How Pro-Social Technology Is Saving Democracy from ‘Big Tech’ with Audrey Tang | TGS 169

Summary

Highlights

• Audrey Tang's role in transforming Taiwan's digital landscape is a beacon of hope for democracy using technology. 💡
• The discussion highlights how AI can assist in civic engagement without replacing human interaction. 🤖
• A unique insight into how plurality can serve as a counter to the potential perils of AI singularity. 🤔
• Taiwan serves as a model for the world on integrating technology in governance, enhancing public trust and participation. 🌏
• The potential of pro-social technology to foster community and ecological consciousness within societal systems. 🌱

Key Takeaways

• Audrey Tang envisions a world where technology fosters human connection, not division. 🌍
• Taiwan's innovative digital public squares demonstrate how tech can enhance government transparency and civic engagement. 📊
• Plurality is crucial; it emphasizes diverse voices and collaborative solutions over a singular, AI-dominated approach. 🌟
• Social media's future could be rooted in fostering community and empathy rather than just engagement and profit. 🤝
• Youth engagement in Taiwanese politics shows significant impacts, leading to real policy changes. 🎓

Overview

Chapters

• 00:00 - 00:30: Introduction to Pro-Social Technology This chapter discusses the concept of pro-social technology by challenging traditional business terminologies, such as 'human resources' and 'incentivizing corporations', which may distort our perception. It suggests shifting our perspective from viewing technology just as tools ('internet of things') to considering their potential for collective human experience ('internet of beings'). The chapter advocates for moving from 'virtual reality' to creating 'shared reality', promoting a deeper, interconnected approach to technology.
• 00:30 - 02:00: Audrey Tang's Role and Impact in Taiwan Chapter Title: Audrey Tang's Role and Impact in Taiwan. The chapter discusses Audrey Tang, Taiwan's Digital Ambassador at large, and their work in integrating technology and transparency into government functions. The emphasis is placed on shifting from machine learning to collaborative learning and from user experience to human experience. The overarching goal is to empower the voice of people through this digital transformation.
• 03:00 - 06:00: The Sunflower Movement The chapter titled 'The Sunflower Movement' discusses the contributions of Audrey Tang, who served as Taiwan's digital minister from 2016 to 2024. Tang was known for advocating radical government transparency, aiming to make information, data, and resources accessible to the public. The chapter explores Tang's successful projects and the guiding philosophy of plurality in their work, highlighting its significance in a global environment.
• 06:00 - 11:00: Digital Public Squares and Civic Engagement In this chapter, the discussion revolves around the initiatives introduced by Audrey in Taiwan, which exemplify how digital public squares and technological advancements can facilitate greater civic engagement and effective policy changes. Through these projects, technology is made accessible to ordinary people, fostering real communication and tangible changes in governance. The speaker expresses their initial naivety on the topic and their amazement at the accomplishments achieved by Audrey and their team. The chapter concludes with a hopeful reflection on the potential global impact if more communities and countries were to adopt similar strategies.
• 12:00 - 18:00: Tackling Deepfake Advertisements with AI The episode focuses on the optimistic perspective of leveraging technology, specifically AI, to address the issue of deepfake advertisements. It emphasizes the potential of technology to support pro-social, community-focused, and environmentally conscious goals.
• 18:00 - 23:00: Deliberative Democracy and AI Facilitation The chapter discusses the role of deliberative democracy and AI facilitation in modern governance. Audrey Tang, a notable figure from Taiwan, is introduced as a guest in the program. The conversation highlights her impressive contributions to successful movements and governance initiatives in Taiwan over the past decade and a half.
• 34:00 - 41:00: Plurality: An Alternative to Singularity Audrey Tang became Taiwan's first Minister of Digital Affairs from 2016 to 2024, and later served as Taiwan's Cyber Ambassador at large. Tang has an extensive background in coding and digital innovation. A significant part of their journey into politics in Taiwan began with involvement in the Sunflower Movement.
• 48:00 - 52:00: Transforming Social Media for a Pro-Social Future The chapter discusses the state of Taiwanese society in 2014, highlighting the polarization and the low approval rating of the president at the time. The discussion touches upon the participant's role and experiences within the movement aimed at transforming social media towards a more pro-social direction, and how these experiences influenced their current perspective and work.
• 79:00 - 82:00: The Importance of Language and Communication Chapter Overview: The chapter 'The Importance of Language and Communication' explores the significant role that language and communication play in global politics and economics, using a case study involving a trade deal with Beijing. It discusses the urgency felt by a parliament to secure a deal based on the potential economic growth, competitive pressures, and the fear of missing out. Additionally, it highlights concerns about the implications of such agreements, particularly in areas like telecommunications, where companies like Huawei and ZTE would be granted access. The chapter underscores the complexity of international negotiations where language and rhetoric can shape decisions with far-reaching consequences.
• 83:00 - 89:00: Innovating Democracy as a Communication Technology The chapter titled 'Innovating Democracy as a Communication Technology' discusses a situation where citizens took proactive measures by peacefully occupying their Parliament for three weeks. This act was not just a protest demanding something or opposing an entity, but a demonstration of taking matters into their own hands. It reflects on the broader impact of communications on the environment, labor, and other societal factors.
• 116:00 - 122:00: Cross-Cultural and Intergenerational Solidarity The chapter focuses on tools and methods developed to facilitate cross-cultural and intergenerational solidarity, specifically through citizen assemblies and online platforms. It describes the involvement of a large number of people engaging both physically and virtually to discuss the impact of trade deals. The chapter emphasizes the importance of providing platforms for individuals to understand and debate the implications of such deals, promoting a collaborative approach to shaping future regulations.
• 133:00 - 139:30: Conclusion and Final Thoughts The final chapter wraps up the successful outcome of a political assembly or movement where daily plenary sessions were held to negotiate and agree on key issues, particularly focusing on achievable goals or 'low hanging fruits'. Over the course of three weeks, a coherent set of demands was established, and subsequently adopted by the Speaker of the Parliament, marking a unique instance of agreement and convergence in such movements. The chapter concludes with the narrator being recognized for their efforts and insights, being appointed as a reverse mentor, a young advisor to governmental cabinets.

How Pro-Social Technology Is Saving Democracy from ‘Big Tech’ with Audrey Tang | TGS 169 Transcription

• 00:00 - 00:30 Repeating the category errors of some business as  usual language, such as saying human resources or   incentivizing corporations just propagates  this category error in our thinking. So it's   like trying to chart out a map, but with like  very tilted lens, you can't perceive the world,   right? 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.
• 00:30 - 01:00 When we see machine learning, let's  make it collaborative learning. When   we see user experience, let's  make it about human experience. Today I am joined by the Taiwanese  Digital Ambassador at large Audrey   Tang to discuss their work championing the  integration of technology and transparency   into government functions with the gold  to further empower the voice of people in
• 01:00 - 01:30 policy decisions. Audrey Tang was the first  digital minister of Taiwan from 2016 to 2024,   where they were dedicated to promoting a  radical level of government transparency with   aims to make all government information, data, and  resources as accessible to the public as possible. Today, we discuss a few of their past. Successful  projects as well as the philosophy of plurality,   which guides all of their work in a  global environment where the topics
• 01:30 - 02:00 of tech and artificial intelligence  can feel esoteric and out of reach for   ordinary people. The projects that Audrey  has introduced in Taiwan and beyond have   resulted in real humans communicating  and enacting effective policy changes. Personally, I was naive on this topic and  I was blown away by what Audrey and their   team were able to accomplish, and I wonder what  the world might look like if more communities,   more countries, the whole world  followed this lead. In my opinion,
• 02:00 - 02:30 this episode highlights the more hopeful  side of the great simplification where   technology could be used towards more pro-social  community, ecologically aware, oriented goals. Additionally. If you are enjoying this  podcast, I invite you to subscribe to our   substack newsletter where you can read  more of the system science underpinning   the human predicament, and where my team  and I post special announcements and new
• 02:30 - 03:00 written Franks and other such snippets  related to the great simplification. You can find the link to subscribe in the  show description. With that, I am pleased   to welcome Audrey Tang. Audrey Tang, welcome to  the program. Hello. Good time everyone. So glad   to be here. So you, um, already have quite an  amazing resume with lots of successful movements   and governance initiatives in your country of  Taiwan, especially over the last 10 to 15 years.
• 03:00 - 03:30 You became the first Minister of Digital Affairs  in Taiwan from 2016 to 2024, and now you are   Taiwan's Cyber Ambassador at large. Mm-hmm. Um,  but from what I understand, you'd been studying   and working in coding and digital innovation  for quite a long time before that. But much   of your journey into Taiwanese politics began  what was called, um, the Sunflower Movement.
• 03:30 - 04:00 Maybe we could start there. Can you tell us a bit  about what that movement was? What was your role   and experience within it and how it affected  your current, uh, worldview and, and work? So back in 2014, the Taiwanese societies deeply  polarized. The president at the time was enjoying,   uh, 9% of approval, which means that  in the country of 24 million anything,   the president ma says 20 million people  are, you know, not so happy with it.
• 04:00 - 04:30 And so at the time, the parliament was  trying to rush through a trade deal with   Beijing and using this. Basically, oh,  it's inevitable. Uh, the GDP will grow,   uh, we'll enter an acceleration  phase. Uh, if we don't sign it,   other people will sign and then we will lose  out, and so on. So forth. Uh, this kind of logic. But then there's, uh, people who deeply, uh,  think about the repercussions that it has,   not just on our system of telecommunications. For  example, Huawei and ZTE will be able to enter and
• 04:30 - 05:00 monitor, um, our communications, but also the,  uh, impact on environment, our labor on many   other things. And so, um, in March at a time,  uh, people took matters with their own hands. So we peacefully occupied the Parliament for three  weeks. Now a crucial difference is that we're not,   uh, protesters who only demand something  like against something we're demonstrators
• 05:00 - 05:30 that showed a alternative. And  so we developed a lot of tools,   like of the half a million people  on the street and many more online. You can show up to a citizen assembly like  conversation. You can enter your company number,   uh, and then you can very quickly see how  exactly does the trade deal affect you. And   then you can have a conversation with a dozen  other people who are also interested in this   matter, to think about ways, uh, to basically  regulate, uh, future trade deals of this kind.
• 05:30 - 06:00 And so every day we read out, uh,  like a plenary what was agreed, uh,   that day. And then every day we push it. Uh, a  little bit more on the low hanging fruits. Uh,   that's, uh, basically under debate. And so after  three weeks, we managed to agree on a set of very   coherent demands and the Speaker of the Parliament  basically say, okay, we'll adopt it, go home. And so it's a very rare occupy that really  converged instead of diverged. And so at the   end of that year, I was tapped as a reverse  mentor, as a young advisor to the cabinets,
• 06:00 - 06:30 basically for each and every incoming  polarized, uh, topic. Instead of fighting   out on social media, which isolates people  into this anti-social corners, we want to   make something like the occupied parliament  space that we did build that year without. Literally occupying the Parliament. And  so I basically built many digital public   squares to tackle things all the way from  Uber in 2015 to counter pandemic in 2020,
• 06:30 - 07:00 all the way to generative AI and so on  in 2023 and 24. And so by 2020 already,   the approval rate is back to more than 70% because  we systemically discovered the uncommon grounds   that can pull people together despite their very  polarized ideologies or political affiliations. Um, I have so many questions, Audrey, so,  mm-hmm. Uh, I'm, I'm glad you're here today.   Let me set the context a little. We, uh, in  the world today realize the algorithms and
• 07:00 - 07:30 social media and the polarization and the  echo chambers, and the inability to really   have civic discourse about the things that  matter, and we don't even know what's true. Uh, I am not an expert on that other than I.  I am an expert in knowing how it important it,   it is to solve these issues  if we're going to have, uh,   any hope of solving the larger issues  that I discuss on this platform. So you
• 07:30 - 08:00 just mentioned that instead of protest,  you wanted to have alternatives. Mm-hmm. And, um, I'd like you to unpack that  a little bit because so much of our,   uh, postmodern, uh, critique of the world is  just pointing out what's wrong and what's bad,   and it's just like an anger sort of thing instead  of actually proactive. So can you mm-hmm. Describe   why, why that's so important, uh, and, and  your experience with offering alternatives.
• 08:00 - 08:30 Yeah, definitely. So I'll use one recent  example a year ago, um, about, uh, March,   uh, 2024. We saw a problem online, uh,  with a lot of deep fake advertisements,   uh, running fraudulent, uh,  ads that pertains to, you know,   sell crypto or sell stocks or so on. In Taiwan  is always from Jensen, Juan, uh, you know,   the Nvidia guy, uh, the richest Taiwanese, uh,  and sometime also from other entrepreneurs.
• 08:30 - 09:00 And then, uh, if you click on Jenssen's,  uh, likeness, um, he actually talks to you,   not just chat, but also, you know, voice  and the whole deal. Uh, and that's because   the generative AI has grown to such a  point where it can run such persuasion,   um, what we call info attacks, uh, with no  human supervision. And so, uh, to solve that. We sent SMS text messages to 200,000 random  numbers in Taiwan from 1 1 1. That's the
• 09:00 - 09:30 trusted number. People know it come from the  government asking just one simple question,   how do you feel about the  information integrity online,   what to do about it? And so people gave  us, uh, their ideas and then. A thousand,   2000 or so, uh, people, uh, volunteered,  uh, to basically have online conversation. And now at the end, we did not engage  all the thousands of people. Uh,   we chose 450 people that is a statistical  representative of the Taiwanese population
• 09:30 - 10:00 in terms of place they live, age,  bracket, gender, so on and so forth.   And so this microcosm, this mini public,  um, deliberated online for almost a day. And the way it works is that people enter,  and it's like a Zoom call with nine other   people. So 10 people each in each room and the 45  rooms deliberated about the potential responses
• 10:00 - 10:30 to this incoming issue of the fake fraud. So  maybe one room would say. Okay. Uh, if Jensen   did not sign off on that advertisement,  it should actually be assumed as skim. We shouldn't assume human unless proven  otherwise. We should assume skim unless,   uh, proven by the human. Another roommate say,  uh, if Facebook doesn't secure the signature   and somebody gets skimmed outta $5 million,  then Facebook should be liable for that $5   million because otherwise they would just pay  the fine, which is, um, you know, negligible.
• 10:30 - 11:00 Uh, and another room says if Facebook also,  you know, doesn't even agree on this framework,   we should slow down. Connection to the Facebook  servers so that the business goes to Google,   uh, and so on and so forth. And so all these  ideas are facilitated not by human, but by the   room itself as a AI facilitator that encourages  the quiet people to speak up and make real time   transcripts and identify what we call sensemaking,  uh, the uncommon ground between those rim.
• 11:00 - 11:30 And then we read it back to everyone and people  agreed more than 85% regardless of their party   affiliation on the package of measures.  And then we check with the stakeholders,   the big tech in April, and they really cannot  lobby against it because there's no fraud,   uh, party. And we can show  that everybody agree on that. And then finally, in May, we push  out the draft. And it's one of the   very rare legislation in Taiwan where all the  three parties now of which have a majority,
• 11:30 - 12:00 um. Just fast track through. And so  now this year, if you open Facebook   or YouTube in Taiwan, you just don't see  any fraudulent advertisements anymore. That's a solved problem. And that is because we  can show that that was the sense-making result   from this broad listening exercise. So this  was an anecdote, but you can get the intuition. That is pretty amazing. I  actually didn't know that,   but let me ask you some questions  about that. So you said, uh,   you started with 200,000, you got it down to  several thousand, and then you chose mm-hmm.
• 12:00 - 12:30 450 mm-hmm. Based on demographics and then they  were in 45 rooms of 10. Mm-hmm. Mm-hmm. And,   and so there would be, because that itself  kind of reflects Dunbar's number of sorts   that you have to bring it down mm-hmm. To a  manageable Yes. Human interaction level and   then scale a little upwards. So did each room  of 10 come up with its own kind of verdict? Yes. And then Exactly. And then you compiled  those 45 verdicts in a, in sort of a, a way that is exactly the case. And so  the, uh, of the 45, uh, 30 rooms,
• 12:30 - 13:00 uh, were from lay people and 15  rooms were from practitioners,   like people who are actually media people or  social media, uh, professionals. And we made sure. That these, uh, cross, um, pollination, uh, works  in the plenary. So, uh, people had one segment of   conversation and during the plenary we weave  together those questions and suggestions and   so on. We read them back, uh, with interpretation  by experts, and then we enter the second segment,
• 13:00 - 13:30 uh, which then, uh, basically ratify uh,  on this, um, plenary, uh, conclusions. The good thing about AI is that  previously you will need a lot   of people to like read individually those comments   in order to make sense. But now AI  can do that without hallucinating,   so you can get a pretty grounded report  based on those 45 rims, individual verdicts. So what about someone that  wasn't part of the 200,000? Um, you said there's 20 some million people  in Taiwan and they see the results of this,
• 13:30 - 14:00 wouldn't they? Their initial reaction be, oh, this  was just some AI scam that put this together. Why,   why should I believe what, what  ended up being in in legislation? Yeah. Uh, part of the reason why is  that we've been doing this for 10 years,   and so starting from 2015, uh, during the  Uber consultation where again, we just ask   people how do you feel about someone with no  professional driver license, driving to work,
• 14:00 - 14:30 meeting a stranger on an app and charging them  for it, um, people already had like more than   100 of those online either petition or the online  sortition, uh, or this kind of, uh, conversations. So people can refer to the prior experiences and  they know they can kind of force a response, uh,   just by going to the national participation  platform and to get 5,000 other people to,   uh, basically produce a counter signature,  uh, so that for any regulation or for any
• 14:30 - 15:00 policy. If they're not happy about  the draft that we come up with, uh,   if they get 5,000 people, they can  force another round of this exchange. How scalable is this? Can't this be applied  to almost any issue in, in the world and,   and technically, um, maybe not politically,  but technically in any country in the world. Yeah, I think the trigger point, uh, really  is that you need a topic that is urgent enough   and politically is not the sole  purview of an existing department.
• 15:00 - 15:30 Mm-hmm. So if it is already a single department,   then they tend to feel that they've already got  a solution figured out. They do not actually   need the collective intelligence. And if  it's not urgent, then it does not warrant   this kind of instant sense making technologies  you can afford to do that over years and so on. So just a couple weeks ago in California,  uh, we launched, uh, engaged California and
• 15:30 - 16:00 the first topic to be discussed, uh, is how  to recover from the wildfire for Eaton and   poly Sade. And that is the kind of topic that  has this urgency for clarity and is far from a   single department's purview. And so I do think  that for this kind of topics like Calfornia is. 40 million people. It's not  a scale thing. It is the, uh,   will of the people and the  actual urgency for clarity,   these two, um, merging together that creates  opportunity to launch this sort of platform.
• 16:00 - 16:30 So there's the technology  itself, like what it does,   but then kind of separate from that is  the people's trust in the technology. Mm-hmm. And you said since you did it for  10 years in Taiwan, there was like a social   approval mm-hmm. Because people were used to  it. What's the threshold beyond which people   believe this? Like, could this happen in the  United States, um, now on some issue that isn't   existential, but, but is interesting to  people and, and relevant to their lives?
• 16:30 - 17:00 Yeah, I think it's also now, uh, ongoing  in Bowling Green, Kentucky, uh, for the,   uh, better bowling green, uh, consultation.  Uh, and so it's not like urgency, urgency,   but uh, obviously people do feel  that there is some value, uh,   in closing the loop of the conversations in the  neighborhood, the mayor paying attention to it,   and then using AI to figure out what's the  uncommon grounds, um, despite the differences
• 17:00 - 17:30 that people have in the society and how those  measures, uh, can really improve people's lives. And closing the loop and telling the people  who initially propose those ideas is because   these words you wrote, and of course the other  3000, uh, people that this measure was taken, was there any evidence that  within the 45 groups of 10 people,   each or any other recent example  that the 10 people themselves. In the, the process of discussion and debate  that was facilitated, facilitated by AI that
• 17:30 - 18:00 they learned and changed their mind, or they,  they altered their position on the issue? Yes, definitely. Uh, if you look for the, uh,  deliberative Democracy lab in Stanford,   uh, which we partner with, uh, for  both Engaged California and for this   information integrity consultation,  uh, they have a lot of research. And the most, uh, important takeaway for  me is that this inoculation works in the   long term. So not just p do people, uh,  entertain. The other side's, uh, visions,
• 18:00 - 18:30 uh, in a kind of surprising validator kind  of way. So, I, I may not like your politics,   but your suggestion makes sense to me.  This actually influenced their decisions   even like a year after such exposure to a  citizen assembly, so that when they vote,   uh, they tend to look at the actual  measures, uh, the actual issues at hunt. Uh, instead of just jumping  into partisan politics. And the people, the 10 people in each group,
• 18:30 - 19:00 did they know that the facilitator  was an AI and not a real human? Yeah, because it's not an avatar  or anything. You just see, uh,   that the transcript, uh, appears, uh, as you  speak. You just see a kind of little poke,   uh, when you've been too quiet, uh, and so on. So it's not like a AI pretending to  be a human facilitator. It's more   like this room itself has a facilitating function. So, so in addition to facilitating, uh,  different, uh, priors and, um, ideologies,
• 19:00 - 19:30 it also, um. Equalizes in a, in a different way.  Because if you get 10 humans together, uh, various   power laws ensue and one or two or three of the  people are gonna do 80 to 90% of the talking. Mm-hmm. But this actually upregulates the  quiet and downregulates the, the chatty. Yes, that is correct. And the reason why is  that we do want the voices, um, that, uh,   reach this uncommon ground, uh, to have some way  of, um. Amplifying their reach. This is in stark
• 19:30 - 20:00 contrast with the antisocial corner of social  media where the only most polarized, most extreme,   the dunking, uh, that gets amplification  because that's a broadcasting network. It's not a conversation network. And so in  weaving together a conversation network,   we want to upregulate the kind of voice that  resonates with the entire room. And to do that,
• 20:00 - 20:30 you probably have to make sure that people,  uh, take turns, uh, listen as well as speak. It's really quite impressive,  and I am not such a fan of ai,   uh, to be blunt, but this, this is  one of the good, good sides of ai. Yes. I think that's because it's using  AI as assistive intelligence. So   just as the assistive technology  you are wearing the eyeglass, uh,   it's not replacing a human in the human to human  relationship. Rather, it is enhancing the human   to human relationship. And this assistive  use of AI also respects the dignity of the.
• 20:30 - 21:00 People, uh, in a conversation so  that they feel they can steer, uh,   this conversation, not your eye glass  steering, uh, the conversation. And so   I think when we talk about ai, we often  think in a kind of automating fashion,   like replacing a human in a human to human  relationship or reducing humans to machines. But assistive kind of intelligence doesn't do  that, is task only and is not trying to be this
• 21:00 - 21:30 general, super intelligent that dictates the  human's logic. And so it's not about aligning,   uh, humanity to. The digital AI logic,  it's about the individual digital tools   like eyeglasses that can align  to the human to human logic. This is very impressive, Audrey, and I  know it's, uh, it's approaching midnight   in Taiwan. Mm-hmm. And, and your clarity,  uh, on this is, is very, um, helpful. Uh,   let me, let me take a step back in, in  your history. Mm-hmm. Eventually, your,
• 21:30 - 22:00 your work with the Sunflower Movement turned  into some other projects, uh, gov Zero and Paul. Mm-hmm. Um, can you give a brief account of what  those two projects were and, and specifically how   they relate to a concept that you describe  as demonstrating rather than protesting. Definitely. So G zero V tw,  that's the domain name, uh,   was registered before Sunflower in  in 2012, um, by some of my friends.
• 22:00 - 22:30 Uh, I joined almost full-time, uh, in 2013.  And the way we work is we look at all the   government services, like something  the Go v tw, and if we don't like it,   whether it's budget or something,  instead of, uh, just, you know,   protesting that it's bad, we actually make a  better version as something that G zero V tw. So I talk about the National  Participation Platform, join the gov tw,   and if you don't like that, you  can change your O to A zero and go
• 22:30 - 23:00 to join the G zero V tw, uh, which is the  gov zero version. But because gov zero is   always. Free software, uh, and open culture,  meaning that our products, uh, are Forex. That's to say alternate versions of the government  versions. But we also relinquish sufficient amount   of copyright so that if a government wants to,  they can always merge it back into government   service. So quite, uh, famously, uh,  during the pandemic, um, the gov zero
• 23:00 - 23:30 people developed a alternate way to do contact  tracing that does not compromise privacy at all. So instead of government version,  the government simply say, okay,   let's use the gov zero version. And  that resulted in Taiwan, you know,   not locking down any cities, uh, during  the three years and actually held for,   until Omicron, uh, which is no mean feat. And  TSMC just keeps running. Um, anyway, I digress.   And so the gov zero, uh, try many different  things, but including Polis and Polis was.
• 23:30 - 24:00 Before generative ai, before language models  for sensemaking, um, you can think of it as a,   um, visualization of where people stand  on a issue. So for Uber, for example,   we ask people to chime in and they go online  and they see a fellow citizens, uh, feeling. For   example, somebody may feel that, uh, undercutting  existing meters is very bad, but search pricing.
• 24:00 - 24:30 During, uh, high demand. That's very good.  Uh, so somebody may, may have this statement,   you can agree, you can disagree or  you can pass, but there is no ance,   so no room for truth to grow. And so it  is in a synchronous way, simulating a   little bit of the 10 people room dynamics by  highlighting what's the most resonating idea. And so you see your avatar being sorted to  one room and this room, uh, have these kind   of agreements, but you also see across all the  different cluster, different rooms, what are the
• 24:30 - 25:00 ideas that are currently gaining grounds that  everybody, regardless of where they're coming   from. Do agree. And so after three weeks in 2015,  we agree on the set, a very coherent idea about   Uber, which we then pass into law so that, uh,  the local co-ops and so on can also operate. And Uber is a legal taxi fleet, uh,  in Taiwan for quite some years now.   So the idea is to use asynchronous,  uh, contribution, uh, and discovery   of the uncommon ground so that even if we  don't have, uh, the language models, uh,
• 25:00 - 25:30 to weave things together, people can still kind  of see the community notes that flows to the top. And the same algorithm has been  adopted by YouTube, by meta,   and by X as the community knows algorithm. Wow. So, um. Embedded in there, uh, is your  emphasis on data about feelings, specifically the   feelings of the citizens living under these laws  and, and regulations that a government enacts?
• 25:30 - 26:00 Mm-hmm. Why is that so important to incorporate,  uh, those values into decision making? And by the way, do you know, um, Nora Bateson and  her work in what's called, uh, warm Data Labs? Mm-hmm. Yes, I've heard of, I've not worked  directly, but yes. Okay. But go ahead. What,   what about, uh, data and feelings?  The integration of that? Mm-hmm. Yeah. Uh, first of all, I think  we're all experts of our feelings. Uh, and so that is actually what can easily  resonate with our fellow citizens. Had we
• 26:00 - 26:30 start our Uber consultation with,  what's your ideal economic model   for sharing economy versus extractive gig  economy? Uh, probably nobody will come,   right? Because it was, um, like. Extremely  abstract, but feeling is not abstract at all. Feeling is very personal. And so based on  feeling, then people want to take care of   each other's feelings. So you can see  like the Uber driver, the taxi drivers,   the passengers, the people worrying  about rural development and someone,
• 26:30 - 27:00 um, they all center around shared feelings. And  so naturally when people start proposing ideas,   those idea that take care of everybody's  feelings will float to the top. And so this speaks to a very different ethical,  uh, foundation of policy making. This is more   about the ethics of care. That is to say,  how much do we want to take care of each   other instead of what's. Single abstract  value, um, like in a scholar value sense,
• 27:00 - 27:30 do we want to optimize? Right? Uh, and care  also has the benefits of, um, its positive sum. So if I take care of your ideas, then you  are probably going to propose an idea that   also take care of my feelings, uh, as opposed  to if you put it to referendum or something   as Uber did in other jurisdictions,  maybe 51% people feel they have won,   maybe 49, feel they have lost, but their  feelings are hurt and are therefore more
• 27:30 - 28:00 likely to engage in negative sum uh,  conversations from that point onward. So what did those projects, um, tell you about  the divisiveness and polarization of the societies   where they were enacted, and did people respond  well to, to these technologies? Like, oh, I, this   feels more, uh, positive, some and, and caring,  and, and did they notice that? Yeah, definitely. Um, so, uh, we can, uh, look  at very objective numbers,
• 28:00 - 28:30 uh, especially the very young people in 2019. We changed our curriculum. So instead of,  uh, the standardized answers, you know,   that the East Asians are very famous  about, uh, we switched. To prioritize,   uh, the civic competencies, uh, namely, uh,  autonomy, that's curiosity, interaction with   people who are unlike you, and also the common  ground, uh, the ability to construct common good. And so the idea here. Is that if we do not have  this shared, uncommon ground in for young people,
• 28:30 - 29:00 young people will feel they're very  detached from politics. They're just 14,   15. They have no way to contribute to  agenda setting, even though they do know,   uh, what is actually better,  uh, for the planet and people. But by making sure that the young people  have agenda setting power, uh, in setting,   for example, e petitions or even becoming, as  I mentioned, cabinet level advisors and so on,   the Taiwanese 15 year olds, according to ICCS in  2022 are now. Populative world when it comes to
• 29:00 - 29:30 the agency. They feel that they can affect  the society for people and planet issues,   and they still, uh, maintain the number  three to number five, uh, PS a score. So people are also happy that their stem isn't  actually degrading. It's not the trade off, uh,   but I think the young people's empowerment as  well as the depolarizing effect across, um,   religious, uh, urban, rural, uh,  age brackets, uh, and these, uh,   Taiwan is also the least polarized among  OECD equivalents A couple years ago.
• 29:30 - 30:00 That's amazing and important, uh,  because there's two issues. One is   using this technology to actually change policy  and regulations and, and things. But the other is,   irrespective of that, this technology, um,  uh, suppresses apathy and provides agency,   which is essential in our current world  because there's more and more people   with, um, mental illness and, and just  checking out because it's so much,
• 30:00 - 30:30 because they don't feel they have agency  against all the things that are going on. So this technology could be really  important just as a, a vector to,   um, to increase the feeling of agency. Yes. Yes. And it also has what we call a pre  bunking effect because if there's   already a polarized fight between the two  memes, uh, then trying to arbitrate it, um,   like especially from the government, uh,  tend to just, uh, kindle the fire even
• 30:30 - 31:00 more and people become even more polarized,  uh, and fuel conspiracy theories and so on. Uh, but this kind of technology  allows us to discover, uh,   the uncommon ground and share it as  pre bunking. So one very early, um,   example, pre is pre bunking. Yes. So it's not  debunking Oh, oh, it's not debunking. It's pre   bunking. Yes. Debunking is After something goes  viral, you say, oh, that's not quite the case. Uh, pre bunking is that before something  goes viral, you already say, by the way,
• 31:00 - 31:30 this is actually like this. Right? So, so it's,  uh. Many people feel that if they pre bunk each   other, they are less likely to be polarized.  And there's many ways to pre bunk and humor   is one large part of it. So in early 2020, uh,  when people are not sure what the coronavirus   interaction with Musk are in Taiwan, we already  observe as in other places, like one side says,
• 31:30 - 32:00 because we had a SARS experience a  few years ago, people feel only N95. The highest gray mask are useful  and every other mask are actually,   you know, a, a scam or something. Uh, and  the other side says it's ventilation is   aerosol. So wearing a mask hurts you  and wearing N95 hurts you the most,   right? So, uh, if we just let these two  polarized memes, uh, grow, then they   tend to fight each other and people will, um,  basically polarized into mask anti-US camps.
• 32:00 - 32:30 Uh, but the science, uh, was still not  very clear then. So we basically. Pushed   out the meme of a uncommon ground, uh, very  quickly, and it's a Shiba inu, a very cute dog,   putting her pole to her mouth, saying, wear a  mask to remind each other to keep your dirty   and wash hand from your face. So, so that's a  uncommon ground no matter which part you are. You probably agree that hand washing is  good. Uh, we actually measure tap water
• 32:30 - 33:00 usage. It actually increased, uh, and because  the dog is just so cute, if you laugh at it,   the next time you see somebody wearing a mask or  not wearing a mask, you would just think about,   uh, you know, hand washing, uh,  which is like not polarizing at all. Uh, everybody washes their hands. So just like,  there's no pro fraud camp in Taiwan. There's no,   you know, anti hand washing camp in Taiwan. And  so it just diffused the polarization into just,   you know, hand washing. There's also songs about  it and the cute dog dancing and things like that.
• 33:00 - 33:30 So, um. Yeah, this is like literally, I'm. Soaking this all up because I think it's,  it's so important and I take our current   social media landscape as a, as a given,  and I, I've stopped using Facebook and, uh,   I do use the other things to post the content  of, of this website, but I'm become really   disenchanted with social media and this, this is  exciting to learn that these things are possible.
• 33:30 - 34:00 LL let me continue. Ultimately, I  believe you've rooted your work, uh,   in the idea of plurality. Mm-hmm. Uh, which I  think is the name of the book you co-authored   with Glen Weil. Mm-hmm. Can you describe  what is at the core of, of pl plurality? Yeah, certainly. So, um, singularity. Means,   uh, an AI that can improve itself, uh,  increasingly without human uh, control. And at some point the AI can automate everything  there is to automate about AI research. And then
• 34:00 - 34:30 either, I guess, grow a self preservation instinct  and refuse to develop the next generation of AI   and kind of see us as competing carbon-based  species, or, uh, they don't get that. Uh,   and just recursively self-improve and serve  not themselves, but maybe, you know, a CEO. And then the CEO becomes. Transhuman, uh,   and then become a very different species than  the rest of us, right? So that's singularity.
• 34:30 - 35:00 Thank you for that. I've,  I've heard that word a lot,   and that was the best description,  uh, as horrifying as it is. Um, but thank you, you for that. Please continue. Uh, it is, uh, kind of  losing the race, um, of humanity, right? Yeah. It's not a race of, um, ascension, uh,  as sometimes portrayed, but for the rest of us,   it's just the humanity race loses. And so  plurality says, um, instead of making an AI
• 35:00 - 35:30 that's even more powerful by the day  recursively, we should, uh, actually   enhance the way that people can work across  differences. So design each piece of technology. It could be ai, it could be immersive  reality, many technologies, um, with this   eye on fostering. The differences, but seeing the  conflict that ensues not as fire to be put out,   but as energy can be harnessed, uh,  for co-creation. And so any sort of
• 35:30 - 36:00 technology that enhances this collaboration across  differences, uh, is in the direction of plurality. So instead of a vertical race, uh,  of takeoff, uh, escape velocity,   you see a lot of space based, uh, metaphors.  Uh, the plurality is entirely horizontal. It is,   uh, a lateral diffusion of technical  capabilities. And each capability is   steerable by the community that's deploying  it. And so the more we invest in plurality,
• 36:00 - 36:30 the better we're prepared, uh,  to face all the emerging harms. That's. Being caused by advanced AI and so  on. And the hope is that at some point people   will just discover that this is a better,  a more worthwhile direction. Maybe it's not   worthwhile at all to replace our, uh, human  race, uh, with some other, uh, silicon based,   uh, stuff. Un unless you're the CEO. Yes. Uh, and  as we have seen, uh, when people said to the CEOs
• 36:30 - 37:00 of big tech, uh, from Taiwan that you need to  be liable for whichever scams advertisement   that you put on because you've been earning,  uh, advertisement dollars from those scammers. Uh, and the entire society is paying the  consequences, the cost of such negative   pollution, externalities. This is the  kind of plurality technology that quickly   lets the decision makers reign in the  CEOs. And so I do believe that this
• 37:00 - 37:30 steerability comes from the button up,  but it also does need, uh, endorsements,   uh, from the regulators, uh, to say  basically, okay, it's not my idea. It's like a trade, uh, negotiation.  It is the people's idea. So is that kind of, uh, a plurality is kind of  like a decentralized singularity. Well, it's a acceleration for decentralization, for  democracy, and also for defense. Uh, so Vitalik
• 37:30 - 38:00 Buting caused this d slash acc or defensive  democratic decentralization acceleration. So it is an kind of acceleration in that we want,   uh, the most possible equitable way  of diffusion. Uh, but it's accelerates   not in the sense of self-improvement,  like the vertical singularity one. This could be applied in a lot of different areas.  I'm specifically interested in how it could be   used for the ongoing battle of what the  future of social media could look like.
• 38:00 - 38:30 Mm-hmm. Especially with our. Aims of,  of this podcast and your work and a lot   of our colleagues and people in the  world for a pro social, uh, future,   what would be specific features of a  social media platform rooted in the   ideas of plurality and how would those look  different than the platforms we have today? I'm, I'm sure you've thought about  this and if not, are working on it.
• 38:30 - 39:00 Yes, certainly. So, uh, I co-authored a paper  called Pro-Social Media that talks about this.   The idea very simply put is that in your newsfeed,  instead of being ranked by the engagement or   addiction that it generates, it can rank instead  by the various communities that you belong to   and how much coherence, how much uncommon ground  each post can generate between those communities. So. Each of us have very different like spiritual,  professional, family and so on circles. And it's
• 39:00 - 39:30 often the case that we ourselves are also figuring  out how to take something that we feel cherished   from one context across to another context.  And the idea is that there are creators, uh,   on social media that specialize in creating this  kind of bridges so that people can understand   the other community more and vice versa just  by viewing and engaging with such content.
• 39:30 - 40:00 And so for each post, you can then see  of the communities you belong to, uh,   which communities find this to be bridging. And  which communities find this to be debatable.   So it's like the polish interface, but  apply to social media. We already have   that in the form of community notes,  but it is kind of a debunking thing. Uh, you already have a trending, polarizing post,  and then you can look at the community notes, uh,
• 40:00 - 40:30 to have this kind of resonance and bridging.  So the intuition is to move this into the   main feed so that the main feed itself becomes,  uh, pro-social. And in the paper we talk about,   for example, I'm involved, uh, in advising  the Project Liberty Institute, uh, who, uh,   works out a new economic model, uh, for TikTok if  the people's bid succeeds, uh, in buying TikTok. Us. And so instead of the advertisers paying  to bid for the highest bid, uh, getting the
• 40:30 - 41:00 attention of each individual kind of strip mining  the social fabric and making each person look   at a wildly different feed, uh, this idea is  recreate this common experience so that people   can know, oh, um, your community and that  community are enjoying this, um, together. So a little bit like those  10 people in the same room,   people will be able to know that this is, uh,  white resonance with the extended communities
• 41:00 - 41:30 and it creates kind of a Super Bowl effect  and things like that. And we conjecture   that the communities as well as brands will,  uh, pay for this kind of shared experiences. So how prevalent are these various technologies?  Some of the things, uh, the project you,   you've mentioned in Taiwan. Mm-hmm. And  is there any evidence that on some, uh,   um, group of issues that Taiwanese population is  less polarized than mm-hmm. Than other countries?
• 41:30 - 42:00 Definitely, as I mentioned, across urban, rural,  across, uh, age groups, across religion and so on. Taiwanese people are the least, uh, polarized. And  we can also simply compare the pro-social ranking   algorithm that's deployed in LinkedIn, uh, versus  say in Facebook, um, LinkedIn Q rates. Uh, its,   uh, feed in a way that is not maximizing  the time you spend on advertisement,   but rather on the, uh, cohesion, the  coherence that we just talked about.
• 42:00 - 42:30 And so feed is quite different, or LinkedIn  In Taiwan? No, no, LinkedIn globally. Uh,   when they first introduced the newsfeed to  LinkedIn, they were very intentional and   then they curated this kind of common ground  bridging, uh, posts from. Business leaders,   uh, from people who follow, uh, who are  followed by a lot of people on LinkedIn. And then they gradually, uh, open up commenting  and things like that. But the whole idea is to
• 42:30 - 43:00 shape a norm where, uh, engaging with the  feed actually adds to your, uh, sense of   social cohesion instead of, uh, distracting,  uh, from, its like Facebook did since 2015. So what are the barriers to this scaling, uh,  pro-social, plurality based, um, social media? What, why isn't this I. Taking off more, this  feels like something that people would want, uh,   of all political ideologies and and backgrounds.
• 43:00 - 43:30 Yeah, definitely. Um, and it is true that I've  been talking with, uh, many different people on   different sides of ideologies and they all feel  that it's time to move past peak polarization. And I do think that what we need now is,  uh, both strategies. One is working with,   uh, free software communities that runs those  smaller but still very respectable sized, uh,   networks such as Blue Sky with 30 million people  on one side, and also choose social on the other,
• 43:30 - 44:00 which is also free software. Uh, and,  uh, in a way to show that we can bridge. The contents so that people across true social  and blue sky can find the uncommon ground,   the, uh, surprising validators. So this  is what we're doing. And the other is   just to take an existing network like  TikTok and just change its algorithm.   And the idea of people spit is  that TikTok needs to interoperate.
• 44:00 - 44:30 Meaning if you post on TikTok, you should  be able to consume the same content and   link to the same friends on Blue Sky or  on true social or on any other places.   And so people will then be able to  curate their own experience instead   of feeling locked in to the core, uh,  recommendation algorithm of TikTok. And so this gives us much more grounds to  experiment with the pro-social ranking. Just like everything else in our world though,  isn't, um, our global economic system, uh,
• 44:30 - 45:00 our national economic system, our corporate  economic incentives are based on, uh, dollars and   we get clicks for dollars. So, you know, when we  use social media, we get some benefit and mm-hmm. A lot of times it's dopamine based  instead of oxytocin based. Mm-hmm. Um,   to, to make a generality, but it  results in an economic, uh, uh,   gain for some individual or corporation.  Mm-hmm. Does, does this still, um,
• 45:00 - 45:30 does this combat that at all? Or how,  how does that play into this? Yeah, the, the hope here is just as LinkedIn has  demonstrated, there is a way to pay for. Common experiences and oxy toin based, uh,  feelings, uh, while still making sure that   whatever advertisement, whatever messages, uh,  that you pay, um, can result, uh, in like Super
• 45:30 - 46:00 Bowl, uh, which is the kind of pinnacle of  common experience. And then you can build   narratives and brands and so on in a way that  individualized dopamine hits, uh, really cannot. Seriously. I, I think our culture has  like a massive dopamine hangover. Um,   they may not know that, but we're so depleted.  Uh, it's like we've all been on this Las Vegas   junket and have lost all our coins and our  brains are kind of fried and we're hungry for
• 46:00 - 46:30 serotonin and oxytocin. Other of our ancestral  neurotransmitters that we've been craving,   and we get that through community and  community engagement and social interactions,   and the fact that we can possibly  get that from social media mm-hmm. Is encouraging. Um. Don't you think? Yes. Uh, and there's a famous  study a year and a half ago,   uh, a average undergrad in the US using  TikTok. If you ask them to move off TikTok,
• 46:30 - 47:00 then you will have to pay them almost $60 a month  so they lose that much utility like fomo. Uh,   while everybody else is still on that hamster  wheel, but if there's a magic button you can   press that can transplant everybody around them  and themself into some other like non dopamine   based, uh, platform, uh, then they're  willing to pay you almost $30 a month. Um, and so it's obvious we're in a product  market trap. Mm-hmm. Everybody lose utility
• 47:00 - 47:30 on the hamster wheel, but the  first one to move off suffers   so much fomo so that nobody want  to be the first that moves off. Hmm. That's quite profound and  dopamine is still worth two x,   uh, serotonin and oxytocin in  our current economic system. But that might change. Yes, that  might change. Um, so, so you are,   uh, in your work, um, you're very specific  in your projects and initiatives about the   use of language and the importance of  it. So why is language so important,
• 47:30 - 48:00 uh, in these movements and for civic  engagement and, and participation in general? Yeah, I think repeating, uh, the category errors,  uh, of some business as usual language such as,   I dunno saying human resources, uh, or  incentivizing corporations, uh, just.   Propagates this category error in our thinking.  So it's like, um, trying to chart out a map, uh,   but with like very tilted, uh, lens, uh,  you, you can't perceive the world, right?
• 48:00 - 48:30 Uh, if you use that sort of category error, um,  where it's, and so in 2016, uh, when I first   entered the cabinet as the digital minister,  uh, I made a word play because in Taiwan,   digital shuway also means plural. So I'm not just  a digital minister, I'm also the minister for   plurality. I. So even though there's no ministry  at the time, the ministry will come in 2022. I still wrote a job description, uh, as a shuway  uh, minister. It goes like this very quick. When
• 48:30 - 49:00 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 the singularity is near,  let's always remember the plurality is here. Nice
• 49:00 - 49:30 work, Audrey. Thank you. Um, I do think law, um,   um, language is so important. Like fossil  fuels. They're not fossil fuels. Mm-hmm. Um,   they're fossil hydrocarbons. We're just  choosing to use them as fuels. Mm-hmm. As, as one example, or we refer to the  United States consumer spent more this month,   like we're human beings who buy food and other  things. We're not necessarily consumers, uh,
• 49:30 - 50:00 unless the true ecological sense, but yeah.  Language is super important. Mm-hmm. Um, mm-hmm. Yes. Because we're marketing to each other. Yes. Consumer of foods is like, you know,  referring to your users and, and it sums this,   you know, drug subscription, uh, case, right? So  I think when I say user experience, uh, should be,   instead, human experience, we're pointing  out the same thing. That is to say there's   much more to being human than just consuming  something or getting addicted on something.
• 50:00 - 50:30 So I've heard you, um, describe liberal democracy  as a sort of social technology. Mm-hmm. That   should be in constant innovation, uh, alongside  other technologies. Mm-hmm. H how would you   describe the current state of innovation for  democracy itself and what is needed for it   to keep pace mm-hmm. Uh, with other things  in parallel that are going on in our world,   like artificial intelligence  and other disruptive technology?
• 50:30 - 51:00 Yeah, that's a great question. So, um, I  analyze, uh, democracy as a communication   technology that has, uh, bandwidth  and latency. Bandwidth is how much,   uh, information can each citizen communicate  to their communities and also into decision   making. So if you have a referendum, that's  one bit of information. If you have a, um,   votes on mayor, uh, with four plausible  candidates, that's two bits of information.
• 51:00 - 51:30 The problem is that the emerging technologies,  they change our world in a way that demands, um,   solutions to what's called wicked problems.  Meaning that issues that require coordinated   action of many, many different parts of  the society. But if each part of society.   Can have two bits, three bits of information  uploaded, then that's not sufficient information   to piece together a solution, a kind of  jigsaw puzzle, uh, to the wicked issue.
• 51:30 - 52:00 And this is one part, and another part is latency.  If you have to wait for four years for the next,   uh, mayor or the next referendum and so on.  Well, um, many incoming transformative threats,   uh, can change the society to the point of no  return in less than four years. And so think,   um, not just pandemic, but also the info  dynamic, uh, the polarization issue,
• 52:00 - 52:30 the, uh, generative ai, power  scams, phishing, and so on. So all of these, you, you do not wait for four  years and start a new referendum or vote in a new   mayor or things like that. You. Immediately  gets people together and very quickly gets   much more bits than just a vote. Maybe you  get conversations which is much more bits,   or instead you get, uh, reflections  on each other's posts and so on. Like in poll, no matter which way, you need to  close the loop very quickly so that people know
• 52:30 - 53:00 that within weeks or at most months, your idea  results in the steering of the, uh, direction   of the technology and its responses. And then  people can come around again and again to learn   the steerability. So I'm the cyber ambassador  and cybernetics in Greek means steering. So this is about the art of steering. I didn't know that. Um, so is there a risk  that if we don't continue to innovate, uh,
• 53:00 - 53:30 democracy as it is today mm-hmm. And all  the liberties and freedoms that we've come   to take for granted, uh, in our generations, that  democracy will simply become obsolete in the face   of accelerating AI towards the singularity  and the changing global political landscape. Um, how worried are you about that and  how do those concepts, uh, interrelate? I think, uh, there are various ways that people  can see the incoming crisis, which is not just
• 53:30 - 54:00 one but many. So some people say PO crisis,  uh, but they're all isomorphic in the sense   that if you see one crisis, you've also seen  the shape of some of the other crisis as well. So like a meta. Crisis. And so I  do feel that, uh, our experience   when it comes to whether it is occupying the  parliament peacefully and keep it peaceful,
• 54:00 - 54:30 or whether it is about countering the algorithmic,  um, dispatch of Uber and of social media and,   uh, the infoam and also the pandemic,  uh, and generative AI harms and so on. Each of these examples, uh, shows that  maybe a crisis, uh, as in weight is both   in a danger and an opportunity. And so  the shared danger is likely to. Make sure   that people see the societal resilience as  not a nice to have, but rather something
• 54:30 - 55:00 that people must contribute to. So  the wildfire, um, recovery issue,   uh, on engaged California is a great, uh,  result of this infrastructure level building. And then when such a topic comes in, then people  can pivot and respond very quickly to it. So I'm   not pessimistic at all. I feel that each of  those incoming threats actually accelerate   the diffusion and the common knowledge  of the people that democracy does need,
• 55:00 - 55:30 uh, improvement as the social technology. Audrey, why are, are concepts, uh, like  responsibility, liability, inclusivity,   and transparency, um, important, uh,  for creating and maintaining an open   democratic governance system of, of the  type that you've been describing? I. Yeah. Um, I learned this, uh, when I entered the  cabinet, um, because, uh, in 2016, uh, I entered   the cabinet with some of that doge energy, you  know, uh, wanting to make everything transparent,
• 55:30 - 56:00 want to make a procurement, like a leaderboard  of people comparing, uh, and things like that. Shortening the, uh, tax filing from three hours  to three minutes, uh, through direct file, um,   and so on. And so all these, like what  we did that in like 2016 and so on,   but we very quickly found out people in the  career, public service, the career public   servants. They also had the same idea,  and they are also like great reformers.
• 56:00 - 56:30 They actually know how to do things  better. It was just they lack a air   cover. There's no one who say,  uh, if you do this well then, um,   it's you who get a credit and if you do this, uh,  but it doesn't work and I can take the blame. Uh,   and so I made sure that we align our, this  energy of democratic innovation to the. Languages and the logic, uh, that the  career, public service, uh, especially the,
• 56:30 - 57:00 uh, planning and research and development  departments use. And so in Taiwan we have   the National Development Council, and to them  always, uh, transparency, accountability, uh,   is I. The norm. Uh, and if we add participation  and inclusive participation at that to it,   they want to know that this participation  is accountable so that we can regulate, um,   this institution into new institutions, not just  challenging and taking down existing institutions.
• 57:00 - 57:30 So we announced our every move,  everything like the join platform,   the participation office and so on. Uh, instead  of just doing it, uh, as code, we said, okay,   six days from now we're going to do it and  here's a public commentary period. And we made   sure that there's no exceptions. Everything  needs to be pre announced publicly this way. And so even though that each of our move takes  like 60 days more, I think we want much more
• 57:30 - 58:00 support from the career public service because  they can see that. I'm designing myself out,   so to speak. Uh, if I'm no longer  the minister, all those institutions,   the new designs are still around because it  conforms to the logic of the bureaucracy. I imagine that there are many other countries in  the world, some countries are very interested in   copying your success in Taiwan. Mm-hmm. And  others are also afraid of, of implementing   some of these things. Mm-hmm. I mean,  in your opinion, should countries, uh,
• 58:00 - 58:30 be doing more to regulate social media platforms  to be in line with these principles and, and what   are some of the, uh, the benefits and risks to  such government oversight and any comments there? So for this kind of broad listening and  sense making, I think the smaller the polity,   the easier it is to implement. Uh, to your point  about Dunbar's number, pretty much any polity,   if it's just 150 people, they don't  have to run a sortition. They just
• 58:30 - 59:00 invite, uh, everybody right to a conversation.  Uh, and we do see that in many countries,   like in Japan there's a long tradition of  citizen assemblies, but on a hyper local level,   like literally township level,  uh, and that, uh, has worked well. Do we have the technology to do that at  a township level now? Yes, we do. It's the same technology. It's just  easier to implement, uh, and gets buy-in   from a mayor of a town as opposed to say, you  know, a federal government. Right. So it's usually easier to start. I want you  to finish answering this question,
• 59:00 - 59:30 but just so I understand, I. In the United States right now, people in  Topeka, Kansas, or Red Wing, Minnesota mm-hmm. Or   Sebastopol, California mm-hmm. Could access some  existing technology right now to do Oh, yeah. Some   of the things You're ta what, what would that  be? Yes. What TE technology. Yeah. As, as I mentioned, the Bowling Green,  uh, process is ongoing, right? So if you just search for Bowling Green,  uh, Kentucky, uh, sensemaking, uh, or polis,
• 59:30 - 60:00 uh, or better Bowling Green, uh, you can see  exactly how it's done. It's all open source,   not just the P platform, but  also the sensemaking tool. Uh,   they're all free software free for anyone to  use. And so there are some US states, uh, with. Citizen assembly tradition already in  an in-person kind like in Oregon. Uh,   and so in that sense then it's not  about convincing them to move online,   but rather using digital tools  to augment the conversation and
• 60:00 - 60:30 to improve its reach. So like Democracy Next  has been working with Oregon people on that. So the Bowling Green and the Oregon, there are,   there are entities that are working  and chaperoning that process. Mm-hmm. But in theory, anyone listening to this  show could look at the Bowling Green example,   access the source code, and start  something in their own community. Yes, definitely you can roll out Polis installations. Uh,  at PL is, and the sales making tools, uh,   you just search for Jigsaw, sensemaking and  Polis, I think now have integrated that logic.
• 60:30 - 61:00 So it can also use language models  to do a very balanced reporting of   people's ideas. So you can close  the loop like literally within,   um, a minute or so for the mayor,  uh, to maybe read every morning. Let, let me ask you a, a related question.  Not, uh, to do with democracy per se. Um,   but I've noticed, um, over the years,  um, decades of convening groups of.
• 61:00 - 61:30 High status scientists and activists, that  everyone's got an opinion and they're very smart,   and you get 80 or a hundred people  together. But what ends up happening   is when you're in person or when your  name is attached to something, people,   since we're social primates, uh, and we  compare and look at status metrics, they   defer to the senior wealthiest, or most famous,  or most influential per person in the group.
• 61:30 - 62:00 And so they don't mm-hmm. Let their, their  real thoughts, um, be known. So I'm wondering,   the technology that you just described  about the Bowling Green, could that be   used mm-hmm. In an institution itself where  there's 200 people and you really wanna know   what people are thinking without fear of saying  the wrong thing and getting demoted or anything? Is, would this apply to those situations as well? Yes. And, uh, there are technologies for  the in-person, uh, kind like cortico,
• 62:00 - 62:30 C-O-R-T-I-C-O, and develop out of MIT. This  tool, uh, you can just put your phone or a   round microphone on table, and then it ensures  that the facilitator is guided by not just, uh,   the conversation guide, the turn taking,  you know, not letting the single senior   person dominate conversation, but can also carry  other conversations from previous, uh, talks, uh,   to this particular conversation pod so that  the conversation network can cross pollinate.
• 62:30 - 63:00 So when the most senior person speaks something,   the facilitator can then press a key and  then a method, uh, place from some other   conversations that counterbalances,  uh, the point that was just made. Why didn't I know about this? And what is, what  is holding this sort of technology back? Is it,   is it awareness, uh, like in my  case, or is it money or is it,   um, big tech is, uh, afraid of these  things, uh, or is it social organization?
• 63:00 - 63:30 Why aren't these things scaling more rapidly? I think, uh, one of the main reason, uh,  was that all these things run on oxy,   toin and serotonin, right? And so it is a,  it is a vibe thing. Once you're in this vibe,   uh, then it's more likely that you  will participate in one of those, um,   conversations and you will discover a very  large rise on like, conversation network. But if you're dopamine bound, it's very difficult.
• 63:30 - 64:00 Yes. So actually we need to heal  people's dopamine addictions, uh,   concurrently so that they move into this more, uh,  um, zen, uh, holistic human experience. And then   obviously this is the type of social media that  I would prefer rather than clicks and likes and,   and unexpect reward of, of some goat that claps  and falls down and a snake crawls under it. And woo, I never saw that before. Um,
• 64:00 - 64:30 which doesn't really give us much meaning  or depth or purpose to our lives Anyways. Oh yeah, definitely. In my phone I have, uh,  turned on the color filter. Uh, you can go   to settings and choose color filters, so it's  almost entirely gray scale, just with a little   hint of color, uh, so that the phone is never  more vivid than reality and it works wonders. Uh, so I cannot get pulled into the do  because, um, this, uh, Las Vegas thing,   uh, this slot machine, uh, simply does  not give, uh, high enough, uh, rewards,
• 64:30 - 65:00 uh, when your phone is grayscale. Oh,  that's a great idea. Yes. I'm gonna do that starting today. It's called color  filter. I'm gonna do that. Mm-hmm. So, uh,   moving on to a more serious topic, not that the  things we've been discussing aren't serious,   but how might the events we're seeing  right now, especially in the United States,   playing out with, uh, with big tech  and tech oligarchs, damage people's. Inherent trust in technology that might limit,
• 65:00 - 65:30 um, some of the opportunities you've been  describing. Um, what do you think about that? Yeah, so on one side, uh, we do  see that people are collectively   feeling it's time to move past. Peak  polarization. On the other hand, uh,   aside from like more people using say  blue sky or true social or signal or   proton or things like that, um, there's yet  to be a very coherent movement out of the.
• 65:30 - 66:00 Big tech dominated social media landscape toward a  more pluralistic, uh, pro-social media landscape.   That is true. So this is partly what we are trying  to achieve, uh, with this paper and advising the   Project, Liberty Institute doing the TikTok  bid. But regardless of whether the TikTok goes   to become a prosocial space, I do think that, uh,  there are pockets of good within those big tech. So the Bowling Green Experiment, for example, is  done by the Jigsaw Group within Google. So there,
• 66:00 - 66:30 the group within Google that try to work, uh,  in a prosocial way to counter the antisocial   damage that the algorithm of say YouTube has  done to the society. Uh, far as I understand,   the Community Forum, community Notes  team within Meta is doing a similar job. Um, and so it's not all. Black and  white, so to speak. Uh, everyone, uh,   who look at these big tech CM monolith,  but what we're doing is that we're also
• 66:30 - 67:00 building a network between the people who  kind of act like conscience within those   big tech so that we can band together  and build a horizontal social network. So I've heard you, uh, in a conversation with  our mutual friend, uh, Tristan Harris mm-hmm. Who   introduced us. I've heard you use the phrase, the  most careful should win the prize. Mm-hmm. Yes. In   reference to how our current systems incentivize  people and companies with dopamine and dollars,
• 67:00 - 67:30 et cetera. Can you unpack by what you  mean by that statement and how is your   work, uh, creating those, those  mechanisms to incentivize care? Yeah, definitely. Uh, I would say it's not  just incentivizing care, it is also assisting   and augmenting care because it is like very, um,  energy and time consuming, uh, to do care work.   And, uh, a facilitator like realistically cannot  facilitate 450 people at once, even if they really
• 67:30 - 68:00 care a lot. There's some wet wear limitations,  uh, to the amount of care you can put. As a facilitator to a conversation. And so think  of, uh, like for like personal care. Sometime if   you want to move, uh, people who are heavy and so  on, you can use a exoskeleton, uh, that does not   automate away your work, but allow you to lift,  uh, better weights. Um, you, you can also think   of cortico and similar conversation network  plurality, technologies like Exo, uh, cortex,
• 68:00 - 68:30 uh, that helps, uh, somebody who perform care  work like facilitation to make sense of more   people or to close the loop slightly faster,  but it's not replacing, uh, the care workers. Um. To replace them would be like, you know,  sending my avatar to talk to your avatar and   have AI summarize all the avatars and  have avatars be the mayor. It's like,   you know, going to the gym and seeing the robot  lifting the weights, I'm sure very impressive,
• 68:30 - 69:00 but it does not help our civic muscles. So at this  care work, uh, pairs with the idea of assistive   intelligence in that it cease the people to  people, promises people to people attention   as the most important, the most cherished,  and then technology is just to foster it. So this is very eyeopening and, and exciting   and, um, we've approached, um, what I call a  species level conversation. Mm-hmm. And almost
• 69:00 - 69:30 a rite of passage for our species at large.  And there's lots of countries in the world.   Do you ever think that there's something unique  about Taiwan and the population of Taiwan, uh,   and the culture that made it a more viable place  for these strategies and movements to take hold? Uh, is it, or is it, is it  uh, applicable anywhere? I think it's applicable anywhere. Uh, I  think Taiwan simply has to innovate along
• 69:30 - 70:00 these domain because all our people,  at least people above 40 years old,   including myself, remember the martial law  and, uh, we've suffered, uh, the longest,   uh, martial period, multiple  decades, uh, in the world. And so we know how it is like to have our  freedom of expression of assembly and moving   and so on taken away. And so nobody want to  go back there. And so when we face. Such,
• 70:00 - 70:30 um, as you put it, uh, civilization skill,  um, threats, existential threats. We have   no choice but to double down on freedom  because we cannot even suffer a little bit   of democracy and freedom backsliding the  people simply would not put up with it. And so, uh, whichever solution we come up with  needs to be with the people, not just for the   people. People do not accept this authoritarian  for the people rhetoric in Taiwan. But that's just
• 70:30 - 71:00 for, uh, the necessity to come up with these  ideas, to apply these ideas. You do not need   the same configuration as Taiwan, and you do not  need the same, um, existential opportunity, uh,   of like facing every day as potentially the,  you know, last day of democracy and so on. As we did since 1996 when we first voted  for our presidents and our not so friendly   neighbors started, uh, missile trails. And so,  yes, so while it originates in Taiwan, it can
• 71:00 - 71:30 work everywhere. It's not just, you know, Finland  or Tokyo, California or Bowling Green or Oregon   and so on. But it can also just be in your family,  in your school, um, and in your local community. So before becoming, uh, the Minister  of Digital Affairs in Taiwan,   you were a very engaged youth activist.  Mm-hmm. Uh, and as I understand it,   you were also a reverse mentor mm-hmm. In  the Taiwanese parliament. Yes. Which is a   role for people under 35 to advise mm-hmm. Older  officials. Yes. So, in your opinion, what is the
• 71:30 - 72:00 role of young people today in governance  and in particip participatory democracy? Mm-hmm. And what lessons do you take  away from being now? Uh, both sides   of the reverse mentor mentorship, uh, in Taiwan. I believe in intergenerational solidarity  where the young people sets the direction   and the senior people provide the support  and resources on the Taiwanese participation
• 72:00 - 72:30 platform. The most active age groups are  the 17 years olds and the 70 years olds. Um, both have more time on their hands, I suppose,  uh, but also both care more about the oxy toin   serotonin thing of sustainability rather than the  dopamine thing of the next quarter. Right? So.   The idea is not to arbitrarily put  them kind of against each other,
• 72:30 - 73:00 but rather to find the common topics where  the younger people see a new possibility. But the more senior people have the wisdom  to see how that can be made possible,   like the adjacent possible, how adjacent  really is that possibility. And so through   reverse mentoring and through this kind  of intergenerational solidarity design,   we incentivize the local social  entrepreneurs and so on, uh, to form   the kind of leadership team that has different  generations, uh, in their board basically.
• 73:00 - 73:30 Uh, and so this I think is a great  way to heal. One of the most, um,   you know, divisive thing currently in our society,   which is the senior people with the resources  think that the society should go this way. And   then the young people already with proof,  the society cannot sustain this way. Do you have any specific  recommendations, Audrey, on how mm-hmm. The listeners and viewers of this program can  create a better relationship with technology
• 73:30 - 74:00 as an average citizen, uh, who wants to be  informed and engaged with their governments, uh,   and institutions. What, what advice do you have  on for the viewers to, to better use technology? Uh, on a personal level  color filter is really great. Uh, I've also seen people  using, uh, like a stylus, uh,   or a keyboard or really anything  that is not a touch screen,   and that also works great. So one of the two  can probably switch you off dopamine. So it's
• 74:00 - 74:30 creating a, it's, it's  creating a dopamine speed bump. Of sorts. Exactly, yes. So making sure that the  slot machine doesn't immediately respond to you,   uh, to increase the latency, uh, and  reduce the bandwidth, so to speak. Uh, so yes, uh, it works, um, very reliably  for me and hopefully for you, uh, as well,   uh, on the community, uh, level. Uh, one can in.  Encourage each other to try like more in-person
• 74:30 - 75:00 gatherings or synchronous online gatherings and  learn about active listening and facilitation. So   the facilitation school that I use, uh, is dynamic  facilitation and focus conversation method. But you don't need to, uh, go into any particular  school, even in a meeting if you say, okay,   now let's speak clockwise and now  let's speak counterclockwise. That   can already break this defer to the  most senior highest status person. Uh,   so that's the easiest facilitation method,  uh, that can be transmitted on a live show.
• 75:00 - 75:30 Uh, but there's a lot of facilitation  methods and so learn about it and also   get into the community of, uh, open space  technology and other ways to scale this, uh,   conversations and facilitation upward so that you  can scale not just horizontally, but also deeply. So you said there's a lots of different methods. Um, where would someone go  to learn about those methods? Yeah, you can, uh, search, uh,  for facilitation techniques, uh,
• 75:30 - 76:00 or group facilitation, and you will see pretty  much everything there is. Uh, or you can also,   uh, reach out to your local facilitation groups  and enter some facilitated conversations selves. So, uh, this, this has been just  an amazing discussion because I,   I realized the importance of this  topic, and I'm not even a novice in it. So I've learned, uh, quite a bit. Um, if  you could take your, um, open society, uh,
• 76:00 - 76:30 software, um, plurality hat off, and  just as a citizen of the world today,   facing the poly crisis, um, and what  I refer to as the human predicament,   what sort of advice do you have for,  for people being alive at this time?   Being aware of the issues that we face and, and  the challenges just as a, as a human to human. Yeah, I think, um, a shared sense of  urgency, whether it's ecological or social,
• 76:30 - 77:00 and. Whichever in between, uh, I think  that helps people to build solidarity,   to build this kind of care. Uh, that  makes it far easier for us to say,   yeah, this is too much for just a single  person. I need your help, and vice versa. And then if we can keep asking each  other, okay, so what's your feeling,   um, right now, uh, around these issues? And  if we can help each other by facilitating
• 77:00 - 77:30 conversations and uncovering uncommon  ground so that like active listening,   you can, uh, entertain listening to  people who are very much unlike you. Maybe coming from very different  background, very different ideology,   but if you can just listen for five minutes  without interrupting them. Even in your head,   uh, and then repeat back what you have,  um, heard with clarifying questions, uh,   also with curiosity and the other person take  turns and so on. Such simple practices of
• 77:30 - 78:00 literally facilitation with just two people  can really get us out of this domine loop. And the topics to explore together  again, is this shared urgency,   this crisis feeling that I'm sure that all of  us have, um, at least some time during the day. With the possible exception of  maybe Daniel Schmucker. I don't   know if I've ever listened to someone for  five minutes without interrupting them. Um, so I, I, I think it's good advice. What about  young people? I, I know you care deeply about
• 78:00 - 78:30 young humans, uh, because you were quite active  mm-hmm. Uh, in your younger years. What, what   specific recommendations do you have for young  humans, uh, in my country, in your country, around   the world listening to this, who become aware of  our economic, uh, social ecological, uh, problems? Yeah. Um, so certainly get organized.  Uh, and the young people of today knows
• 78:30 - 79:00 a lot about horizontal organization  of discovering a shared purpose and   how those shared purpose can bring people  together. And so if you are organized. Then   just as the Taiwanese 15 year olds,  you feel you are already a adult. You feel that you can already contribute  meaningfully to the agenda setting of the society.   The Taiwanese people, even before they turned  18, started some of the most, uh, impactful,
• 79:00 - 79:30 uh, petitions. Uh, not just changing, you know,  the, uh, recycling or plastic straw policy or   things like that on the ecological sense, but  also changed, uh, like their school schedule. So they go to school one hour later, uh, because  they prove that one more hour sleep is better   for grace than one more hour of the study. And  the Ministry of Education just accepted that,   uh, or, uh, even funding, uh, one of  the kind, uh, menstruation museum,
• 79:30 - 80:00 uh, in Taiwan and just slashed that  taboo from all the society in just. Two or three years and so on and so forth.  So any of these contributions, um, made,   uh, cabinet level advisor, reverse mentor,  uh, status, uh, but even without a status,   just organizing yourselves enable you to  have this kind of, uh, conversations that   are societal scale. And again, organization  starts by listening, uh, towards shared purpose.
• 80:00 - 80:30 And I recommend, um, people power, uh, from  Marshall Guns, uh, on how to get organized. So I have a couple, uh, closing questions  that I ask, uh, all my guests. I hope you   don't mind it. I know it's approaching, uh,  midnight mm-hmm. Uh, in where you are. Um,   what do you care most about in the world, Audrey?  I care the most about our ability to care. Thank you. Um. If you could wave a magic wand,   what is one thing you would do to  improve human and planetary futures? I
• 80:30 - 81:00 would make sure that, uh, anytime people speak  of utilitarian, uh, logic, uh, they automatically   have some care or virtue or, uh, spiritual, really  whichever edition, uh, Intuit. So, uh, a little   bit of infusion or inception, uh, of a different  ethics into the current utilitarian logic. And that, uh, as we have been  observing is what we've been doing,
• 81:00 - 81:30 uh, for the past hour and a half. Jao? Yes. Um. So what are you working  on now and what are you most, uh,   enthusiastic about? Mm-hmm.  That, that you can share? Yeah, so, um, I'm going to South by  Southwest, uh, in a couple days from   now. And, uh, my short biopic, uh, good enough  ancestor, uh, will be premiered, uh, online. Good enough, ancestor. I love that.
• 81:30 - 82:00 Yes. Uh, and, um, so potentially also working  on the film links, uh, adaptation. Uh, but yeah,   I encourage you to check out good enough  ancestor, uh, go how, as we say in Mandarin,   because if we were perfect, we actually robbed  the future, uh, from the creativity and the   canvas. But if we're just good enough, then  we can make peace with future generations. I love it. I love it. If you were to come  back on this show sometime in the future,   6, 9, 12 months from now, what is one topic,  um, that is relevant to our future that you
• 82:00 - 82:30 are personally passionate about that you  would like to take a deep dive on? So we talked about, uh, this idea of a vertical  takeoff singularity when it comes to ai,   and we also talk about this horizontal care  based diffusion of capabilities of plurality. So a deep dive of how these two directions  work with each other, against each other. Uh,
• 82:30 - 83:00 the dynamic between those two approaches,  I think we can do a deep dive on it. Awesome. Um, this has been great. Audrey, do you  have any closing words, uh, for our viewers today? Yeah, definitely. So I often quote, uh, from  my favorite, uh, singer songwriter Lena Cohen,   on the importance of being just  good enough but not perfect. Because if you're perfect, there's no way  to say I need help, and no way for others   to express care. So to quote Lena Cohen, um,  my favorite stanza from Anthem goes like this,
• 83:00 - 83:30 ring the bells that still can ring. Forget  your perfect offering. There's a crack,   a crack in everything, and  that's how the light gets in. Thank you for your time today and for your  very important work and, uh, to be continued,   my friend. Thank you. Take care. Take good care.  If you enjoyed or learned from this episode of the
• 83:30 - 84:00 Great Simplification, please follow us on your  favorite podcast platform. You can also visit   the great simplification.com for references  and show notes from today's conversation. And to connect with fellow listeners  of this podcast, check out our Discord   channel. This show is hosted by me, Nate  Hagens, edited by No Troublemakers Media,   and produced by Misty Stinnett. Leslie  Balu, Brady Hayan, and Lizzie Sir.