 All right, welcome everybody. Thank you for joining. With so many of you, essentially being new to the community today, I wanna take a minute, introduce myself and the event series that we have here today. My name is Eric Anderson. I lead operations here at round and this is a roundabout. And for those of you that haven't attended a roundabout before, it's a really great series that brings together some of the world's foremost thinkers and doers to lead conversations on some really relevant and hot topics shape in our world. And the whole point about this is to bring these topics to the round community so that we can build and have better informed perspectives that are helpful for us as we lead within the tech industry where these topics are really relevant. And so today we have a really fantastic discussion. We'll spend about 30 to 40 minutes with our guest today and we'll have a conversation with Audrey Tang and we'll keep the remaining time available for Q&A. So as we go through the conversation, there is a Q&A chat on the right hand side of your screen. So I encourage you to drop your questions that you have into that Q&A function. And in addition, there's a chat function there as well. And so encourage you to engage within the chat let us know what's resonating, what things are super interesting and so on as you go through and hear the conversation and you can engage with the other participants in the audience as we do that. In just a moment, we'll bring our guest in. I'm super excited for our guest today who's Audrey Tang. But before we do that, I want to just take a minute and share a little bit about her fascinating background. I have a bio here that I'll read through and just share a little bit about why she's such a relevant person to come talk to the round community. Audrey Tang, she oversees Taiwan's social innovation as the digital minister of Taiwan but is also a super accomplished software programmer. No snow worthy, she revitalized the computer language Perl and Haskell and in collaboration with Dan Bricklin she built the online spreadsheet system EtherCalc. She worked in many jobs, computational linguistics consultant with Apple on crowd lexicography with Oxford Press and on social interaction design with social text. She actively contributes to GovZero which we'll learn a lot more about today but it's a vibrant community focusing on creating tools for civil society with a call to fork the government and again, we'll learn more about what that means. And she's no stranger to the stage. Her work has been featured at TED Talks in Wired Magazine in the Economist and many more places. So join me and please welcome Audrey Tang to the roundabout stage. Good luck coach. Hi, Audrey, welcome and thank you so much for being with us today. Really happy to be here. Yeah, it's, we're super excited to have you. Number one, the work you're doing is super impactful but I also think it's really fascinating from a tech standpoint and I think the audience here will find that super interesting. But before we get into all of that, I really wanna start with your role and your title as digital minister of Taiwan. And when I learned that you're gonna join us and I saw that was your title, I went and looked and I don't think I've seen any other government in the world that actually has a digital minister. So it's kind of cool to say, not only are you a digital minister in Taiwan, potentially the only one in the world, but also I learned that you actually had the chance to write your own job description when you did that, which is something that when we think about our jobs, we all dream of getting to come into a job and write exactly what that's going to be. But you did it in an interesting way, right? Instead of sitting down and writing a typical job description, you created a poem. So I'd love it, we're gonna bring that poem up and I'd love if you would recite that poem for us, which I think is really powerful. And then tell us a little bit about what you were hoping to accomplish as the digital minister of Taiwan. Thank you. Yeah, the poem or the prayer I wrote in 2016 because Taiwan didn't have a digital minister and so I get to write my own job description. It goes like this, when we see the internet of things, let's make it an internet of beings. When we see virtual reality, let's make it a shared reality. When we see machine learning, let's make it collaborative learning. When we see user experience, let's make it about human experience. And whenever we hear that a singularity is near, it's always remember the plurality is here. It's a plurality, collaboration across diversity. That's the work that I do in Taiwan. Amazing, super powerful and a really cool way if someone asks you, hey, what do you do for a job and you can actually recite a poem back to them. I think that's really neat. So as this digital minister, if that's your job description, what were you hoping to accomplish when you were appointed back in 2016 into this role? Mm-hmm, yeah. So I actually worked as a reverse mentor, a young person younger than 35 to advise the cabinet for a couple of years before becoming the digital minister. During those two years, we worked with, for example, Uber when he first came to Taiwan, made sure that Uber now is a Taiwanese taxi company, but the innovations, for example, the surge pricing and the dynamic dispatch and so on eventually benefited the local temples and churches to serve the people who previously could not operate their own taxi fleet. So things like that, Uber, Airbnb, distributed ledgers and things like that in Taiwan, we don't want it to be a zero sum like a showdown between opposing values but rather a conversation around diverse values to leave no one behind and to promote innovations based on the rough consensus like the widely agreed shared values. So that's my job description. And I think for the past couple of years, when we thought of the pandemic with no lockdowns, thought of the infodemic, the disinformation crisis without administrative takedowns, I think that logic of collaboration across diversity extended beyond platform sharing economy now to the realms of cybersecurity and many other emerging topics. Fantastic, fascinating. And what resistance, if any, did you get and face as you started this role with, as you described it, a pretty big and ambitious goal that you were trying to do. Yeah, so in 2016, when I entered a cabinet, I made it very clear that I'm not working for the government and work with the government so that the government doesn't just work for the people, it works with the people. So with that, I had and still have three working conditions. First, it's about radical transparency. So everything the journalist lobbies, including this conversation, will post to the internet free of copyright at most attribution after each meeting of this kind. So everybody can learn what I have learned. So the second is voluntary association, meaning that I never issue commands and never take commands among peer to peer relationship with the administration. And third is location independence. Anywhere I'm working, anytime I'm working, I'm working, I only enter the cabinet office, I think twice a week or something physically. So all of this is before pandemic. So initially, of course, there are some resistance just to the, the galaxy of all this. Like, is it possible for a minister to work in the cabinet office with such three conditions? But fortunately, I think with our president, Dr. Tsai Ing-wen, very willing to support this kind of grassroots civil society connections. We eventually found legal interpretations to make not just personally for me, this possible, but for it to be possible for all public servants. So I think for the past four or five years there's more and more public servants taking on these more horizontal leadership, servant leadership role, instead of the traditional top down way. Great, and you talked a lot about things there that use the word servant leadership at the end. But a lot of the things you were describing leading up to it was kind of a leadership approach that it sounds like you took. And you also mentioned how young you were when you were appointed this role. In fact, I think maybe you're still the youngest cabinet member in Taiwan. But you've always been young when you've done things, right? You started coding early on at age eight. You started your first company at age 15. How do you think being involved in business and government at such a young age influenced that leadership style? Do you think, did that have any influence on that? Yeah, I think the internet, in particular IETF and W3C and the pro community were really my first experience in democracy. Because when I seriously engaged free software and later on open source communities, I was just like 15 years old, right, 16 years old. But across the internet mailing list, of course it doesn't really matter how old I am. But it will not be another few years till I actually get to vote in the democratic system. So although Taiwan is a democracy in 1996, we had our first direct presidential election, I can only support the campaign, but I cannot actually vote. So unlike many other people who got their first impression of democracy in a traditional system of very high latency, like every two or four years, very low bandwidth, like just first winner, past the post and so on. My first association to the word democracy actually came from the internet engineering task force, rough consensus, running code, a very high bandwidth low latency, wide connection kind of democracy. So it always shaped my belief that I don't have this not invented here when working within the government or within larger organizations. My impulse is always go to IRC or in our days matrix and say, this is what we're facing and we don't have the solution, but this is the end point. We believe those end points are valuable. So join us and form a horizontal grassroots group to tackle this together. So that's my leadership style. Got it. And I think the audience here being part of the round community in some ways it's not that different than what you just talked about, right? People wanting to come together with like-minded people with an objective or a mission in mind and bringing the right people together to do that. Any advice you might give a room of senior leaders in the tech industry based upon what you've learned to really help do that? Yeah, certainly. So I think a lot of my sharing is on say it at the say it platform view type, say it, RJ town, Microsoft senior leadership, you get to see everything that I had talked to Microsoft senior leadership a few months back. And a lot of it involving just like my advice to the president and premier in Taiwan about trusting our citizens. So basically you get no trust when you give no trust and as leaders in large organizations, the more that we trust the citizens, that is to say not just users, but fellow collaborates us to innovate with us, the more trust we actually get back from the ecosystem. And nowadays to give trust means to ensure that the citizens have agenda setting power in the sense that they can set the agenda of what we talk about instead of just making mods and add-ons and things like that. So sharing agenda setting power earlier on the decision-making stage, that would be my main advice. Got it, and that concept of trust between the government and the people, is that something that's always been in place in Taiwan or is that something that's been established relatively recently as you just described it? Yeah, it was pouring into the martial law. So obviously it was not always like that. When I was young, there was pretty much like, there's no freedom to associate to foreign political policy. The newspapers were heavily censored and things like that when I was born. But later on, I think the peaceful democratization of Taiwan has a lot to attribute to the emerging personal computing, internet, and later on the war web, and the war web's popularization in 96 coincides with the presidential election, as I mentioned. So for us, internet and democracy are pretty much the same thing because it's the first time that we get exposed to this entire community that really care about making decisions together at the same time that we gain our democracy. Now, I think another thing I want to highlight is that in Taiwan, we had constant earthquakes, typhoons and so on. We're caught between the Eurasian plate on one side and Philippine seaplane on the other, which bumps into one another very quickly and often we just get earthquakes. But that also push Taiwan upward toward the sky, a couple of centimeters for the highest point in Taiwan every year. Now, in every earthquake, there's just this unlikely alliance between people of different faith, different background, different language, and so on who all have to figure out how to rebuild after around the turn of century, one of the largest earthquakes and so on. So I would say that way before the Marshall is fully lifted and the democratization fully realized that the civil society organizations, the temples and churches I just mentioned, that also was part of their conversation. They gain a lot of trust, a lot of legitimacy simply by working with people on recovery together. Got it. So certainly when there are situations where people need to come together and they get a chance to work together, then there's no way of building trust better than actually doing that. Let's kind of switch gears and go a little bit back to the technology at the core of things. And Taiwan's democracy is really known as the strongest in all of Asia. And a big part of that is the work you're doing via digital technology. So can you share the concept behind what you refer to as digital democracy and what that is? Certainly. So as I mentioned, it's all about increasing the bandwidth and reducing the latency of democracy. For example, we have a actually functioning ePetition website. Join the GOV.tw where people can just go to the participatory budgeting, regulatory pre-announcement, all sort of budget auditing and so on and start their own petitions. And the most active ones are usually like 17 years old or 70 years old. These two age groups have a lot more time on their hands and care about sustainability and the next generation's welfare. But all in all, in Taiwan with 23 million people, there's around 10 million users to that website, which is a lot. And any petition that gathers more than 5,000 signatures meet a ministerial response. And if this is interministerial, then I get to meet those different ministers as well as the petitioners to figure out to co-create the solutions to get. So instead of waiting for two years or four years, if people, for example, find out the text filing system is, and I quote, explosively hostile to Mac analytics users. And I quote, instead of voting somebody in or out or a participatory budget, actually the people who complain go into a workshop we co-create in a few workshops. And then we made the next years in 2017 the text file experience together. So imagine TurboTax doing that in the US. So actually it really made the embracing of relatively new technologies very quick because when people push for that to be adopted, for example, in the text filing systems case, it may be the Ocean, the Coastal Guard facility in the conversation because in each ministry we've got this team of participation officers and they host the conversations across the silos. So when we talk about, for example, lifting up the surfing or amateur fishing restrictions, then maybe it will be the text agency office hosting the conversation because after work they also have good fish, right? So is this the senior public servants but in the petitioners role in the position because it's not directly related to their subject matter in an entirely cross-functional way holding collaborative workshops. And we've held more than 100 of these conversations that led to a lot of improvements in public service including telemedicine or bending plastic straws on bubble tea takeouts, many interesting cases. Got it. So it seems like all of that is really around how to use technology to get more people involved in all of the issues that they care about. Is that a good way to think about that? Yeah, the idea is that we bring technology to the people instead of asking people to come to technology. 10 million people using join platform means about 13 million people not using the join platform. So my work has always been to not replace face-to-face gatherings but tour around Taiwan to meet with people where they are. The local cooperatives, the social innovation organizations, social entrepreneurs and so on with the help of my reverse mentors because I'm more than 35 years old now, I'm 41, I'm old. So we've got 35 people under 35 who can summon me to all those different regional meetings to meet people where they are and to surface their ideas to the various different ministries. So everybody can learn their local ways of doing things and our president even runs a hackathon every year to surface the top five ideas into her next year's presidential agenda. Amazing, amazing. The COVID pandemic obviously was something that impacted everybody in the world. And I think in Taiwan, you maybe had a very different experience there than lots of other countries because of the way that you took advantage of digital democracy. Can you share a little bit about how you use that to help combat COVID? Certainly. So I remember my last travel before COVID was I think around end of February, early March to the DC where I showed this mask rationing map to people in the DC saying that getting access to easily accessible medical grade mask is very important. And in Taiwan, since the beginning, the very beginning, like the 6th of February, we've got this civic tech tool built not by the government but by GZero, VRGavZero, the GavZero folks that shows in your nearby pharmacies how many medical grade masks are available and anyone can see very quickly as they queue in line. People queue in before them using their universal health cards. They make a purchase and the stock gets decreed. The decrement is in real time, like in every 30 seconds. And it's like a distributed ledger because there's more than 100 different tools built by the civil society or seeing the one of the more popular one based on open street map, I believe. And serving, for example, voice assistants, people speaking different language and so on to ensure that people don't panic. And when people like more than three quarters of our entire population got access to the mask and put them on, then we reduce the basic reproduction number of the original strain to well below one, actually. And so we enjoyed almost a year without a single locally transmissible case. So that's how we responded so quickly because of the collective intelligence on our local equivalent of Reddit, the PTT, which is like Reddit except it's, it doesn't have shareholders or advertisers. It's entirely in the National Towering University as part of the academic network. So the epidemiologists, the public servants and so on took the conversation on PTT which was last day of 2019. And on the very first day of 2020, we start health inspections for all flight passengers coming in from Wuhan to Taiwan. So it's fast and fair and that enabled us to build a response that until now, we've not ever had a single day of lockdown and enjoyed a record economic growth in the past couple of years. Great, and you mentioned you messed in fast. So being able to react quickly almost in real time, it sounds like with the tool like that's on the screen here that's showing real time kind of mask inventory and fair in that everyone has access to it. The other thing that came about in COVID, at least especially here in the United States was kind of this information that was flowing and this question of what is real information and what is rumor and all that sort of thing. How did you deal with that in Taiwan? Yeah, we call it humor over rumor or out meaming, the infodemic. The idea is that this information and more information misinformation often gets viral because it has a higher basic reproduction number, higher R value, R number than the science and the clarification. So well before the pandemic, we already honed again collaboratively the design of zero counter disinformation way in which that just like flagging incoming email as spam people in WhatsApp like communication channels long press a viral disinformation as spam and reports to any of the entire virus or counter solicits call vendors they trust, but they all get aggregated like spam house into a public dashboard. So we can see which disinformation are going viral which has a highest R number. And then for the one that have the highest R number of the hour, our preservation offices also double as comedians. So they work on memes. So the idea is that after two hours at most we post two memes each with at most 200 characters. So this is a pre-pandemic example. Our premier head of cabinet picture here says that he noted that there's a popular rumor that says if you perm your hair many times a week the state will find you a million anti-dollars which is not true. And the younger version of the premier says, and I quote, I may be bought now but I used to have hair. I would not punish people with hair. And the fine print that says what we've actually introduced is labeling requirements for, you know, priming ingredients for hair products taking effect on July, 2021, but on the lower pot which I didn't translate is our premier as he looks now, he's in his 70s and said, if you perm your hair repeatedly in a week it will not damage your bank account but it will damage your hair. Your hairstyle will resemble mine. So this is very funny. It has a very high R number. And so within hours, you know, everybody becomes, is like a viral vaccine. People become immune to this information. They stop sharing this information or if they do it's clearly labeled and associating with the meme. So during the pandemic, we also had, for example we introduced mass rationing, which is great but there was a rumor that said, okay so the state is nationalized in mass production and the masks are made of the same material as toilet papers. So we're going to run out of toilet paper soon. So people didn't rush to buy masks but they rushed to buy toilet papers and there was a real kind of a rush to buy and it was quite chaotic and there's some pictures and so on and there's information manipulation from overseas in hostile jurisdictions that try to paint democracy as chaotic. So within a couple of hours, our premier again, now this time she was his backside, weakling his bottom, same very large font, hard to translate. Each of us only have a pair of bottoms so we can't use that much but it's a war play because twin to stockpile sounds the same as bottom to it in mandatory. So it doesn't pay to stockpile and the table says that the domestic materials, the plastic materials are for masks but for tissue paper, toilet paper it's entirely different material is from South America. So don't confuse the two, the toilet paper will not run out. So and with all the required information from administrative economy, but again is this spike protein right here is this packaging that has a huge basic reproduction number. And so within a couple of days, pretty much everyone in Taiwan have seen this meme and they stopped panic buying. So again, outmitting the trending disinformation that has been our way to counter disinformation crisis. Yeah, that's a really great story. And I have to ask like on staff in the government you mentioned that some of the workers are comedians like what does that actually look like in the government office? You have on staff comedians that that's their job. Yes, we have on staff comedians and we pre-clear for example, personally I pre-clear all the copyrights including the attribution rights or whatever. So any of my photo can be used by the meme makers to say pretty much anything without my consent. So they have a limited material to work with and I'm sure our premier head of cabinet pre-clear that wiggling bottom picture as well. So we put a lot of trust to our engagement participation officers to work with professional comedians and as comedians themselves. Yeah, it's an amazing tactic. I just, the thing that's going around in my head is in the US there's a comedy show called Saturday Night Live that often makes fun of the US government and I just keep thinking in my head what if the government actually had control over using Saturday Night Live to out meme the rumors. So it's an amazing, amazing approach. So beyond COVID, obviously you're using lots of technology to do other similar things and drive a bunch of solutions and you've introduced this thing called Polis and you sort of insinuated before this concept of getting people all on the same page around certain topics. And I think that's what Polis helps do around driving something you call collective intelligence. Can you share about that, how it works and some of the results from that platform? Sure. So the experience if the projection can get the 2015 UberX conversation on the screen. So that's how it looks like for anyone participating in a Polis conversation. It's the friends and families resonating with each other on how people feel about the UberX case. So that's me in the middle and the circled avatar represents an anonymous person entering for the first time. And they can see their friends and families in initially four different clusters. But the importance is that the area here is not a headcount, it is the diversity, the plurality of that cluster. So the idea is that you see a fellow citizen sentiment like liability insurance for passengers is very important. If you agree, you move towards me, the person who proposed this. And if you disagree, you move away from them. But there's no reply button, so there is no room for trolls to grow. And after answering a few, you're asked to chime in with your own reflections again for other people to vote. So obviously this on-screen is a two-dimensional reduction of the K-means clustering for the clusters and principal component analysis for the division devices points. But the scoreboard shows consistently what works across this initial division of the sharing versus gig economy and so on. So people were quite surprised to find this report which is updated continuously in Polish that on the TV or on the more anti-social corner of social media, the divisive statements keeps getting repeated. But it's actually just like 5% of the statements. Actually, we all agree with most of our neighbors on most of the things, most of the time. And this is not just a Taiwanese thing. This actually, this report on-screen is from Bowling Green, Kentucky. So if you search for Bowling Green Polish and you get to see what they probably agree on across party lines. So once people get this like shared reflection of the rough consensus here in plain sight, it then become very easy for us then to basically ensure that the taxi companies, the local unions, the temple churches, and Uber all commit on the top agenda as said by the people who broadly agreed across the initial divisions so like not undercutting existing meters and so on. So very swiftly we were able to basically table the differences, the ideological differences of sharing versus geek economy and pass something very practical. That is now the multipurpose taxi law of Taiwan. So I hope that illustrates the main point. Yeah, it's amazing. It's almost like this, you're taking the glass half full view and instead of putting all the focus on where people disagree, you're putting on the focus on where they agree which actually gets things moving forward, right? Because then you've got a situation where collectively people can report on something. I think it's amazing. In all of this, there's obviously some products that you've created or have been created and used. I think a lot of technology leaders might look at it and say, hey, the products are really not all that complicated, but it's the way that they're actually applied and the way they're actually used that is important. And so there's certainly some principles that are at play there and how you think about bringing these products to the people. For example, in democracy, inclusiveness seems to be a very important one. How have you succeeded on making sure everything is inclusive when it comes to technology in Taiwan? Yeah, so a lot of the experiences that we delivered didn't originate from the government, not even from Taiwan, like Polis initially from Seattle and later on in New York, right? They're the Computational Democracy Project now. And the petition website I mentioned is actually from Iceland, better than Kavík, participatory budgeting from Barcelona and from Madrid. So basically I'm just one of the tentacles of this digital open government tribe and Taiwan just happens to be one of the most connected places, I think we're top actual broadband usage, mobile usage, so that we do not have to worry about whether somebody has connectivity because in Taiwan, broadband has been a human right for quite a while now. Any time, any place in Taiwan, you're guaranteed to have video conference rate broadband for just, I think, $15 per month for limited data. So going beyond the broadband's human rights, the digital competence education and so on, we can then empower the 70 years old and 17 years old to innovate and discover new possibilities. For example, the mask rationing map, it's repurposed into rapid testing, rationing vaccination finder and things like that. None of this is state capacity. All of this is civic capacity and the state's role is not to do a procurement but rather reverse procurement. Basically we are the vendors and the people are one giving the specification and that is the manifestation of the principle of trusting the people first and also sharing early decision-making power on the agenda setting stage. Yeah, that's a really, really great way to think about it. And another principle that I think is really important, especially when you talked about the people are the ones that are actually telling you what you need. How do you think about privacy? Privacy is obviously a hot topic when it comes to people and government and technology. How have you dealt with that within the things that you're introducing into Taiwan? Yeah, so I'll take contact tracing as an example because if it's not privacy enhancing, people will not use it in democratic societies. In Taiwan, in 2021, when we finally faced the first wave, which was very quickly squashed, eliminated, we had two different implementations of contact tracing. One is based on the exposure notification interface co-developed with Google and Apple, what many other jurisdictions use as well based on Bluetooth and proved by mathematicians as good enough, pretty good privacy. But the other one is invented by GovZero, which is much simpler to explain. And the GovZero one basically said all the venues get to post 15-digit random numbers that they don't have to send to the state just generate a random number. And then the QR code maker puts that 15-digit and anyone who don't have a smartphone can just manually text that 15-digit to a toll-free number 1922 representing the center of the main command sensor and finish a check-in and that's how you do contact tracing. But if you do have an iPhone or something, you can just point it to the QR code which will pop your building SMS that text the 15-digit to 1922 with a line that says this must be used for pandemic control only. So in technical terms, this is the oblivious multi-party federated storage. In a sense that the number doesn't go to the state. It goes to your local telecom. The local telecom deletes that after four weeks. And when there's local outbreak in any venue, finally, work with the contact tracers to send notifications to the people who have frequented the same venue in the same time period by working with the five telecoms in a distributed way to send such exposure notifications. And there's also a reverse lookup accountability portal where people can go to their telecom and then just sign with SMS and see for the past four weeks which contact tracers in which municipality have looked at their data. So Bluetooth wasn't rolled out before the pandemic. Very hard to explain, quite technical. But SMS and QR code everybody had experienced. You don't have to download everything. Numbers speak for itself in the first week after rolling out the Civic Tech solution more than two million venues voluntarily printed out such random codes in the QR code and it took more than a year for the Bluetooth solution to reach critical mass in Taiwan for people to download it to the Play Store. And so this shows that when we trust the people, the citizens, the privacy away and privacy enhancing human organizations, they get to design something that is privacy enhancing and far easier to explain and easier to win the trust from people that successfully reduce the contact tracing for each case from more than 24 hours into just 24 minutes. Very amazing and interesting. You talked about Government Zero that built this. And you've referred to Government Zero a couple of times. Can you explain a little bit what that is so people understand what Government Zero is and how they're actually generating these solutions? Yeah, it's a domain hack. So one of my friends, registered a domain in 2012, G0v.tw, which allows for each and every Government Service we didn't like. We simply forked the Government. So if you don't like, join the G0v.tw, just change it O to a zero, join the G0v.tw and you get into the Shadow Government that's always more fun. And so it's all open source free software and so any improvements that G0 has to the state capacity, the state can simply click Merge and merge it back into actually the official Government. So that's how we got the budget visualization at first, but very quickly expanding to the collaborative dictionary and educational, cultural, science, air pollution monitoring and so on. For each and every project that worked on the Government Zero site, there's then a corresponding office on the Government site that simply turned the Civic Tech into a maintained by the Government. G0v.tw, so that the Civil Society can move on to do other things. There are exceptions. For example, the collaborative fact checking of facts never got merged because it's not the administration's job to fact check the journalist. That must remain in the Civic site. We just focus on the comedians. So if you search for G0v, you'll see the manifesto and for the past 10 years, there's more than 100 or so such innovations got merged by the local and national and international governments from the G0v initiative. Got it. That's pretty impressive. And where do you think that sense of civic duty or what compels these people to become a part of this and actually build these solutions for the Government? Or for the people, I would say. Yeah, it's just very satisfying, right? For the young people, for the primary schoolers, I think all primary schools have the G0v co-designed air box now that measure PM 2.5. For the young people, the very young people, their measurements, the data stewardship that they play affected whether their parents go out to hike or jog in the morning because of the PM 2.5 level is a civic duty to the family and to the community and the fact that the middle schoolers when they fact check the three presidential candidates during their former and debate and if they found a factual error that one of the candidates made, their name actually may be on their leaderboard in public TV and live streaming and so when they get older they can live stream the counting ceremony in our mayoral candidates counting or presidential candidates counting boxes and so on and it was very popular to just have a kind of democracy as a celebration so all of this is intrinsic reward. The intrinsic reward means that the more people contribute we see democracy like something like a design so it's not something that we just perform like a ritual every year but rather something we can actively contribute to and really it's just fun. Got it. That's great. So with a lot of the audience here being from the United States I think I'd be missing the opportunity to ask for some advice from you with your experience in Taiwan and the successes you've had so what do you think are some of the biggest learnings that other democracies like the United States can learn from what Taiwan has done? Yeah, it will start small the designs like Polis join platform can work very well in a town hall like literally in a township district and that's where we started remember when I said before we had this different local community builders co-ops, social entrepreneur who really banded together when there's a natural disaster like earthquake or COVID or things like that but even before that we worked to empower them so that they can employ such pro-democratic social media tools instead of having to rely on an advertisement field social media so try something like that in your local community maybe in your department as easy as holding a town hall powered by Slido or Polis and just make it a everyday habit and once you make it a everyday habit you will keep seeing more and more ways to scale it up, scale it out and scale it more deeply. Great, there's some questions from the audience that have come in and there's one that's actually very similar to what we just talked about but maybe an extension of that do you think the approaches that you've talked about would work in other countries or is there something unique to Taiwan's culture that makes them work more effectively there and are there any examples of other countries that are succeeding similarly to Taiwan? I mentioned Iceland, Estonia of course and Madrid, Barcelona there are pioneers around the world for digital democracy I would also like to say that in Taiwan we in short broadband as a human right and digital competence as producers of data not just consumers competence not literacy in our basic and lifelong curricula. I think these two are the two fundamental pillars on top of which to build digital democracy if you just jump to e-petition without broadband as a human right without media competence then of course that may create a very divided it will widen the digital gap, right? So I think for the US in the jurisdictions the states and parts of the US that actually had those two I would say that there's no reason why it cannot thrive in such jurisdictions but in the places where that still needs to be resolved maybe through lowest orbiting satellites or things like that maybe we can focus on that first now in our region the mass creationing map once we introduced that in Taiwan in just four weeks the South Koreans deployed that actually using the same API so the very first map that worked was the triangular map with a lot of triangles that you just saw in Korea even though the author doesn't speak Korean he speaks JavaScript, right? And then the dashboard we also collaborated with Japan so at least in South Korea Japan and Taiwan there's a lot of work going on and Casero used to hold hackathons in Okinawa which is like the midpoint between the three jurisdictions so we can hack together Fantastic let's look at some of these other questions that have come in from the audience so this first one is web users around the world are shifting from an era of absolute anonymity to building out their own identity as humans in the digital world what opportunities do you see for governments to interact with their own citizens but also the rest of the world outside of their own citizens thinking specifically about electronic visas and allowing travelers to move freely across borders and things like that Yeah of course Taiwan is both part of the UDCC Digital COVID Certificate and the SHC of US Japan Australia so we're again like two tectonic plates we're part of both and during the COVID we offered gold card visas you can search for Taiwan gold card a lot of people I used to encounter in Silicon Valley were actually in Taiwan for those past couple years because they were able to become Taiwanese residents for three years and renewable based on nothing but the fact that they successfully launched a startup before so we hand out gold cards even for people who have never been to Taiwan so it's like the Estonian E-Resident except you actually can be a resident here and you and your family enjoy healthcare and so on if you spent six years five years here contributed a lot you can also become also Taiwanese get our passport without sacrificing your original one so we take a very plural view on national identities and we're very eager to work with the for example the centralized identity community I mean our upcoming Ministry of Digital Affairs will have our entire website on IPFS interplanetary file system real inquisition on copyright CC0 so at any point you can fork our public code we of course learned a lot also from the USDS the 18F design systems and things like that so I think with time the idea of identity may just go entirely post Westphalia and I sincerely wish that we can accelerate that progress great another question here it sounds like this is definitely a question from a product leader how do you measure the impact of your disinformation meme campaigns sure well we measure the basic reproduction number basically it's exactly like pandemic monitoring right in the risk of taking the metaphor too far or basically taking the mRNA of any trending disinformation and neutralize it with a different spike protein that's comedian to make sure that the outrage doesn't turn into anger or polarization but the outrage turns into something outrageously funny and once people left the palace really there's no place for the original toxic virus to take hold so you can measure it using as I mentioned the co-facts the antivirus campaigns people flagging incoming email messages the spam and scam and so on using exactly the same way that we measure the scams and spams and virus of the computer sort and we can reuse the same apparatus to measure on the dashboard the virus of the mental sort got it kind of on long the similar lines this next question is sometimes the right policy answer is unclear have you used any real life but fast experimentation techniques define what works in real life versus theory are governments open to A.B. testing or other approaches before policy implementations yeah I mean when there's more than 100 different mass creation and visualizations not all of them work well but it's really a smart right as a government if we do a procurement we have a prefer vendor and if that vendor doesn't work or doesn't respond to the emerging threats then we have a problem but it's not procurement we just open up the A.B.I. and ensure that we lower the cost for experimentation by providing free computation in our national center for high speed computation to anyone from the civic tech world to fork one of the 100 maps and make another of their own so basically the civil society which excels in producing norms can converge those technological solutions without moving behind the so called minority requirements because in Taiwan we have more than 20 national languages including the Taiwanese sign language so there's a lot of need to make the website accessible so we basically say okay if we make a procurement website it must speak the language of open A.B.I. from the next foundation it must interact with robots otherwise it's discriminating against robots I guess we piggyback on the language of the US would be section 508 accessibility so you must not discriminate against robots and what this means is that anyone who wants to then fork the front end experience can just take this robotic A.B.I. interface ignore the website altogether and recombine in interesting ways so it's not us to converge what product wins or excels it's basically a public code experimentation ground and any 17 or 70 years old can fork that solution to make it work better for themselves so it's like entirely open innovation with the people not for the people hope to answer the question great we've got two more questions we're getting down to the last couple of minutes but I think this is a really interesting one it's all about the future of technology and democracy at the age of A.I. specifically when A.I. is successful based upon the data set that you're able to build the A.I. on top of and often times democracy is an inhibitor to generating some of that data so what are your thoughts about how optimistic are you about A.I. type technology in democracy is it a inhibitor I just read on hugging face that Bloom a very large open science multi linguistic data set has been made publicly available and it's very democratic right without democracy and in particular freedom of speech and associate I doubt that they were able to generate that much corpus in a democratic manner so I think democracy is great to generate data sets especially in large language models with that said I think the question may be implying that it must be symmetrical in a democracy people in primary schools need to also like nothing about them without them need to also understand the bias the distribution and so on behind a model so that they're not locked in to any particular stereotype because after all they are leading our future like zeitgeist so we don't want them to be locked in particular was to interpreting the world so I often say to young people that I understand A.I. as assistive intelligence meaning that is assistive technology like this eye class is aligned with my values of my vision like literally my vision to see you more clearly but not push advertisements to me and it's accountable meaning that if it breaks if it has blurry positive has bias I get to fix it myself or take it to the repair person we don't have to pay a lot of licensing fee or reverse engineer a cyan DA or anything like that so I think for the foundational models it must work like my eye class and to what becoming assistive intelligence in order for democracy to thrive with A.I. instead of kind of democracy versus A.I. great great and a last quick question for you that will sneak in here is it sounds like you've had so much success so far what's next for you and the digital ministry? Yeah so Taiwan didn't have a digital ministry so for the past five and half years I'm the chief information officer working with other ministers to collaboratively do digital transformation but the recent experience with COVID disinformation and Ukraine showed that maybe this approach can also work for example the US equivalent would be cyber command the heart cyber security pod or the like starlink side of things the 5G and beyond 5G the 6G development the spectrum resource planning and rationing as well as the platform economy originally part of the ministry economic affairs so in addition to work on digital democracy now I also get to work on setting up the new ministry that works also on the resilience part of things and also the heart cyber security part of things and also on the platform economy and resource planning for the spectrum kind of things so we'll see if we can take a servant leadership that work with around a hundred people participation office all work very voluntarily to well over a thousand people and around I think one billion US dollars in annual budget and see if the same philosophy still holds and wakes well I'm excited to watch and see where that journey goes and what success you can have I think you've proven that there are lots of innovative ways to go do things so we're at time I think I would love to keep the conversation going but want to thank you so much Audrey for leading this incredible discussion for taking your time I think it's been truly inspirational and really appreciate you you coming to share some of your experiences with us thank you really really good questions and live low and prosper everyone great we're going to say goodbye to Audrey thank you again we have lots of great opportunities coming up and like to share just a couple of quick highlights for you to plan to attend so next Wednesday August 3rd join the round community in discussing how teams can increase output with both conviction empathy so that we can get our teams operating at peak performance I think that's something that everyone is always trying to do and that's an exchange format which is really focused on walking away with with things that you can go apply in your day jobs right away and then later in August we'll be hosting Eric Sprunk who's the former COO of Nike to talk about their digital transformation and for those that don't know 2022 marks 20th anniversary so maybe Eric will give us a history on on technology throughout all those 50 years we'll have to see how that goes and then in early September we're launching a brand new series that we call the front page which is a no judgment exploratory series dedicated to exploring kind of those buzzword tech topics that are showing up in the news today and so Tanya Kiri part the former head international product at Patreon will be joining us to lead that conversation focused on the creator economy so super exciting things coming up hope to see you there and thanks for joining today