of became that, but it was definitely not obvious. So, I'm curious how given that kind of initial story, how you think about storytelling, particularly at the scale of h products that hundreds of millions of people will use. I mean, in Airbnb's case, storytelling was essential.

16:46 there's certainly the origin story which helps people understand why does this thing exist? >> How should I orientate to this? You know what I've seen over the years is everybody loves a good origin story. >> They all want to know like what was the the genesis? Where did this come from? Why does this exist in the world? They want to know the why. >> Yeah. >> And I think any great company has a great why story. Nike, eBay, Apple, right? And it's kind of personal, too, even despite the scale.

17:20 >> Sure. You know, in my experience, some of those best stories are somebody solving their own problem. >> Which is a great way to start a company, by the way, because they it's what I used to say is like you're married to the problem. >> Whatever the personal problem is that you you face the frustration every day and you can't stand it anymore, you're going to do something about it. You're not going to wait around anymore for somebody else to fix it.

17:44 >> You're going to go be proactive and take this problem head on >> and and you know it firsthand. It's personal to you which gives you insights on how to approach it and how to solve it. It also gives founders very personal connection because when it gets really hard and you may want to stop working on it, the people who who are personally attached to the problem, I think are the ones that that continue through, you know, all the ups and downs and reasons to quit. So with Airbnb, the story was obviously the founding story of Brian and I and the apartment with the air beds. and even to this day, it's so I'll be on social media and there'll be some animation that pops pops up around the founding of Airbnb and somebody's retelling our story with some AI animation or something where they've, you know, interpreted it in some way and they're now promoting it, you know, on their own. And so, I think it does have one of those qualities of being repeatable.

18:44 >> Yes. Most people have had to pay rent at some point in their life and that's a relatable thing. In some cases, the rent's gone up and people have to figure out, well, how are we going to make that rent check work? Also a relatable thing. And then the ingenuity was, well, what if we rent out this extra space on these temporary hairb mattresses during this conference? And and the idea was born. Now, the story of Airbnb has evolved beyond the origin.

19:12 I remember one of our early taglines was travel like a human and that became our story. Yeah. And it was this idea that instead of the generic mass tourism way of traveling where you're going to wake up in a hotel room that looks very similar to the last one you stayed in in a city where you couldn't tell exactly what part of the world you're in. you know potentially alone if you're traveling by yourself. versus staying in a home that has character and life to it and in most cases feels like the place that you're you're staying in. you know, it's this this idea of giving people you human humanness back that they're not just a number moving through a cattle line at the airport to get to some generic hotel that's in a box sort of lifeless and lacks personality. We used to have business travelers tell us that they specifically use Airbnb because they'd be on the road like 200 plus days a year and when you wake up in those rooms night after night, you lose a bit of yourself.

20:19 >> Yes. >> Yeah. You're nodding your head as if you've experienced this. >> I have I've done the commuting hotel thing. It is is a little tough. >> Yeah. It can wear you down. >> Yeah. >> And they stay specifically Airbnbs and specifically not even entire Airbnbs. stay in private rooms, which means the host is there >> because it gives them a sense of dignity back that they can wake up and there's a kitchen where they can, you know, make a coffee and put a bagel in the toaster and maybe have a conversation with the host before they head out to the reason they're traveling.

20:51 So, that was a very unexpected I remember that that research came in. It was business travelers love staying in private rooms. Wow, this is really interesting. You'd think that they want just the most efficient get me in get me out quickly experience, but the ability, I remember, to do laundry in a laundry machine gave them some semblance of home. Totally. Right. And so it's like enough of those little things added up where they're willing to pay for our experience versus a hotel.

21:22 And so there's these different stories that would emerge of in that case travel like a human. over the years then it became belong anywhere. that you could travel anywhere in the world and there was a host waiting for you with a home to welcome you in so that you could go from feeling like an outsider to a foreign city to an insider. And I had this experience. I remember the f very first time I was in Copenhagen many years ago and I got into the Airbnb pretty late at night. It's for a work trip for the company. We were opening our Copenhagen office. That's what it was. I got in really late. I had the whole place to myself. The host left a note on the table, welcome note, and said, "You have to try this coffee place in the morning. If you're jetlagged, this is the place you got to go." They drew a little map. It's just around the corner. And sure enough, I woke up like 4:30 in the morning, still on East Coast time or West Coast time. And got the key, put on my jacket.

22:26 It was dead of winter. It was freezing outside. I step outside. I I lock the door. The neighbor, it was like a townhouse. The neighbor next to me was also leaving at that exact same time early in the morning and looks at me and waves. I was like, "Well, that that's such a cool feeling to have. I'm not from here." >> Right. Right. >> There's somebody who lives in this building waving to I waved back. It was It was like, "Oh, that's that's a nice feeling.

22:53 >> I kind of belong here a little." >> Yeah. Then I get to the got down the street, get to the the corner coffee shop, go down some stairs. There's like a a woman with her child and somebody at the counter serving coffee. And I walk in, I feel like, oh man, I'm going to stand out like a sore thumb. I'm not a regular here, you know, American. What am I not Danish? And I open the door and they just give me the the warmest smiles.

23:20 And when my counter was was so friendly and so nice and it was these little touches that I'm walking back, you know, throughout the course of the day, I've got this metal key to a house in my pocket. You can feel it. It's a different feeling walking around a city >> with a house key in your pocket than a plastic card. >> It's a very different experience. >> You would never think about it until you felt that though. It's like so subtle.

23:45 >> So subtle. I remember another time I was staying in in New York. I come out of the Airbnb and somebody approaches me on the sidewalk and goes, "Hey, do you know how I get to the the L train? I'm trying to find the entrance to the subway." It's like, "I don't know. I'm not from here." But they thought I was. >> Yeah. >> And that was like, you know, versus walking out of a hotel in the middle of a tourist zone, some downtown city where you stand like a sore thumb.

24:12 >> Yes. There's an experience of staying in a home where you get to feel like you belong there. You don't you don't have to feel like an outsider to a foreign place. >> And so we say belong anywhere. That's where we're getting at. Feel like an insider instead of an outsider. Why would you use Airbnb or hotel? It's for that. Yeah. It's funny. Both of those stories have experience. Like it's it's not you're not really talking that much about the product. It's almost all about the experience. Mhm.

24:44 >> I think that and there's one other input that I think frames a lot of the other stuff I want to talk about which is you've spoken you gave me a great TED talk years ago on designing for trust and I think trust continue to come up kind of over and over again in my preparation as I was thinking about this and I think there's something interesting which is when you're designing for trust at scale you there's there's an old quote you said design can overcome our most deeply rooted stranger danger bias Yes, it's this kind of fascinating thing that like actually a story of a story of a story told so many times about the founding of Airbnb can get somebody to let a stranger live in their house. I was thinking about this in terms of the current work like at and people might disagree on to what degree but I think there's a common narrative that a lot of Americans don't trust each other and certainly that a lot of Americans don't trust the government >> and you are now tasked with designing services that every American every American either gets to use or has to use depending on how they see it and they need to trust you in order to have an experience that hopefully is frictionless or positive or any of these things. I'm curious how you think about when you can design for trust >> and when it actually takes it has to sort of be earned over time and and ideally you're doing both of those but Airbnb to me seems like an amazing story of that where at first it was like what levers can we pull and over over time it's like oh yeah obviously you would use an Airbnb but that I assume that wasn't instant or obvious. That was not a norm when we started.

26:21 >> in fact, I remember I told my dad about the first three guests that stayed on the air beds >> and that they were coming and he goes, "Oh, Joe, I I just don't think that's that's not a good idea." >> You know, people would have a knee-jerk reaction >> when it came to hearing about Airbnb in the early days. They'd be like, "Wow, what a cool idea. I used, you know, I went to Italy once and I stayed in the home with somebody picking me up at the train station cuz I didn't have a place to stay and, you know, had a local experience." And other people would be like, "Are you crazy? How could you stay in the house with a stranger?

26:52 >> Stranger danger." Literally like internet stranger, by the way. >> Yeah. >> Yeah. What's interesting is that Airbnb could not have launched years earlier. >> The internet had to grow up to prepare for Airbnb. Totally. >> We didn't know this at the time, of course, but I could look back and I could see, oh my god, >> you needed e-commerce to work, you know, starting in the late 90s where people could trust putting their credit card into the internet, which took some time, by the way. Like my mom didn't trust e-commerce for >> Yeah. Yeah.

27:18 >> pretty long time like into the mid200s. You needed people being comfortable sharing their location. So Foursquare that was a a big deal when that came out. People were like, "Whoa, you're sharing your location? You crazy?" >> People need to be comfortable sharing their photos online. So if you remember Flickr and Facebook in the early days, they were like, "Whoa, did you you posted your picture online?" Even real name, by the way. like ahead of all those things. I I distinctly remember it was like you know you never use your real name and then Facebook kind of changed it but yeah. Wow.

27:49 >> Facebook changed that and then the the dating sites you know the early ones match.com and Cupid. Okay, Cupid and the others. Well, you're gonna meet a stranger in person somewhere. And so you needed we needed all these things to stack on each other so that by the time 2008 came around, it's like, okay, it wasn't it was a stretch, but it wasn't >> a giant stretch to say, okay, wow, you're going to transact in line, share your location with somebody's picture, and go see them in person, and it's their home. That was the the wo right >> moment, >> right?

28:23 >> and so you know the timing really really worked out. All right. First of all, you are now in public service. and that's a word that I think to go back to our experience idea feels resonant feels resonant with being a designer. as far as I understand, you were sort of initially engaged to help out a little bit with Doge when they first got started. But I'm curious, first of all, what went into that decision, and two, what your relationship to that idea or that word is, particularly at a at a national scale.

29:02 >> Well, I've been designing websites since I was 14 in my dad's computer in his home office. And I remember the magic of the first time I published my first website. was on geio cities and this idea that the thing that I just, you know, cobbled together in HTML was now available to the entire world, not just my small town in Georgia where I grew up. And there was something about that where I was deeply hooked. I was like, "Wow, this is so powerful."

29:34 >> Yeah. that you can have this this vehicle of ideas and creativity that anybody in the world with a web browser can participate in or view or interact with. And you know, it's it's incredible from that moment to where I am today to think what what an amazing responsibility to get to bring everything I've learned about building great internet products to a part of society that historically hasn't had a lot of investment into, you know, really high standards with regards to user experience.

30:13 And that's that's why I'm doing what I'm doing is I I I feel very strongly that the United States deserves better. Don't we deserve the best? Why are we okay with subpar, frustrating, aggravating, annoying web experiences when people try to get things done in the government? I'm not okay with this. I don't think others should be okay with this. It's similar to I remember growing up as a kid in Atlanta. We'd be driving around the city going to Braves game or something and it became known to me that we'd pass like a really dull gray building and it was a government building. It just somehow became known.

31:00 No one ever said like, "Hey, look at that." It just it became known. And so I I think I've certainly grown up with such a low expectation of government websites. You know, they never really are held to the same standard of the private sector. And I think there's something wrong with that. I don't I don't I think it's an injustice to Americans. Why why should the American public deserve anything less than the best?

31:33 And that's that's why I'm here doing this work with National Design Studio. >> What did you learn in the I don't know exactly how long it was in the kind of initial engagement and a lot of the work with the retirement kind of process, but I'm curious. Doge was complicated. I think there were successes, there were things that didn't work. I'm curious what if anything you took from Doge. and what if anything you're kind of doing distinctly differently as you kind of went into the properly forming out the NDS and the new role when I came into this experience it was only supposed to be 6 months right >> to fix a very yeah no kidding I like 18 months later here we are I have to give a special thanks to Elon I wouldn't be where I am today doing this kind of work if he didn't lead the way and attract acts and bringing people like me into the government and you know I think he really put his neck out there and you know made us all take a second look at this idea of working in the government and so you know I'm I'm here thanks to him the the reason I came in was to fix a very specific broken part of the federal employee experience which is that every year 100,000 federal workers retire. It's one of one of the great perks and benefits of being a federal employee. And the process with which they go through to retire is stuck in the 1960s >> and really hasn't changed since the 1970s. That was the last time there was a major improvement made to the retirement experience.

33:14 It's all based on was all based on paper. It took upwards of six months at times. And I also think this this is like an example of an injustice. It's so complicated that federal employees had to go through two required training programs just to learn how to retire from the government hours long. >> Oh my gosh. >> It's it's such a headache. I think it's the injustice is that these civil servants give us the best years of their working life and then they go to retire to get their their benefit and they have to go through freaking obstacles.

33:46 >> Not retire. >> Some don't. I learned >> some put it off because they're like, I don't want to deal with that. So, they stay another year. So, I joked it's it's a retention program. >> Yeah. >> Because it is so frustrating. >> And so, I had a chance to to dive into this, you know, build a small team around it, some ex Airbnb engineers and designers and, >> you know, brought in Silicon Valley experience where this problem, by the way, they've been trying to solve it since the Bush administration in 2002.

34:14 They've had 20 years of contractors trying to crack this code. They spent over hund00 million over those 20 years to no avail. And so I show up and I'm working inside of OPM, Office of Personal Management. It's basically the HR and benefits department of the government. And suddenly I have to become a a quick expert on retirement. and so just like what we learned at Risie with design and just like what we employed at Airbnb to truly solve a problem at its deepest level, you have to put yourself in the shoes of the people you're designing for, which is what I did. Yes. So, we spent time talking to recent retirees, upcoming retirees, the HR teams that process them, the payroll teams that process them, so we can really holistically understand the landscape of the problem at a very personal level.

35:08 And this is, you know, the old Airbnb playbook. We used to go talk to our hosts, >> talk to our guests, >> hold meetups to invite them in so that we could learn about their experience, hear from them face to face. >> Yes. >> About what are the problems that you're facing with our service? What would make it better for you? What do you wish we did that we don't do today? And so through honing that research approach at Airbnb over 14 years brought the same spirit into retirement and we learned a lot. People were left in the dark when they went to retire.

35:43 people there so confusion about how to calculate their retirement in the first place. Nobody really knew exactly what the number was going to be. >> Yeah. >> Like what is the amount of the check that I'm going to get for the rest of my life so I can plan my life, >> right? I can plan that vacation with my grandkids I've always wanted to do or get the bed and breakfast I always wanted to buy or whatever the thing is.

36:03 >> And you're just you guys are just interviewing these people. >> Yeah. >> It's the process that >> I learned at Risie and brought to Airbnb and now we're doing it inside the government. >> you could call it user center design or there's a thousand ways to define define it. And so we take those learnings and we look at the product and the the user flow and say okay well let's let's simplify this what's actually required you know what's truly a law that needs to be adhered to and let's create an amazing user experience here and let's do it all online. Why are we using paper in 2025?

36:41 >> Gosh, how is it possible that this process is largely still paperbased? And the reason it took many months is the paper moves by mail. And so one department would get it, fill it out, they'd mail it to some other part of the country so they could fill it out. It would bounce around the country. Why was it so hard to could you? Had they tried to like digitize stages of this and there's restrictions around it or >> it was so antiquated. They would get it in the mail. They would scan it, put it on a computer, fill it on the computer, and then print it back out and mail the paper out again.

37:19 >> Gosh, >> it is so And you guys presumably had to go down this rabbit hole like or rabbit trail like see every single step. I remember we on my second week we had a couple of career folks who've been in the government for a long time in the room and we mapped out on a whiteboard >> each of these steps and it looked like some crazy calculus >> you know equation >> the meme with all the strings and >> Yeah. It was that but the retirement process and I remember thinking holy smokes this is crazy. There's just so many outdated databases and antiquated processes that no one ever really looked at from a high level and said people just over the years had been solving little tiny pieces but never >> yes >> they end up cobbled together viewed it as viewed and so >> we had a couple of the career folks around the table and one of the other guys on the project we we asked the question >> how long do you think it would take for us to take one retirement just one the basic one and move it through the system completely digitally. Meaning that we never had to print out paper.

38:33 >> Got it. >> It would be manual in that like we'll email >> we're watching it. >> Yeah. We'll handhold it digitally. >> Yes. >> How long do you think it would take for that to work? >> And everyone around the room had a you know was like some person said 4 months, another person said 6 months. Somebody in the video call said, "I'd be it would probably be a year to figure that out." And my partner on this and I said, "Well, what if we did it in a week?" There's an audible gasp in the room.

39:08 And then they sort they realized we were serious. >> Yes. >> And then they started to think, well, what if we were to do it in a week? And everybody got incredibly excited about it. And the remarkable thing about this is that they didn't do it in a week, they did it in four days. And we had for the first time ever a proof point of an entirely digital retirement that involved no printed paper whatsoever. >> And of course it was a one-off. And of course it it was not scalable in that form.

39:36 >> Yeah. You have a seed and something you could point a reference. Guys, we did that one. What if we did 10 more like that? >> And this is this is the startup process. >> Yes. >> This was Airbnb in the early days. This is pretty much any company that's you see today if you trace it back to its original >> NYC Paul Graham stuff. >> Exactly. This is, you know, go meet your host in New York City in Airbnb days, take pictures of their place, like do the things that don't scale.

40:00 >> Yeah. >> In order to learn what you're going to need to eventually build out the right technology and software to support an all digital process. >> Y >> and so that's exactly what happened. >> It's cool. We we the next couple months we built the software from the learnings. The career teams that we partnered with origally very skeptical about our presence and you know fairly quickly realized wow these guys are here to really help us. They're here to help accelerate what we've always wanted to try to accomplish which is create a better experience for retirees by making this process way less painful.

40:35 and they very quickly became you know our closest partners. So it was did it as one team and I'm really proud to say that by that June we had issued a memo out to the government which stated that all new retirements could only use our software. >> Wow. >> And then >> and this was all within Doge. >> This is Yes. This is all within >> Wow. And so then a year after that, this past June, OPM issued a memo that said there will be no new paper accepted.

41:11 >> Wow. >> To the storage facility. >> Oh my gosh. >> Which is a mine in Pennsylvania. You may have heard about they store all this paper. It has to go somewhere. There's a mine in Pennsylvania where they store paper. >> Oh, what are we doing there? So much room for activities in the mine now. >> I I know. Well, >> guess they have backups, but Wow. >> Yeah. And so we we've really and now the process through our software if you have a basic retirement it can take you a couple minutes to retire.

41:36 >> So I say from 6 months to six minutes in some cases. >> Wow. That's cool. So it was a a real success story of what happens when you take government experience and you overlay it with a private sector tech mindset with highly experienced engineers and designers. You can dive in, do the research, understand the needs and the problem set of the actual customer. look at holistically prototype you know in ways that inform what eventually became we call it o online retirement application >> and it's a huge success story of that process that's cool there's a there's a thread here that is in inside a lot of that answer which is Silicon Valley broadly not the only place but Silicon Valley especially is like yes like what is possible we can do it yes that story and DC at least from the outside looking end. But I think a lot of people would would agree kind of defaults to no like big committee. I don't know.

42:40 I heard from somebody that you guys at somebody's birthday party or something had hats that said no fake blockers, which maybe exemplifies that. On one hand, that is like the classic what people have heard about Doge like crudely. It's bulldozed through things. On the other hand, you are someone who cares deeply about experience and about craft and care and design. And so I I would love to hear you talk about kind of the tension between that sort of like speed, scrainess, how do we just get stuff done, which is clearly a even if some of the Doge stuff was left behind, clearly is tremendously valuable in the example you just gave. And understanding that like it isn't a startup, it is the government and certain things have are are going to take longer and you can't break certain things. You can break some things like that. That tension makes sense.

43:29 >> It certainly does. I'm reminded of Steve Jobs quote, "Real artist ship." >> Fair enough. Fair enough. There's a ton of tension as a designer. You want to perfect things and make them just so elegant, right? Every pixel and every, you know, rad, every detail. And if you spent your whole time making things perfect, they would they would never ship. And you know in in this context of a startup there's always some outside pressure whether it's just your funding's running out or there's competition or other that you have to move at lightning speed.

44:06 >> Yeah. >> And in this context there's a different pressure which is that there's a countdown clock every day this administration will end with it you know things cycle through. So there's an incredible amount of pressure to move quickly, make smart decisions. We're not cutting corners obviously, and find that balance between the refinement and the we have to ship gets in the hands of people. You know, there's you'll learn a lot of course when something's out there in the world.

44:45 People give you feedback. and so I think to me the line is around what are you proud of? What can you stand behind and say I'm I'm proud of that? I'm not I'm not embarrassed by that. I was going to say you seem to have a very high bar in basically all aspects of how you think about things. You you certainly don't have a mediocre bar and so that that is not that default doesn't isn't conducive to like moving incredibly fast and clearly you're I don't think people deserve mediocre.

45:15 >> Why would you put something mediocre into the world especially at the nation level? >> Yeah. Yeah. Yeah. >> Is that how we respect Americans? >> I mean, >> is that how we show respect? Is that how we say what we value? To me, design is is value. >> And the level of design that a company or a government puts into the world communicates its level of value and of respect to the people receiving that design. So why wouldn't the United States communicate the highest level of respect to the American people when it comes to the products that they have to use, not that they choose to use. It's a captive audience.

45:56 >> Yes, there's not a lot of places to, you know, enroll for Medicare and to get your social security. You know, these are fundamental things that I think the United States can increase its level of design to increase the level of respect that an American feels when they're using these products. Yeah. You you have this metaphor that I think is really beautiful where you talk about giving an Apple store-l like experience to all of these services. And while I think it's a great metaphor, it's also I have to admit like the Apple store on some level gets to choose who they serve. the DMV does not. And so in some sense, I'm curious what it's like to go from effectively directly serving customers to the American people are almost less your customer and more like your ultimate boss, but they're also kind of a captive audience and and how that changes maybe the approach, how you think about feedback.

46:55 and and and maybe I should I should add the most important point which is you are serving the widest possible demographic. I mean maybe you're not serving the whole world but at least within I would it wouldn't shock me that on some level even if Airbnb has more touch points. The US population is a wider and more distinct or excuse me wide demographic. Yeah it's the widest. And I have said when I started this group my vision was what if the government was like an Apple store like experience and what I mean by that is it's a metaphor.

47:32 What does the Apple store stand for? Stands for good usability, great design, running on modern software. >> What if our government had those three qualities? What would that look like? How respected would people feel when they entered a.gov of domain into their browser and instead of met with a confusing interface where they don't know where to start, they agree with something that was easy on the eyes. The exhale and communicate clearly and was presented in a way that they actually wanted to engage with that they could find the thing that they were looking for quickly and easily versus today's world where it's like a maze. And I think it's incredibly disrespectful that we have people, we just currently have to try to navigate a maze >> to find what they're looking for. The burden is on the American, >> right? Right.

48:20 >> To figure out the org chart of the government, right? Like which agency do I go to? What's their website? I'm on their website. How do I find the link across a sea of links and buttons to get to the form to get to the submit button to get something done in the government? I don't think that's fair. And so we're fixing that. >> As I understand it, there's sort of two shapes to the type of work you guys do.

48:50 One is things that maybe not to a complete degree, but you seem to be relatively owning yourselves and then the other is like the broad range of partnership across various organizations inside of the government. I would imagine the first one has some perks, which is you own the ship date, you can move super quickly. It's relatively vertically integrated. The second one has a totally different set of constraints. And maybe we can talk about some some specifics, but I'm curious what those kind of two modes of working are and how you think about kind of optimizing for success in both of them.

49:26 National Design Studio includes software engineers, designers, a few operation folks, research and there's two tracks of work that we do. The first track I call it the policy work >> where there's a new policy that's created and there's needs to be a bridge to the public. >> Cool. >> So, how do people touch and feel what this policy is? >> The interface. >> Yeah. What's the interface? The literal how do I access the benefits of that policy? So a couple examples one would be the new food pyramid and dietary guidelines. So USDA had a 200page PDF and they came to us and said how can you help us communicate these new guidelines >> y >> to the nation and so we got to work and we worked with them and came up with the the new food pyramid and then the website realfood.gov that tells a story and educates people about what these new guidelines are.

50:21 Then there's Trump accounts. Incredible policy where every newborn gets $1,000 into a bank account. And it's in that account until the child turns 18. >> They can take it out or keep it in at that time, but it grows interest alongside the stock market. And so a kid grows up with the ability to learn about finances and about investing, about compound interest, about the stock market, about capitalism. And so what's the way that a parent and child can interface with this cool policy and this gift from the treasury?

50:54 Well, we made the app and the website and did the storytelling around it so that people have on their phone. They can see their child's account growing over time. They can see the balance. They can see the stocks that the fund is invested in on behalf of the child. Trump RX amazing policy. For decades, Americans have been overpaying for prescription drugs and basically subsidizing drug prices for the rest of first world countries. Meaning in Canada, you can buy the same drugs in the United States, but you'd pay 10 times more in the US than you would in Canada. Exact same drug.

51:34 every president has tried to battle the pharmaceutical companies to level the playing field. They call most favor nation pricing. And President Trump in this term was the first one to successfully bring drug prices down for the first time I think in many many decades. And so how do people experience this policy? Well, Trumprex.gov is that we built the website. it's incredibly intelligent. It connects to, you know, hundreds of pharmaceutical companies to bring in real-time drug pricing. It allows people to find the drug they're looking for and have access to the coupon to show the pharmacist at the counter where they get the Trump X price. And to date, this launched in February, so this couple months, Americans have already saved $700 million.

52:26 >> That's that's insane. >> Yeah, >> that's crazy. Like if you were a startup building an app that within six months of launching seven months >> you would have you know transactions or booking value of $700 million can you that like that never happens. so the same is true with Trump accounts. We had there's over 7 million accounts that have been created for kids and over a billion dollars of contributions.

53:00 >> Oh, cool. >> Separate from the $1,000 per child. >> Yeah. >> A billion extra dollars >> put into this by parents and family members and companies that launched January 6. We're barely six weeks out from the launch. What what other app could has a startup built like that in such a short period of time can achieve that kind of scale? It's profound. >> For either of those two examples, are you guys also acting like a startup that's in charge of growth and ramping up those numbers? Are those handed off to another organization within the government? Typically like a Silicon Valley company would be like have all these dashboards on how many kids are signed up, how do we get from 60% to 80% whatever. and and be totally reasonable if that's outside of the jurisdiction, but I'm curious how that works.

53:47 >> Well, there's agencies that handle we're doing this on behalf of agencies. So, for that one, it's it's HHS. >> Got it. Got it. >> right. It's in that domain under Dr. Oz. >> Maybe there might even be a product team there eventually that's maintaining. >> Oh, yeah. There already are. >> Cool. >> So, Trump Trump accounts is in the treasury. >> Cool. >> So, you know, we're we're this complement to these >> but also kind of ignition in some of these cases. Yeah, we're initial lift.

54:14 >> we help give these things shape and I think set them up on a trajectory that the teams in those agencies run with and and we still help out with things of course they they come. >> You're way more zero to one. The goal the success in the NDS in many ways is about zero to one for at least on the new initiative side of things. How can we get them going? >> That's where we seem to be needed the most.

54:34 >> That's where we were asked to step in the most. >> Yeah. >> And we're happy to do that. You know, I I think it's been really interesting to to see how building these kind of products inside the government, the least likely place, can get scale so quickly. And it's actually become my number one recruiting line when I'm talking to designers and engineers. >> We got a lot of users baked in or potential users. >> Yeah. They're like, "Why why would I join the government? I'm not thinking about the government." I go, "Well, first of all, I wasn't either. Nobody on our team was thinking about the government at all. wasn't on anybody's career plans yes whatsoever and you know they'll tell me the other options thinking about go and that startup I'm sure it'll be fine and you'll get an experience and it'll you'll get to scale eventually if you're lucky what if day one you started here you shipped an update to fix a product and you had an install user base of 340 million people and they go I never thought about it like that yeah it's almost like not burden, but it's almost like a it's heavy when you think about it that way.

55:39 Like the potential energy there is high. >> Of course, I tell people come here if you want responsibility for that. >> Yeah. Yeah. Yeah. >> Like come here if you're excited by that scale. >> Yeah. >> And you're excited to ship at the highest standards to, >> you know, exceed the expectations of what the nation's received so far. The other thing I tell really talented software engineers and designers is that those 340 million people include your parents, your aunts and uncles, your grandparents, your siblings, your friends, your co-workers, you literally get to design for everybody in your life.

56:13 >> Yeah, because that's who we get to design for. It's cool. You said there was you started talking about the policy side, new policy side. Was there a second component or maybe to follow up on my kind of the other side of my question which is as I understand a huge part of the work too is just helping other existing government services organizations whatever modernize even if it isn't a new product. >> So the US has changed a lot since our founding 250 years ago.

56:43 We've gone from the quill to the microchip, from the cannon to the drone, from President Washington to President Trump. And despite 250 years of extraordinary progress, the greatest unrealized opportunity of our nation is the citizen experience. And how do people experience the government today? It's through 29,000.gov websites. Oh yeah, they try not to. Basically, when I started this role, it was somewhere in the magnitude of 25 or 26,000.

57:24 It just keeps going up. Gosh, it keeps going up. And at some point in history, people used to walk into government buildings to access the government. And when they did, the lobbies were beautifully designed. The investment from the government to have an incredible citizen experience is still visible cuz they're the lobbies are around. You can walk into around DC. Beautiful marble floors. >> Yeah. >> Brass inlay, fresco ceilings, great lighting. The reality is the lobby of the government today is 29,000 websites through a browser or a phone. That's what people touch the government.

58:02 And the number of people that walk into that lobby every day blow your mind. It's 40 million people >> daily. >> Daily. >> What are the most common versions of that? Like what are the sites one through? Maybe you can't share exactly, but as a ballpark like 41 million a day. 40 million people a day. I looked up the viewership to the World Cup finals last month, Spain versus Argentina, and the number of Americans that tuned in to watch that finals match. 38.9 million.

58:35 >> Wow, it's a lot. >> Every day, the government has a World Cup finals audience walking into its lobby. And what are they met with? red card after red card of bad experience, poor design, slow page load time, frustration, and unnecessary suffering. So, luckily, we know the only guy in history to reverse a red card. Oh gosh, President Trump set up this team and gave us the mandate to effectively make sure no American gets a red card again.

59:10 And so National Design Studio has been working to reimagine the relationship with Americans. And a year since our executive order, we've come up with not another website, not another way of bureaucracy, but an entirely new front door to the United States. It's called America.gov. Whoa. taking 29,000 entry points and saying, "Let's just make one."

59:43 >> Whoa. >> Say login.gov. Wow. That's a That's crazy. >> One place where you can get things done like enrolling for Medicare, managing Social Security, your student loans, getting a new passport, changing your name, all the way down the list. I was going to say I I don't even know what like the main American website would be. Wow. This is what we've been working on. Yeah. It'll have launched by the time this comes out. >> Yeah.

60:14 >> This will be one of the biggest digital product developments in the history of the country. >> Was this part of the original mandate? >> The original mandate was to improve the user experience of the nation. >> Yeah. Yeah. Yeah. >> This is our answer to President Trump's request of us. >> It's thanks to President Trump that America.gov exists. This was his mandate. This was his wish to improve the user experience to fix the issues that people have when it comes to the citizen experience of the United States.

60:44 And just thank him for his vision and his trust in me to help bring his idea and his wishes to life in the form of America.gov. >> Oh man, I have so many questions. maybe I guess the first one that might sound uninspired but I think is relevant is like how is this more than a sort of like an you think about like those old web page like Yahoo or something like an index of all the websites. I'm so glad you asked.

61:16 There's what I call legacy websites which follow patterns that go back to the 90s when the internet was developed that have a homepage and a bunch of links and a search bar that you get to get to the next page to funnel in to what you're looking for and you end up going to a user flow to accomplish something to transact at the end and make a payment or submit a form whatever it is and that these very linear experiences.

61:44 what if we flipped that all upside down? What if instead of the burden being on the American to try to manage that maze and hunt through it like on an expedition to find the link and the form and the thing? What if they could just tell us what they needed? Right. By the way, if a Silicon Valley or really any real company did this now, you would the designer would get fired, the company would fail, the users would turn. Like it's it's funny how on in some context things can be so intuitive and expected and to go back to your earlier point in other contexts we can just tolerate mediocrity.

62:22 >> Mhm. >> Okay. So it's like the front door. >> It's the front door. You just tell us what you need and we meaning the government go get it done for you. >> And and to take that literally like is it like a Google or chache style text box? >> Yes. >> Wow. >> Yes. >> Wow. Our LLM is the 29,000.gov websites. Wow. Because we're inside the government, that's our that's our LLM, which is source of truth data. And we've benchmark our search results across the other frontier models. And we objectively have a better search result. And the reason is very simple. We're only pulling from source of truth information. We're not scraping the internet. Yes. Yes. Yes. We're not going through Reddit forums and other places where there's inaccurate information and things that are out of date. Wow. Wow. So this I mean it's a in the weeds, but it you could totally imagine how something like Rampart plugs into something like this. Pretty Wow.

63:22 Okay. So this is like hey I don't know I got this tax thing or I got the social security thing. You just you just straight up tell I mean also to be frank I would imagine many Americans for better or for worse are telling that to Chad and Claude and Gemini today in a way that is at the very minimum security nightmare. Wow. What what is the kind of the initial formulation of a product like this? Like one how does how do you figure out if it's possible? Two how much like backend indexing do you have to do? I mean, yeah, we could talk about this for many hours, but maybe a better question is when did this idea first show up and how we talked about this last October, >> okay?

64:02 >> And I remember writing something down in my sketchbook related to this and it was still a bit of a stretch at the time of like, wow, you know, how would you pull that off? And you know that the models are still evolving and you know but this this you know platform is evolving so quickly at a certain point not that much later everything clicked and it was like wait a second we can embrace these frontier models in a secure way of course >> that's safeguarded around the government to deliver an experience that you can't get anywhere else and And the reason is this. Because we're from within the government, it gives us the ability to tap into databases across the government. So, not only can you search for a question you're trying to understand about the government and get a result faster than you can snap your fingers that's higher quality in the frontier model, you can also action it right in the thread.

65:06 >> Right. Right. And so if you go on Chat GBT, great product, and you say, "What are some good vacations in Italy, it'll pull up Airbnb listings and hotels and flights, but to book any of those, it's a link that takes you outside of ChachiBT." And and maybe even it gets more agentic, but that information, my hotel booking information in Italy is very different than my social security number. That's right. One maybe less sexy but I think important question this brings to mind is for a lot of people it's going to be as you say it's like kind of how people use Chedg it's like asking Google a question I need I need to figure out how to get to this whatever assuming the second part of what you're saying which obviously is where the potential lies people I said it once earlier people many people don't trust the government many people don't trust this specific president and and administration there's that whole spectrum but one, people don't trust the government, and two, people don't trust AI.

66:10 How do you increase the likelihood of people really getting the upside of what you're describing in a way where they have a baseline of trust and maybe that's storytelling, maybe that's technical. As I understand it, what you're really describing is like an agent. What could become an agent for how people work with everything related to the government, which could be so magical. I suspect most American I'm I'm a tech guy. Most Americans are probably gonna be like, I don't want to give the the government's AI. the first part of your question sounds very familiar to the early days of Airbnb.

66:44 >> How are we going to trust that? I don't trust that company Airbnb. I don't trust that host that's advertising their home on it, their platform. >> And in some ways, there's, you know, a parallel of having to navigate through a gap in trust. >> Yeah. And if there's any lessons from Airbnb, it's using good design to reassure people that what you're engaging with is trustworthy. You know, it's like when you see it's the broken windows theory. You know, if you come across a website and it's kind of slow, doesn't really load.

67:21 >> Yeah. >> Things aren't really lined up and you can tell it's like somebody didn't spend a lot of time on this website. It it it it's rickety. >> It was rickety. Yeah, sure. Janky, rickety. It's wobbly, flimsy. >> Yeah, >> it undermines trust. Well, what else is flimsy about this website? How they handle my data, what they do with my credit card. >> The way something's designed is essential to communicating trust. >> I've been asked over my career a thousand times, Joe, what do you think good design is?

67:57 And I actually that went down. >> Yeah, I I have a I have a new definition based on this experience at developingamea.gov. Good design is when a user actively thinks or says out loud, "Wow, somebody really put a lot of care into this to make my experience better." >> Yeah. Yes. Bang on. >> That is good design. That is what we're delivering with America. I want the nation to have that experience. where they feel the 16-hour days we're pulling over in our studio with some the brightest engineers and some of those talented designers to craft an experience that gets our nation to a whole new level of user experience. One that respects the American, gives them dignity back so people don't feel like an idiot when they get lost in a government website.

68:56 Something that respects people's time to give them time back in their life. Something that people can take pride in and say, "Wow, my government really did that. Wow." Like, "Yeah, heck yeah." And and and the fourth thing is they can feel the sense of wow somebody really cared enough to put in this effort to save me time, give me dignity, help me feel pride in my country, make my life easier when it comes to getting stuff done in our nation.

69:29 It's hard to imagine. I I'm very very excited to see it. It's like but it is so far from the default, right? on some level it's funny. I I love that description of good design and I also think there's a kind of similar version of good design that is much more commonly stated which is like it's invisible. I was thinking about this as I was prepping like in some sense certainly more than the average company or product you are it's it sounds like the wrong way to say it. almost trying to design things that are kind of boring.

70:09 they're supposed to be they they should be at le when I think about like a great government tool or website. It's like really reliable, easy to use and also kind of boring. like you're not trying to dazzle people necessarily or maybe maybe the what I really mean by the opposite of boring is like you're not focused on engaging me and keeping me using this product. I want to get what I want and get it done and get out of there. Mhm.

70:33 >> dealing with LLMs broadly, like LM are not deterministic. It's like the feature in the bug. how do you think about like broad-based reliability and testing and sort of like consistency for you? You want it to be this like super boring reliable text box that does exactly what I want and nothing more. Mhm. We work very closely with the agencies that we partner with both to integrate actions so that when it comes to renewing a passport, you can get it done right inamea.gov.

71:06 >> Right? So when it comes to comparing Medicare plans and then picking the one you want to enroll in, you can get it done right inamea.gov. You don't have to leave the site to go back into the maze >> y >> of >> government websites. So, these backend integrations are what makes this so magical. And it's interesting enough, it's something that the frontier models haven't had to deal with yet because who are they integrating with? It's like they cover the whole world. We just cover the government. So, we're vertically integrated.

71:39 >> Yeah. Yeah. Yeah. Maybe there's a passport app that is really reliable. Doesn't even use LMS. It just plugs in. But you're route. You're really the job here is you're the ultimate router. >> Yes. We're the ultimate front door. We don't store anyone's information. It lives in the databases where it already exists in State Department or HHS or Medicare. We're just a front door that allows people to access it, manage it, action it, and we send it back to the agency. It's certainly in the technical weeds, but can you speak a little bit about Rampart, which as I understand was kind of like a side project? but either that specifically or that philosophically it's easy to say oh we don't we we're just passing through your data or whatever like why is something like that one important and and what actually is it rampart wasn't a side project rampart was preparation for american.gov of Rampart is an open source model that is for the mobile web view because we we knew that people would be using an LLM on their device on their phone and so what we want to make sure is that if people put in personally identifiable information PII that it's encrypted and trustworthy and it stays on the device.

72:52 So if you search Rampart open source model privacy on the internet, you'll come and you can read about how we've programmatically figured out a way to keep people's PII on the device in a web browser, mind you, too. Right. >> Correct. Which is cool. >> And the the size of the file is small enough. I think it's about 12 megabytes. >> Wow. >> That allows it to be on a device. You know, the other Frontier models have something similar, but I think they're in the gigabytes of size, which is completely unrealistic to ever get on somebody's phone. So that's what that's I think that's why it's trending on hugging face right now.

73:28 >> We've been talking a lot about websites websites for a long time and I think for the average person especially the experience on a phone like they're not really using websites as much. Everybody has an app. We're past the app era in some sense. >> What is so powerful about websites? >> Yeah. I mean what makes websites amazing is the they're so universal. any device, any screen size, any operating system, you can reach the world. >> Yeah. In some sense, it almost feels like in a in a non-intuitive way, it's respecting me cuz it's light enough to do exactly the job.

74:01 >> That's why it's I think it's so cool that you guys can do the the the private model stuff through a browser. It's pretty amazing. >> Wow. America.gov. So, you're done and like this is out. Oh, it's this is just the the you know version one. There's many versions ahead. >> This is just we're just scratching the surface of the capabilities of this platform to really deliver convenient experience for for Americans that I don't think anyone's ever really seen before.

74:31 >> Yeah. >> When it comes to changing your name for as an example, this one comes up all the time. I've met so many women through this project who confided in me very quietly. By the way, I still haven't changed my last name. I've been married for years. I go, "Why is that?" They go, "It's just it's too complicated." Yeah. I can't figure it out on the websites. And they say with shame, too. And it's like they're kind of like the subtext is I couldn't figure out a website.

74:58 You know, those days are over. We can integrate with the agencies required to change your name and and a couple keystrokes with, you know, validation of your marriage certificate. we can we can agentically go change your name for you rather than the burden being on you going door to door to each agency filling out paperwork mailing it in waiting months etc those days are over you know I really think as americgov matures and the platform builds out more there'll be a certain point where people won't go back to the old way I was going to say between now and then this is an amazing huge launch you've wor also worked on a bunch of other presumably are going to continue to there are 29,000 different services surfaces I should say.

75:46 even if you get amazing at routing I'm curious how you think about back to this countdown clock prioritization across various different things. Again maybe someday it gets to the point where I don't even have to go to any of those individual websites. It's all done agentically. In the meantime, you're balancing new launches like whatever the the food pyramid or or Trump accounts, big long-term projects. You have two more years. You have two and a half more years. Is there like a preference stack? Is there like a is it like websites numbered one through 150 that we're going to tackle? Will some of that start to get automated using AI tools or otherwise? Is it about setting up design systems?

76:30 The easy answer is we don't have to redesign any websites. america.gov is the front door to all of them. >> So let's take an example. The the name change. I go to america.gov. I need to change my name. >> What happens? >> you would put in, hey, I just got married. How do I change my name? >> Yeah. >> A response would come back and say, congratulations. And go log in through login.gov of and we'll find all the places in the government that you have your name already exists and we'll show you and then you can opt in to the ones that you want to change your name.

77:08 You'd click next. The component would update. It would confirm your existing name. You put in the new name. You put in the reason. You hit next. If you said marriage, it's going to ask you for your marriage certificate. So, you either upload it from your computer or hit the QR code to take a picture of it on your desk. output it from your phone and then you hit done. You can my my jaw isn't on the floor but okay that is amazing.

77:35 I believe that that will be possible in some circumstances. It seems impossible to me that at least upon launch that is broadly true across anything I might need to do with the government. >> Oh for sure on launch a lot of these capabilities won't be available. These are require deep integrations through >> but so so maybe maybe that's the right way to ask the question even when it comes to integrations passport change TSA prejack change my name IRS madness are you then is the would the is the plan of attack sort of like go down the biggest surfaces and do the agentic integration how do you think about well a lot of what we want to do in America.gov of agencies have been trying to do already.

78:15 >> I see >> it just you're building almost like an API that the agencies can tap into >> this and vice versa. They're building APIs for us or have existing ones that we can accelerate a lot of what they're trying to accomplish anyway. >> Wow. >> And do it in a way where the end experience for the American is simplified, easy to understand and manage and ultimately convenient. >> Wow. It's a it's a amazing world to imagine. It can't can't come soon enough.

78:45 >> Yeah. Look, version one of this will have basic functionality. >> Yeah. And over the next two years, we're going to keep integrating and there'll be more and more integrations and more and more actual you know implementations right in the chat so that you know I imagine two years from now when a student wants to get a student loan and somebody really wants to manage every component of their social security and somebody's looking for HUD housing and somebody wants to really do you know the endto-end experience of Medicare, they'll actually be able to do do it all in one place >> securely, safely, in a way that respects their privacy, secures their data, and and ultimately gives people respect.

79:30 Also, none of this would be possible without Susie Wilds, the chief of staff, believing in us and me to lead this work and the directors and the cabinet members across all the key agencies that have already integrated to America.gov or will be integrating into America.gov. This convenience that's being delivered to the American people is only possible with their partnership and with their help. >> A lot of people, I would imagine, a lot of people. You spoke a little bit about the opportunity when it comes to people working on this.

80:06 How have you actually tactically brought in the level of talent needed to build something this ambitious when it comes to to be frank working for the government not not everybody's dream? >> Yeah. No one on the team had any plans to be in the government and the the scale of what we're doing and the ambition has attracted some of the most experienced designers, most experienced engineers, most experienced researchers that I've ever had a chance to work with.

80:34 And it's it's for somebody who who likes to scale and they they like the responsibility. They like the high stakes that this these systems have to work. and ultimately I think deep down they like this ability to serve. They like this idea that they can take what they're really experienced at software engineering design research and actually direct it not towards a you know for-profit or nonprofit company but towards the infrastructure that runs our nation and they can actually put their fingers on and stamp their name to one of the most massive transformations that's ever happened >> in this country from a citizen experience standpoint.

81:20 And this is why people are joining. This is why people are leaving highpaying jobs and >> >> you know great positions and great titles to come contribute to this. And in some ways it's like you know there's there's a parallel to the Olympics which is most Olympians leading normal lives. Then the Olympics come up they they leave their job and their family. They go train somewhere for certain period of time >> away from their home. Then they go travel to where the Olympics are and compete and they have this incredible life experience where they're operating at the best of their talent and their sport and their competition. At the end of the Olympics, they go back to their normal life.

82:04 >> Yeah. This is that experience for software engineers and and designers. >> That's intense. >> Yeah. Come work with the best in class. >> Yeah. >> On the biggest problem. >> Yeah. One of the exciting things about getting to do this work like America.gov is it's one of the most apolitical things you could work on. And what I've learned here, one of the biggest things that surprises me in the government is how unifying websites are because everyone can agree >> totally.

82:35 >> They wish government websites worked better, looked better, and loaded faster. It truly unites the country, unites the left and the right >> around a single common cause. And I think what we're doing for a designer out there who's watching this or for a software engineer who's now thinking about, wow, I never considered working on such a project at this scale inside the government. it's important to know that this is one of the most apolitical things that anybody could possibly do. It's for every American, north to south, east coast to west coast, young and old, men and women. Doesn't matter who you voted for.

83:11 This is about improving the citizen experience for every citizen. >> Right on. Do you do any work in the category that you would broadly call brand for the US? >> Not directly, but I remember something Jeff Bezos said once that a brand is defined by what people say about it when you're not in the room. >> And so, are we working on the brand of the United States? >> In that context, yes, we are. >> Yeah.

83:42 great brand in certain ways and websites they're on the up. I was thinking about how to talk about sort of the challenge of doing this thing on a national scale that everyone has an opinion on. You brought it up at the very beginning. as I understand it, I didn't go to art school, but a huge part of that culture and particularly risy is crit. and this idea putting your work on display, sort of like taking the arrows as they come.

84:08 Obviously, Silicon Valley has a different culture of that, but certainly like very, very tight feedback loops. you have a big tent of what you would call feedback and criticism today. and that is colored on this kind of extreme level, which is that a lot of it is politically flavored. So, there are certain people who think everything you do is amazing, and there are certain people who think everything you do is the worst thing ever. Maybe to start, I'm just curious for you to reflect on year one, I think, at the time of filming, your first year, and your birthday. happy birthday recently passed. what do you think went really well in year one? What do you wish you had done better or are you looking to improve in I suppose?

84:48 >> Yeah, year one was recruiting was timeconuming for the first couple months. >> Yeah, >> you know it had never existed before and so just like a startup you got to convince people to join your team of which there's no precedent. They had a lot of baggage. >> A lot of bag. Well, you know, baggage and dynamics. >> Yeah. >> And so the first couple months was really figuring out how to recruit for this unproven promise that we would actually have authority to get stuff done, that we working on some of the biggest things possible, that really experienced people would be joining this team, which they have. And so, you know, the engine got going in this first couple months and eventually it's, you know, been racing down the road ever since.

85:39 you know, I wish we come up with America to go sooner. That's one thing. I think it, it took a little time to match the right technology and the right stack and >> yeah, >> do all the testing and really proof of concepts. Just like we took that one retirement and push it through the system digitally in a week or in four days, so too do we try American.gov out. It's like how do we prove this concept quickly and give ourselves a reference or a new baseline of what's actually possible?

86:06 >> And and I was going to say part of that is even believing it could be possible on this scale at this time. Yeah, that's that's cool. >> Yeah. And there's rigorous research that goes into all this. tons of testing on the back end, tons of security testing, like you wouldn't believe. more than, you know, I've ever seen before. It's it's incredible. and and doing something that I learned from Airbnb actually, which is I remember we were in the apartment when we were 10 10 employees. We're in the living room. We had one long table. I'm sitting here across from me is Nate, my co-founder. Next to him is one of our early engineers. And the early engineer gets a phone call and we're all seated at the table. We can over hear the call.

86:49 He goes, "Oh, hey." Yeah. Yeah. okay. Go to the top right. Yep. Okay. Now, click over. Yeah. Click where it says profile. No, no, no. go back to page. Go down. No. No. Mom. Mom, go back. And this went on for far too long to the point where he got off the phone and we all looked at each other and we go, "Guys, if our moms can't figure out how to use our website, >> like we're never going to go mainstream.

87:18 This will never become a mainstream company." >> Wow. >> As 10 guys sitting in the living room with this big dream, we have an artificial ceiling on our growth if our you user interface is not clear enough for our moms to figure out without calling their techsavvy sons. Yes. >> And so we employed this idea of our site's got to be mom friendly. And that drove our design. It was a, you know, subtext of our design for a very long time. I'm sure it probably still is at the company in some ways where there's this universal usability where things have to be so clear and so intuitive that, you know, a mom could figure it out. In fact, I remember years later in our much bigger office, it's a much bigger company. We used to celebrate Mother's Day. And so, in our, food service area, we had, we had all all the moms of anyone any employees in the company come visit. And we actually set up a a research station.

88:22 >> Wow. >> Where we could >> My mom would have gone crazy on that. She she got opinions. We would put our latest designs in, you know, prototypes in front of moms there for for brunch >> and we got actual feedback from moms and, you know, through different user testing over the years, we would do that. Now in the government, it's different because it's not just moms. We've actually upgraded and we operate the team that has to be grandma friendly, which is true.

88:51 And so this is the bar that we hold ourselves to. And with the things that we make, start certainly with America.gov, we actually have a research cohort of mothers over the age of 55 that we put our work in front of. >> Wow. >> For usability testing. >> What does that what does that practically look like? Is it like I don't know 10 of them in a room at a time and you've got your team in there like >> we use a service to facilitate it and so they you know recruit and they find you know these mothers over the age of 55 >> and we do research user research beyond that. Of course, >> there's something to say that if you can, you know, have your website working for a non- techsavvy grandmother or grandfather, that you've you've really reached some high high bar of usability, of clarity. Yep. in fact, you know, accessibility is a very important thing for us. All our sites are accessible as they need to be.

89:58 but there's there's an element to accessibility that's not required that we actually go the distance for, which is accessibility includes a website that's clear and that people can comprehend because you can have an accessible website with really confusing navigation and really confusing language and copy. That's it's not required to do that. You can takeick all the other boxes by compliance standards, but that's not enough for us. our like what we put into the world, it's got to have a level of clarity to it so people can actually navigate it. And to me that again comes back to respect.

90:38 I I love that answer. One of my questions was going to be like especially given all the political stuff, how do you sort of actually like just seek this sort of like classic truth grounded feedback and like what a cool what a cool frame the default that particularly considers the longtail of Americans from a technology standpoint as technology increasingly is part of our lives but not always a thing that people love in their lives is is important and cool.

91:06 You mentioned the clock. You have started many companies. You've built an incredibly incredibly successful scaled one. that involved all the classic things and managing money and fundraising and all that. People who have started companies, especially ventureback companies, have a relationship this idea of the time ticking. but it's very different here. there there's nothing that you can do that is going to have this two and a half year timeline at least go away. There might be questions about what happens after and it's worth talking about.

91:45 what is the right frame and disposition around that? Like I guess part of it is just keeping it in mind but like you have two and a half years left. Is it a mad dash to cover as much ground as possible? is about investing in something like America.gov, America.gov and having it stand the test of time. Maybe it's worth adding in the second question which is like with that timeline in mind, how much are you thinking about endurance and continuity and the idea of whether the be these products or the NDS itself enduring past your your term?

92:21 >> I think the answer for America.gov of will be people voting with their feet. It's but people choose the best product that they want to get done what they're trying to get done. >> That's also quite uncommon with with government tools and websites, right? >> Yeah. >> We're creating a new option for people. >> Yeah. And it's there for them to use it or not. If they want to use, you know, traditional method, they can. It doesn't replace any website.

92:52 It's just an additional option to create a more convenient way, time-saving way, >> right? >> User friendly way to accomplish what they're trying to accomplish, to find the answer that they're trying to find, to get the information that they're seeking to make a decision and lead their life in the best possible way. Also, maybe a a a pretty strong case regardless of who wins the presidency and otherwise, a pretty strong case to keep something around is that a whole bunch of people use it.

93:23 beyond that which I I I don't I don't want to trivialize I think that matters a ton more broadly the work design systems and and very specifically this is you're the first chief design officer of the US the national design studio is a really ambitious idea regardless of nitpicks people have or how much people love it the idea of something like this existing as you've talked plenty about is I think pretty remarkable. >> >> do you have any thoughts or please or ideas on its future and and how you would hope that it could continue as particularly maybe like independent of the politics of it all. It's hard to separate. Right.

94:08 >> Well, if anything, I hope I've I've kicked k kicked the door open and shown that design can improve society at this scale if you find the right type of experience level to bring in. We're here on executive order and those you know come and go at the the whims of whoever's in the office next. I hope that we'll have proven that design has a role in the White House. Design has a role in the government. the highest possible level to help inform and introduce new policies to the public, improve things like access to government services through something like America.gov. Just make it simple, make it user friendly. remove these stigmas that people just automatically know that government websites suck.

95:04 let's just get rid of that. I don't think any administration going forward has to accept that anymore. And I hope we've proven that. >> What do you think would be most surprising to people about working in government? >> I'll speak for myself. The most surprising thing is how much fun I'm having. I I never would have imagined that the people I get to work with, the cabinet secretaries, Susie Wilds, the president, directors of agencies, been so receptive to our services and to coming to understand the value of good design as it relates to what they're trying to accomplish.

95:44 AI and design. probably not that relevant for normal people, but back in San Francisco, people are talking about, thinking about it. >> Yeah. I'm just curious if you have any reactions based on what you've seen so far, what you think is going to happen. In some sense, >> it feels like to go back to the thing, it's like somebody really cared. >> That's not mutually exclusive with using amazing tools to go faster and do more. Product development as I had known it in my previous life is over.

96:14 >> Whoa. It's done. Gone are the days where you have this waterfall process design. You sweat through every pixel. You finish the mockup. You hand it to engineering goes to some front end devs that whittle away at it for, you know, days or weeks. Maybe build a prototype in there somewhere, test it out, then back in devs, you know, connect it to databases and other things. I've been in meetings where we're discussing an idea and in the meeting it's already built.

96:54 It's done. Not done, but it's brought to life in a way where like the meeting has shifted. Now we're talking about the thing we were talking about 10 minutes ago, but looking at a screen and clicking through it and the speed with which engineers can build changes everything. >> Yeah. Not to mention the speed with which engine designers can bring ideas to life through similar tools. But the the cost of engineer software development is plummeted plummeted.

97:28 And I don't we're never going back, right? Like it's everyone's using AI tools, right? It's like in this the speed with which they can code is is phenomenal. So those those old days >> Yeah. are a >> where do you think the care comes in? >> I think in the polish, you know, vibe coding is one way to move very quickly. it's not a way to ship a rigid product and so the care comes into sometimes you got to go back and hardcode it.

97:59 >> Sometimes you end up going back to the the old days of how you would >> create something. The benefit is you got to realize it sooner with lower cost investment to bring it to life and realize, okay, the idea actually wasn't as good as we thought. Let's evolve it. And so in the same span of time that you would have built one thing in the old days, you can now build 15 versions of it and you've already iterated to a point where like you're ready to like really double down and invest in it and then you can go, >> you know, really make it rigid. Yeah.

98:30 And really clean up the code. It's about the extra detail. >> Get it for ready for production. Exactly. >> any open-ended, but any thoughts on why the EMS is were so inspiring to you as a designer? >> The the EMS were inspiring to me on on many levels. one is that >> >> one of their quotes from Charles Emmes is that the details aren't the details, they are the product. They cared so deeply about a holistic design that, you know, does have quality and does have that level of perfection. And I think the coolest proof point about them is that their designs are still relevant today >> where the things they made 50, 60 years ago are still as treasured and still as relevant and still as robust. I have in our office here EMS chairs from the 1960s and my, you know, our engineers who are in their 30s or the 20s, they can't believe it. They're like, "How are these chairs from 1968?

99:39 It's the EMS weather management chair." >> And I like to point it out to everybody as an example of good design lasts. I think the best design, if you want to talk about the most sustainable design, it's the thing that becomes an heirloom that you both treasure and value and you want to pass on to somebody, but it's also built in such a way that it can endure and it can last. >> Yeah. >> Over time. It's not flimsy.

100:08 It's not cheap, right? There's a quality to it that allows something a chair that we're sitting in today in 2026. was 1968 inspiring for how we think about digital stuff. Certainly the digital maintenance, digital durability is is not held to the same standard. >> Just a few final things. I'm just curious if you look back anything that you particularly learned from or were inspired by from Brian and Nathan over those all those years building Airbnb. I got incredibly lucky with my co-founders.

100:46 There's a lot of stories in Silicon Valley of founding teams that don't exist anymore. You know, some drama, some quarrel, some disagreement, and things imploded. Brian and I still to this day have a friendship and a brotherhood that hasn't changed. And and throughout the company, we would refer to the three legs of a stool. And that if you removed any one of those legs, stool would fall over. And when I look at the success of Airbnb and how we endured those early days, which were really difficult, a lot of forces against us, there was this notion that, you know, we had a triangle, right? One of the most robust pieces of geometry.

101:37 and in fact we did a personality test one time where some consultant came in and did the whole leadership team and they would plot your results in these concentric circles and each quadrant of the circle or you know section of the circle had you know leadership and creativity and self-awareness and they plotted ours on the circles all three of us and we made an equilateral triangle. The facilitator was like shocked. She's like, "Never seen anything like this before."

102:09 >> Planned it. I'm just kidding. >> That's awesome. >> Yeah. So, there's something really special about the two of them. Of course, we didn't always agree on everything, but because of the respect that we had for each other and the friendship, whatever we were trying to make a decision on, we we'd net out somewhere in the middle of the triangle, right? which was typically a very balanced and well-honed decision about where we taking the company. >> Brian's done a phenomenal job as a CEO.

102:38 >> Nathan's done a phenomenal job as our, you know, technical founding technical partners, seeing the company through growth and now running strategy and supporting the executive team and a lot of different initiatives to the company and it really could not have happened and I think all three of us agree with any of the two of us. M >> it took the three of us to bring this you know unusual concept >> into mainstream >> quite a heavy lift it took three guys.

103:07 >> Do you think outside of block do you have any advice or thoughts on finding great collaborators like that? I remember when I was at Risie during orientation, the guy on stage, the orientation leaders, looked out to the the auditorium to all of us freshmen and said, "Look to your left, look to your right. These are your future collaborators, your future employees, maybe your future spouses." And the message was to spend the next four years getting to know people like creating relationships. And I remember thinking that moment, I was like, "Oh, wow.

103:42 They're right." And during the course of my time there, I had a sense that I wanted to start a company one day. I known that since I was a kid. And so my radar was always on in the background of meeting people, starting friendships, and building relationships. This is somebody I'd ever want to start something with. And by the end of my time there, it's a very short list. and Brian was at the top because we had experiences with each other where I saw how we compleimemented one another. We have different personalities, different styles, ultimately both care about design and creativity and and we had this quality where we were on a project together. We got paired on an assignment to redesign products for this company called Conair, which makes like hair dryers and curling irons. Your mom probably has >> Conair.

104:36 And so the summer assignment him and I went off and we we went so far outside the bounds of the brief. We went so far beyond designing like consumer products. We were designing Kanye or the company. We were coming up with a new framing for the company. I remember we went to their headquarters in Connecticut >> to present to the head of marketing and a bunch of these guys directors. and we get up there and the look on their faces who the hell are these two kids? But there was something in that experience specifically where I remember Brian and I, we could come up with big ideas together. He put us in the same room and we' bounce off each other and we would end up at a place that neither of us might have got to on our own. And this also manifested in our epic ping-pong matches which became a metaphor for us which is we ping pong ideas back and forth.

105:34 >> We make each other better. >> Yeah. >> And both of us got really good at ping pong the best >> by playing each other. >> There's like this healthy competition. And so I think within that story there's qualities of what people should look for. >> Yeah. >> Someone you want to spend time with. >> someone who you compliment. They can't be exactly like you. They got to be kind of cover your bases on, you know, your strengths are here and your weaknesses are here. How are they complimenting you?

106:00 And I think ultimately like a deep respect for each other >> for your strengths and weaknesses. And that's what Nate, Brian, and I all had together. We were all brought something to the table that the others didn't have as much of, and the others filled the gaps of where we had deficiencies in. Bing bong. One last thing from your Airbnb resignation letter or not resignation but farewell farewell. the data proves that the golden rule is actually human nature. In other words, people are good.

106:34 How how has working at this scale affected that view if at all? I wrote that because the numbers of Airbnb proved it that the sheer scale of people who have stayed in strangers homes over the last now 17 18 years. Wow. It's been that long. It's been that long. You know, of course there are incidents and trust and safety things, but the the the fraction of the bigger picture and you kind of look at the numbers and you go, well, it's objectively clear here that by and large people can trust each other if you create the right vehicle and platform to exchange that trust with other people.

107:24 >> Yeah. All right. What are you going to do in 2029? What's next? >> Great question. >> You have a few things to do before then. >> Yeah, >> Joe Jebia, thank you very much. This was an honor. a little bit outside of my normal wheelhouse and very cool to hear about the thinking and I think yeah it's just there's a broad bucket of politics and outside of that what I think government reminds me of and you talked about some of it is it's an it's it's a reminder and an inspiration to think about how we can serve think about how we can serve the people right next to us who we know really well or like us and the grandmas us. and so I I admire it a lot. Thank you.

108:09 >> It was an honor. Thank you. >> Thank you for listening to my conversation with Joe. All of the links in the transcript related to this episode are available in the description at dialectic.fm/joe-jia. If you enjoyed it, please share it with a friend. And I'd like to thank Notion one more time for presenting Dialectic. Being able to work across all the AI products and agents you use makes Notion the ideal place to run everything through. That includes any of the agents you're working with, including new ones like Muse and Instinct, as well as every AI model the day it releases. So, anytime there's a new model from Anthropic, OpenAI, Google, Grock, and others, as well as openweight models, you can use them with all of your context right in Notion with your teammates where your work already is.

108:52 Notion also recently launched the skills API, which allows you to share skills with anyone on your team and allow them to use AI tools just like you do. This can be especially helpful in helping the power users on your team give leverage to everybody else. Notion has also spent a ton of time recently speeding up the product across web and mobile, including a rebuild of the mobile app. And he's doing the same thing with Notion Calendar as everything moves faster and faster, having one place for you and your team where you know your contacts lives, that works with every tool and agent you use, and that is a reliable place to see the work as it's actually happening is a great way to futureproof your work and your process. You can learn more at notion.com/dialectic.

109:33 And thank you to Notion for their continual support of me and the show. Thanks again for listening and I will see you next