2021-03-09 GovAI Webinar: The Frontier of Democracy

Welcome. I’m Allan Dafoe, the director of the Center for the Governance of AI, which is organizing this talk series. We are based at the Future of Humanity Institute at the University of Oxford. We research the opportunities and challenges brought by advances in AI and related technologies, so as to advise policy to maximize the benefits and minimize the risks from advanced AI. Governance, this key term in our name, refers both descriptively to the ways that the decisions are made about the development and deployment of AI, but also the normative aspiration that those decisions emerge from institutions that are effective, equitable, and legitimate. If you want to learn more about our work, you can go to governance.ai.

I’m delighted today to introduce our conversation featuring Audrey Tang, Hélène Landemore, and Ben Garfinkel. Hélène Landemore is my former colleague from Yale University, where she is an associate professor of political science. She is halfway through a four-volume examination of democracy, in which she provides a justification for democracy, clarifies its meaning, and suggests ways to innovate on it. Hélène has been thinking about the meaning of democracy for as long as I’ve known her, and she is that especially valuable kind of political theorist who is trying to solve pressing real-world problems, and in so doing, exposes her thinking to the mess of empirical reality.

Ben Garfinkel is a research fellow at the Center for the Governance of AI and a DPhil student at Oxford’s Department of Politics and International Relations. Ben’s intellectual contributions span many topic areas, including the implications for democracy of cryptography and of AI. I’m always eager to read Ben’s analyses, as he has a gift for distilling out the key issues from large, messy topics.

Audrey Tang is Taiwan’s digital minister, Taiwan’s youngest minister in the country’s history, and also the world’s first openly transgender minister. Audrey’s contributions are hard to summarize. Audrey, at various points, has been a startup entrepreneur, open source software hacker, political activist, and poet. Audrey is one of those rare people who is innovating in governance in a major way in the real world at the cutting edge of contemporary culture and technology. Audrey is doing this through big things, such as through collaborative civic technologies in Taiwan, combating disinformation campaigns, and globally exemplary policies for addressing COVID. But also through simple innovations, such as being attentive to the virtues of sometimes removing features from apps and devices to better steer our use of them.

Audrey reminds us that some of the greatest positive impacts from digital technologies come not from what they do, but from what they enable us to do. To quote in part from Wired, “While the world is torn between twin dystopias of post-trust information chaos amongst some democracies and the other dystopia of authoritarian, technologically mediated surveillance and censorship regime, Audrey is making and demonstrating the radical argument that digital tools can be used to build stronger, more open, more accountable democracies.” For those of us who believe that improved governance is as much about empirical learning as it is about theory, real-world policy experiments, such as Audrey’s, and those of the people of Taiwan, are the critical input to governance innovation.

Today, we will aim for a conversational format, in part where Hélène and Ben will offer comments and pose questions. I also want to encourage all audience members to type your questions in the feature below. We can’t promise that your questions will be answered, but we will see them and try to integrate them into the conversation. So with that, Audrey, we look forward to learning from you. The floor is yours.

Good local time, everyone. Really happy to be here and sharing what we have learned in the past few years in Taiwan about digital democracy. And I will begin with a quite short, fifteen, twenty minutes at most, presentation that walks through first how we counter the pandemic with no lockdown and counter the infodemic with no takedown, and then all the while increasing the power of the social sector through what we call the people, public, private partnership. But I will go through the material very quickly and only revisit it after our Q&A session as part of the conversational format as our moderator have just introduced.

So to me, participatory democracy, deliberative democracy, they are great, but they are too many syllables. So as a poet, I try to use monosyllabic introductions at beginning. So I usually just say it’s democracy that’s fast, that’s fair, that’s fun. So fast, fair, fun are the three principles of conversation in a society that enable us to counter the COVID, for example, last year.

For example, we have a social sector run Reddit equivalent that has no advertisers or shareholders. It’s called PTT. It’s an open source project co-governed by National Taiwan University students and so on, who triaged the Dr. Li Wenliang’s whistleblowing in 2019, actually, COVID-19. Dr. Li Wenliang’s message reached a lot of people in Taiwan after being triaged and within twenty-four hours enabled us to begin health inspections for all flight passengers coming in from Wuhan, because in Taiwan, we enjoy the, according to Civicus Monitor, only open, fully open society in all of Asia, meaning that is, freedom of speech, of assembly, and so on, are total, that a journalist’s word holds the same currency as a minister’s word, actually, rather higher. And because of that, people were able to talk about SARS openly and triage such emerging collective intelligence.

And we add to that a real-time response system, the Central Epidemic Command Center. So it’s actually not cutting-edge digital technology. This is simple technology called call center. So anyone throughout 2020 can pick up their phone and call 1922 and ask to their heart’s content anything about the epidemic. And they can also suggest, like, concrete innovations.

For example, last April, there was a young boy who called saying, “I don’t want to go to school because you’re rationing our mask. All I get is these pink medical masks. And I don’t want to wear it to school for I am a boy, all the boys in my class have navy blue,” and things like that. And because of that, on the very next day, daily 2:00 PM press conference, all the medical officers wore pink, regardless of their gender. And so the boy become the most hip boy in the class, for only he has the color that the heroes wear. And Minister Chen even said Pink Panther was his childhood idol. So also heroes wear.

And so through these two very simple things, like daily press conference that goes live stream and a hotline that receives suggestion from each and everyone, more than two million calls throughout 2020, we enable this very quick response system that ensures the collective intelligence gets into the decision cycle and rolled out every Thursday because we use a agile methodology to all the, like, PPE distribution and vaccination and quarantine policies, and you name it.

So this is another example. We have a civic technologist, name is Howard Wu from Tainan, who early February last year wrote this thing out without asking anyone’s permission, which is a crowdsourced map that displays availability of medical masks in pharmacies. And I look at this app and talk to the head of cabinet, our premier, saying, “We need to trust citizen with open data.” But this is not open data that’s pre-approved by his public servants. This is real-time open data or open API that’s published as soon as anyone swipes the national health identity cards, the IC cards, in exchange for the medical-grade mask in any pharmacies. The real-time count decreased by two at the time. Nowadays, it’s by ten. And everyone queuing in line can check for themselves that this system is performing as expected.

This also enabled evidence-based interpolations, like in our parliament by a previous VP of data analytics at Foxconn, MP Kao Hung-an. And she interpolated saying, “Your rural and urban distribution looks fair on the map, but it’s actually unfair if you take into the time opportunity cost.” And that prompted us to co-create the next distribution mechanism using pre-ordering in convenience stores and so on within the next twenty-four hours. So the idea is that it’s everyone’s business with everyone’s help. Anyone who makes a evidence-based suggestion, then the minister will simply say, “Okay, legislator teach us,” and then we roll it out the next Thursday. So again, a very agile mindset that ensures fairness of all kinds.

So this is the head of the cabinet, our premier. Now, of course, during COVID, people are anxious. There’s a lot of conspiracy theories, disinformation. For example, there was a popular rumor that said, and I quote, “The state is confiscating all the tissue paper material to make medical-grade masks,” end quote. But we detect such disinformation within a couple hours, and always within two hours, we roll out two different pictures, each 200 characters or less, that dispel these rumors by turning the outrage into humor. This is called humor over rumor.

So this communication style, you see the backside of our premier now, and says in very large fonts, “Each of us only have one pair of bottoms,” because in Mandarin, bottoms twin sounds the same as stockpiling twin. (laughter) So this is reminding us that, of course, first it doesn’t make sense to stockpile, and then the South American materials that makes the tissue papers are actually very different from the medical mask material, which are domestic. Because it went absolutely viral, people who see this and laugh about it actually becomes immune from the disinformation. So people gets vaccinated against the virus of the mind.

And this is doubly useful if we have a cute dog, the spokesdog is Shiba Inu, talking about physical distancing. When you’re outdoor, keep two Shibas away. Indoor, three Shibas away. Wear a mask to protect your own face against your own unwashed hand, and so on. Again, in a way that makes this memes, this idea was spreading spread. And once people laughed about it, they get into this co-creative mood. And we’ve documented this part, the Taiwan model in taiwancanhelp.us, which is also a crowdsourced and crowdfunded website.

Now let’s dive in a little bit more. How do we get the sixty minutes on average time, timely response from each and every ministry? And why would a cabinet’s premier makes fun of his own head? Because there was a popular rumor that says, “Perming your hair multiple times a week will start to be subject to one million dollar fine.” Of course, that’s not true. And he said, “I may be bald now, but I would not punish people with hair. It’s just a labeling requirement.” And the premier, as he looks now, says, “If you perm your hair multiple times a week, you will not, you know, damage your bank account, but you will damage your hair. And just look at me for what will happen.”

So what we actually do is like creation of vaccine. We take this disinformation, this popular rumor, and then we bracket it in a notice, in public notice way, and make sure that there’s a very funny memetic payload. And this is empowered by civic tech. The same gov-zero, g0v community that supported the mask rationing map, in this case also supports the collaborative fact-checking so that people in even end-to-end encrypted channel like WhatsApp, it’s called Line here, can long press a message and flag it as spam, as potential disinformation.

And because at any given point there’s maybe only total societal bandwidth for like three popular disinformation that has a R value of more than one, that is to say, going viral. So we focus our energy, the International Fact-Checking Network’s energy, on these, like, trending rumors, and then they fact-check it. This is an independent organization that also fact-checks the administration. But upshot is that once they did their journalistic work, now with the contribution from like young primary school or middle school people, we just wrote this meme out so that people understand that, hey, this is actually something that you can laugh at and then co-create, but then the outrage is spent off, and then people don’t get this into this conspiracy theory field thinking anymore.

And this very strong social sector mandate also enabled us to negotiate with the more antisocial corner of social media, namely Facebook, and many other platforms saying, “Hey, the social sector already pressured the public sector into disclosing, for example, the political expenditure and donations and so on, and the political advertisement on Facebook need to be treated the same, publishing as open data, banning foreign sponsored ones for the investigative journalists to do their work, otherwise they may face social sanction.” And Facebook relented in 2019. And for our presidential election, we then have a market of ideas that’s remarkably free from either foreign interference or specialized advertisements.

And this is a really good example because Hong Kong situation was shaping up to be a deciding factor in our presidential election, and this is a actual disinformation campaign that’s being spread there. And so the fact-checkers unveiled that this alternate caption on the right is actually based on a real Reuters photo, but the alternate caption is probably sponsored by the (laughter) state organ, the central political and law unit of the Beijing regime. But we didn’t take anything down. We rather put a public notice so that people understand the framing anytime they share it, and therefore develop things like the antibodies of the mind.

And the same applies to the, like, voting. The voting is rigged rumors. Again, we counter this by inviting YouTubers to come to the counting process, and they have different apps in each different parties, but it counts the counting process in real time. So people choose to receive information from their affiliated party. Of course, there’s four major parties, but all of them are working with the same crowdsourced accountability mechanism that detects anything like invisible ink or whatever in real time. So again, the conspiracy theory have no room to grow. Their trolls also have no room to grow. And this I already mentioned, this disinformation about medical mask is countered very quickly by people going to the pharmacy and checking for themself in more than 100 different tools based on the same open API.

So this part, I would like to say that we treat the infodemic as the epidemic, making sure there’s broadband as a human right, the digital competence, not just literacy, media competence classes in middle school and primary school, so they can check all the presidential candidates’ forums and debates and so on. And then we innovate to flag the trending disinformation and also ban, like, precision advertisement fields, money fields, more antisocial corner of social platform conversations. And all this is to create the necessary condition for digital democracy.

And as Dr. Tsai Ing-wen, our president, said in her inauguration speech 2016, she said, “Before we think of democracy as a showdown between two opposing values, but now democracy must become a conversation between many diverse values.” And to me, this symbolizes this thinking that instead of letting 49% of people feeling they have lost every four years or two years, depending, by uploading three bits of information per voter per four years, which is very low bandwidth, we would indeed see democracy as a type of technology and increase the bandwidth of democracy.

So this is literally my office where I hold my Wednesday office hours, and it’s in the middle of Taipei, heart of Taipei City, called the Social Innovation Lab. And we invite all the social innovators to present their work like for self-driving vehicles, even before we have laws for self-driving vehicles. But there are tricycles, they’re open source, open data from MIT Media Lab, so people modify it to fit the local community’s needs, thereby co-domesticating with AI so that people can see AI as assistive intelligence instead of authoritarian intelligence, and therefore build effective partnerships.

So a lot of our work is to make sure there’s a institutionalized way for such co-creation to happen outside of the election cycle. For example, every year we have the Presidential Hackathon where we invite people to innovate digitally, like building assistive intelligence to save water because we are facing a water shortage this year. We had no typhoon last year thanks to climate change. And so they use AI system to detect water leaks very effectively. And while they only have the budget to try it out with the social sector in the Keelung region because they won the championship. We give out five of those awards in the Presidential Hackathon.

It’s a trophy that’s also a micro projector. If you turn it on, it projects Dr. Tsai Ing-wen handing you the trophy, saying that whatever you did on a local small scale in three months, the president promised that it’s as good as a executive promise and will make it into national level policy with all the budget, personnel, and law required within the next 12 months. But where is the democratic mandate? Well, it turns out each of the five winning teams all need to go through incubation period where we make the social, private, and public sector partnered data coalitions.

So we collaboratively make such, like, air boxes, this is a good example, where people in primary school measure PM 2.5 air qualities, write it into a distributed ledger, and then show that people even as young as seven years old can be good data stewards and contribute to environmental sensing. And because we are a very open democracy, we can’t beat the environmental activists, so we must join them. So the government dedicate the resource to make such climate sensing community and networks a reality. So it’s almost all with collective intelligence or assistive intelligence, and more often than not, it’s assistive collective intelligence that solve one or more of the global goals using this co-creation.

So I mentioned like choosing a small area to experiment, but how do we discover people’s needs? Well, we go on social innovation tours using the video conferencing so that I’m the only facilitator that travels. But the 12 ministries in the central government in the Social Innovation Lab look at the rural places and where I facilitate the conversation, so they respond to the people’s need in the here and now, and make sure that the ministries do not copy each other, but rather co-create with each other in a idea of sandbox where we try out such solutions for like six months, three months, and so on, and see whether people likes it or not.

But how do we listen at scale and find out whether people like it or not? Well, for self-driving vehicles and the 5G spectrum allocation for the local 5G sandboxes, as well as this is the original one from 2015 about UberX. Some people call it gig economy, some people call it sharing economy, some people call it platform economy. We don’t dwell on these abstract conversations. We ask what people feel about the fact that there’s some unprofessionally licensed private vehicles picking up strangers and charging them for it. Now, of course, there’s no right or wrong about feelings. You may feel happy, they may feel upset, it’s all okay. But with this pro-social digital public infrastructure, which is called Pol.is, it’s open source, we make sure that the ideas that reflects people’s common feeling, what we call rough consensus, translate into the agenda of a face-to-face real-time deliberation that’s also live streamed.

So the experience is like this. Someone says, “I feel passenger liability insurance is very important.” You may agree or disagree. If you agree, you move toward me. If you disagree, you move farther away from me. But there’s no reply button, so there’s no way for trolls to grow. And after three weeks of such conversation, we always see that ideological or divisive statements are there, but people don’t spend calories on it. And then the rough consensus as revealed by the Pol.is conversation, these are the statements that then we hold ourself accountable to deliberate with the stakeholders as the agenda that we understand that everybody can live with it, right?

So this is the consensus statement that drives the multipurpose taxi. So the law is now that Uber is a legal “Q Taxi” fleet, but it also enable platform co-ops and Line Taxi, many other taxi company to not undercut existing meters, but also work in a way that reflects new innovation like surge pricing and so on. So this is like KPI measurement progress, but it’s also crowdsourced. This is crowdsourced agenda setting. So the upshot is that with this way, people discover there’s far more in common when we look at things like a rough consensus than we originally imagined in a more antisocial corner of social media, so that we can find common values and deliver on innovation that fulfills those common values. So that’s my opening, and I look forward to the discussion.

Thank you so much, Audrey. Hélène, would you like to share your thoughts?

Can you hear me? Yes? Okay. So first, I’d like to thank my former Yale colleague, Allan Dafoe, for including me in this conversation and for the kind introduction. It’s a great honor to — and an immense pleasure — to be able to meet Audrey Tang, whose work I’ve been admiring for a long time from a distance, for its incredibly technologically visionary aspects, certainly, but even more so for the philosophical principles of openness, transparency, fairness, and fun behind it, as she presented them, which very much resonates with the philosophy of politics I have tried to develop in my own work, including in my recent book, Open Democracy.

So I first heard of Audrey when I was visiting Stanford in 2012, 13. I think the Silicon Valley was perhaps more, you know, aware of what was going on in Taiwan at the time, though I can’t exactly remember who first pointed out her work to me. Around that time, I was discovering and exploring myself the merits of practices like crowdsourcing. I was involved in the design and running an analysis of an experiment in Finland on a crowdsourced law reform of off-road traffic regulation involving the regulation of snowmobile in northern Finland.

And that really taught me something important about how radical principles like openness and transparency can generate a free flow of ideas, collaboration, and genuine creativity, in a way that much more traditional, top-down, rational, planned, centralized methods actually can’t. So from my perspective, there’s a magic to these kinds of principle that it’s actually hard to explain and understand unless you’ve seen them deployed and at work. I think only then you can fully appreciate that it’s something worth pursuing.

And I noticed how hard it is, at least in my home — in my own country of France, and I think in the US as well, how hard it is to sell these principles and the technologies that try to implement them to government representatives who only believe in familiar, rational, top-down, opaque structures where most of the control is in their hands. Most politicians, as it turns out, are deathly afraid of losing control. Audrey Tang, from that point of view, is a rare politician or poetician, as she probably is better labeled, whose ambition seems to want to give control away and empower other people. Again, I think you can only fully understand the radicalness and potential of that approach when you see it working firsthand in some of those experiments.

So for those of you who may not know fully who Audrey is, and despite the brilliant presentation she just gave, I thought I would summarize briefly what I take her work to be about, because it’s so wide-ranging that maybe her presentation didn’t quite cover it all. So I knew some of it myself, but I discovered a lot more while preparing for this encounter.

So the first aspect is definitely the technological aspect. As digital minister of Taiwan, Audrey is famous for using very innovative technological methods to bring legislative agenda questions from the periphery of power to the center and helping solve identified issues in consultative, efficient, fair and fun and legitimate ways. Many of you have — had probably already heard of vTaiwan. So vTaiwan is this process that combines a website where anyone can register and put forward an issue and try to gather votes to push it on the deliberation plate of Audrey and her team. But it’s also a series of meetings and hackathons for problem-solving. And it’s also, vTaiwan, the use of an AI-powered system called Pol.is, as she presented it just before, that allows them to identify the preferences, judgments, and in fact, underlying evolving consensus among large groups of people.

And the goal is both to allow more access for ordinary citizens, especially the youth, into the process of shaping the legislative agenda, and to tap the collective intelligence and creativity of the group to offer better solutions to collective problems. She mentioned the pandemic obviously, but one of their big successes prior to that was the management of Uber and Airbnb’s arrival in Taiwan’s economic ecology. And I love the tagline of that — of the vTaiwan website. It says, “Where do we go as a society? Let’s go and think together.” And to me, that’s really the spirit of my own vision for open democracy, which is, well, as a first take on the problems that the world throw at us, we have to ask ourselves collectively, where do we go as a society, and involve absolutely everyone in the conversation.

So beyond this technological aspect, she’s indeed pushing for new design principles which double as governance principles, and that’s where I think the radical potential is the most visible. So among those are radical transparency, what you could call anarchy or at least flat hierarchy, pluralism, decentralization, and fun. So in terms of transparency, as a minister, unlike most of her peers in advanced democracies, she really makes a habit of posting all her talks online, so there’s no secret conversations that only some people are privy to.

Anarchy or flat hierarchy, as far as you could tell from the pictures, her whole ministry is not really headed by her as much as inspired by her. This reminds me a little bit of the philosophy of the former leader of the Pirate Party in Iceland, Birgitta Jónsdóttir, who herself called herself a poetician, meaning a crossbreed of a poet and a politician, and refused to be seen as the head or representative of the party. So it’s really in these figures, Audrey or Birgitta, we see a different kind of leadership. It’s leadership by example, leadership by kindness.

As to pluralism, it’s this idea that there should not be a monopoly of authority in any area of life, that polyphony is essential in any decision process, which doesn’t mean conflict. And even as we are sort of aiming for a consensus or a consensual solution, we don’t want to silence any voice. So it’s important for her to always include an alternative to the mainstream narrative or the main algorithm or the main source of authority. And all of that seems to me, it seems to translate in a further principle, that of decentralization. There’s not one path to the center of power. In fact, there should be many paths and many centers that sort of self-manage and develop their own logics, which feeds in turn into the pluralism.

And finally, fun is a huge part of her design, both for instrumental and intrinsic reasons. She showed you cute dog pictures. She uses cute cat GIFs, because they make people happy, but also because they can help make viral a government’s tweet about, say, the need to wash your hands frequently. Or because fun is the best way to incentivize or rather spur creativity as well as fulfillment and a sense of community.

There’s a third prong to Audrey’s work, which I think is also essential, which is education. In a world where we know that online experiences are going to be more and more threatened by the presence of deepfakes, technologies that create addictions and manipulate people’s attention, et cetera, et cetera, she aims to inculcate and cultivate digital competence, as she calls it, not literacy, among children, from kindergarten basically, to equip them with the right cognitive capacities for the world they are going to live in. So it’s very much a revolutionary form of education that they’ve piloted in Taiwan, and they’ve tried different kinds of approaches. Many is basically promoting autonomy and self-reliance and the pursuit of one’s personal projects from a very early age on.

So she talks also in other presentations of a pluralism of AI mentors, if we’re going to now live our lives with the help of personalized AI mentors, in the sense that we need to have a way to forge and refine the values that we want, including by playing one set of advice against another, a little bit like we can play one parent against another.

So I now turn to some questions I have for Audrey. There are some questions around design that, you know, uses dichotomies of deliberation versus aggregation, random selection versus self-selection, anonymity versus publicity. I understand that vTaiwan and the other processes she uses blend those things and have different components at different stages. But I have to say still for me as a deliberative democrat, so someone who’s committed to the view that the quality of deliberation crucially matters to the legitimacy of laws and policies, I have tended to prefer in my own conceptualization the central structure of a large randomly selected assembly of citizens, such as, for example, the one that just concluded in France.

A body is called a Citizens’ Convention for Climate, which — and it was a body of 150 randomly selected citizens who spent around 12 months at this point carefully debating ways to curb French green gas emissions in a socially fair way. It was a long, costly process. Can’t say it was a sort of fast iteration, 48 hours hackathon that Audrey is familiar with. Much of it was face to face until basically the pandemic forced the group to move on Zoom. And overall, it was quite low tech, apart from indeed the use of Zoom and a platform where they could exchange some ideas. And it also only involved the larger population in an indirect way.

In the vTaiwan process, by contrast, Audrey relies on the participation of a much larger but essentially self-selected group whose demographic representativeness is questionable. Additionally, whatever deliberation happens is of a more decentralized, anarchical, and distributed nature, taking place over several months, between often anonymous participants who come and go on the platform. So there’s no guarantee that the viewpoints represented are those of the larger population. I think she showed us a slide where I think the number of participants was something like 2,000 people, so a far cry from the 27 million people in Taiwan.

And yet, I have to say, this seems to work beautifully to generate consensus, to work basically. The same way that the process that I observed in Finland worked, even though it was also based on self-selection and sort of little bits of deliberation here and there, but nothing like a centralized moment of actual exchange of arguments. So I have a question for Audrey, which is how do you account for that magic? And what would you say are the benefits of your model compared to a more centralized citizens’ assembly model? If you had used a citizens’ assembly model to deal with the Uber case, for example, or the pandemic, how do you think things would have turned out differently, if at all?

And finally, what’s the value of anonymity of participants versus transparency about their identity in online platforms like vTaiwan or Pol.is, especially regarding the resulting quality of their exchanges? Because in my experience in the Finnish experiment, we didn’t have too many issues with lack of civility because there was facilitation, but we know that anonymity can trigger behavior that are largely reprehensible and detrimental to the quality of other people’s engagement. So how do you decide yourself whether to require anonymity or publicity on any given platform?

I have now another set of questions about accountability. A common objection to citizens’ assemblies use is their assumed lack of accountability because members that are randomly selected are not under the threats or the sanction of elections. So I suppose you must hear the same type of objections to vTaiwan. Who are those people on the internet? What gives them the right to influence policy-making when we hardly know who they are? They don’t have a mandate. They can’t be sanctioned and made accountable by the threat of elections. Who says — who knows if they’re not being captured by large corporations, et cetera? What do you say to people who have these doubts about the process?

On a related note, one hour in a talk with historian Yuval Harari, which I just recently watched, you mentioned the idea of a distributed accountability, whereby every one of us is responsible for keeping track of mask stocks, for example. And all together, our collective efforts end up tracking reality quite efficiently. And I would like you to elaborate on this idea because it’s very hard to sell to, for example, French officials who only trust in electoral accountability and very long chains of commands with a final arbiter at the top, et cetera, et cetera.

So how do we guarantee that individual citizens will do their job instead of free riding on the work or that of others? Sorry — instead of free riding on the work of others. Is there something perhaps peculiar to Taiwanese culture that makes it so that people are responsible and invested? And how do you jumpstart such a culture where it’s lacking or assumed to be lacking, as in France, for example?

I have a question about plurality as well, which is related to this question of accountability as a design principle. In various talks, you mentioned the value of plurality as a design value to encode in institutions and algorithms. And I would like just to hear more about concrete example of how that works in practice, or even better, about how it sometimes doesn’t work or when it creates new kinds of problems. Because I have a hard time imagining it doesn’t sometimes.

The way I hear you describe it, I think of plurality as something somewhat similar to the idea of separation of powers or the checks and balances in the American Constitution. The idea is that if we create multiple poles with equal power and we allow them to check each other, we save ourselves from totalitarian risks or capture by one algorithm or viewpoint. But at the same time, in the American context, this pluralism or this divide to rule kind of approach has also been a recipe for institutional paralysis and status quo bias. So how does your design pluralism avoid such outcomes?

Now I turn to a question about the relationship of mini-publics to maxi-publics. So you say in one of your talks, and in the one you just gave, that if there’s a sufficient amount of cute cats and memes, the ideas worth spreading will spread. And I suppose that’s been true in your experience, and that’s absolutely wonderful, but again, what accounts for this experience on a causal theoretical level? What’s the theory that explain that ideas worth spreading will actually spread when we know that bad ideas and wrong ideas and falsehoods and fake news also spread?

So there’s an optimism running through your work, which is very reminiscent of John Stuart Mill’s faith that the truth will eventually emerge from the free exchange of views in a free market of ideas, or Habermas’ ideal of deliberation in the wild, being magically able to set the agenda for the deliberation of officials in the central decision track of the public sphere. Or I should say my own optimism in Habermas’ idea that the forceless force of the better argument will triumph. So obviously, I buy this vision myself, but I’d like to hear your explanation of that magic and how we can convince people that this can be expected. So in other words, in your experience, what else explains the failure of good ideas to spread besides the absence of cute cats? What’s the key ingredient for this spreading of good ideas to work?

I have another couple of questions, and then I promise I’m done. A question on the transition from here to there. So how do we get from where we, most of us are, not Taiwan obviously, meaning stuck in dated representative governments that are quite closed in their operational principles, to the there of open systems like Taiwan’s or open democracies, as I like to call them? You once said that instead of fighting an old system, you have to make a new system that will make the old system feel obsolete. You don’t have to convince everybody that the old system is obsolete. It’s enough that you convince just a few people.

I like that answer, but at the same time, if we look at the US, these few people have so much power that in such a massive conflict of interest in promoting a reform of the status quo that this is not going to happen. We are not gonna convince Zuckerberg or the GOP or economic elites to relinquish power or at least share it, because they just don’t have an incentive to. So how did you do it in your own country? How did you get past those hurdles and get to involve people that were not included in originally?

Similarly, there’s another more psychological obstacle, I guess, among people, sometimes of older generations, who fear technology or among groups that fear technology sometimes for very good reason, because technology can be racist, can be exclusionary. And so there’s a whole category of people who, for very different reasons, some of them very reasonable, refuse to engage in this participatory modes. So what do we say to them?

I have a question about the translation to developing societies. Okay. Excusez-moi. I apologize. My kids are back from school, and I told them not to rush in, but they did anyway. So a question about translation to developing societies. Many of your recommendations seem to apply to advanced societies with a tech-savvy population. So how do your ideals translate to developing countries with fewer resources and greater digital divides and illiteracies?

Finally, and this is the last question, I promise, I was thinking of the kind of advice you could perhaps give to a government like the French government or perhaps the organizers of this upcoming global citizens’ convention for climate, which will gather 1,000 people online in parallel to the COP26 in Glasgow, I believe, this year. How would you design a citizens’ assembly in a way that connects it to the larger public, the French public in one case, the entire, you know, humanity in the other, in a way that builds on what you’re familiar with basically, all these AI-assisted technologies? Can we have at this point AI facilitators for the deliberations among smaller groups on Zoom or in breakout rooms?

And more importantly, what should a governance platform for a self-ruling body that can’t rely on the classic structures and hierarchies of parties look like? ’Cause a big puzzle for deliberative democrats is how do we get this citizens’ assembly to self-rule as opposed to be managed from the outside by either professional companies that organize these things or a body of outsiders appointed by government in the case, for example, of the French convention for climate. So if you have any idea about how to facilitate the autonomy and independence and basically sovereignty of these randomly selected assemblies, I’d be really, really curious. So I’ll stop here. Thank you so much.

So, do I just answer them in reverse chronological order because I’m a reverse chron? (laughter) Or do we first move to Ben?

I think you should probably start discussing things.

Right. So, and I really am a reverse chron, so I will (laughter) do it the reverse order. So I think, theoretically, when I say plurality and when I say social sector, it means that literally any civil society organization can run such deliberation spaces. Just like non-violent communication, open space technology, facilitated dynamic facilitation, and things like that. These are not monopolies, but these are technologies. These are social technology just as social science is science.

We make sure that people who are civil society organizations, for example, during our occupy of the parliament in the Sunflower Movement in 2014, there’s more than 20 NGOs that occupy the parliament together, and each one occupying a corner near the parliament. And all we did is making sure that it’s a safe space, safe not only from police, because people counter surrounded the police, but also safe from disinformation and from rumors. But from that point onward, each CSOs take their own open space technology conversation, their own facilitated dynamic facilitation without any centrality of it.

And that in a large amount is thanks to the wide availability of broadband, so there’s no marginal cost in live streaming people’s speech all the time. And there’s also a reliance on free software, like Discourse and later on Pol.is, and so on, which could be easily self-hosted and then just forked. If a group of people doesn’t like the way it’s governed, they don’t have to convince some central authority in order to change the way that a deliberation is done. They can just take the deliberation, because it’s all open data anyway, into a clone or a fork of a conversation and bring it into another direction.

And so in that sense, people are in a non-rival situation when it comes to those mini republics. If you don’t like a particular way that a deliberation is done, you can always fork into a different mini republic until it feels more cohesive and more sovereign. And chaotic and swarm-like that it may feel like from my description, it actually works. And with half a million people on the street, many more online, I think we’ve proven quite conclusively to all the major parties in Taiwan that, well, this way actually works. We can scale deliberation at scale, and we did deliver the cross-strait service and trade agreement deliberation into five very concrete demands, not one less, and all of them gets accepted by the head of the parliament.

And so I think my main suggestion is just to make it very simple to start such kits of conversation in a very small mini republic, maybe as small as like 20 people or even five people, and then just scale out and deeply instead of scaling up, on the first principles.

And for developing societies, as I mentioned, the toll-free number, the TV or radio daily live streaming, these are appropriate technology. I wouldn’t call them low tech, by the way. (laughter) They’re appropriate technologies for the people there. And the memes, well, they are digital, but we also have young people just drawing them or printing them on posters or graffiti on walls, and things like so. So the nature of memes is that it’s adaptive to the format. So I’m never prescribing any particular broadband requiring, like, video conferencing as we’re doing now. We only do that when the people are comfortable with that. So we bring technology to the people. We never ask people to come to technology or to fit the way the technology works.

And that address the psychological fear, uncertainty, and doubt of technology because people feel just like the very cute self-driving tricycles that I showed middle of my presentation. People feel that this is something like a shopping cart (laughter) that they can also just modify. It never runs very fast. It never run someone over. So instead of the, you know, platooning trucks or things that looks very scary, people feel that self-driving vehicles is something that they can co-domesticate, and this feeling of participatory design also very important.

And this also ties into the Buckminster Fuller quote, I didn’t invent the quote (laughter), about building a new system that makes the old one obsolete, because you’re absolutely correct that there were, just as in Taiwan when we occupy the parliament, there were powers that be that really didn’t like this new form of deliberation. But we didn’t quite convince them, though. We worked with the mayoral candidates, just like with 15-M and other large scale movements. And basically all the mayoral candidate that didn’t supply their support to open government principles lost the election at the end of 2014.

And so there’s always a outside game, which is like in mayoral elections, in referenda or whatever. If people don’t play by the open democracy rules, then, because this is a social sector norm that people already felt, well, if you have participated in a real occupy, you know what I’m talking about. They get transformed from within so that they will actively bring down any candidate that didn’t continue pushing forward the open democracy idea.

So just today, this morning, in the parliament and all the major four parties just threw their weight behind the open parliament national action plan. So each party competes on being more open to others on open democracy because after 2014, they understand if they advocate a platform that is stuck in the old slow bit rate, like dial-up speed of democracy, well, they will get no votes whatsoever. So a outside game is always very important. It’s not about convincing. This is about just bringing down people who didn’t support that.

And the mini public to maxi public. I think the theory is a simple theory of really epidemiology, right? The idea was spread if they have a high basic transmission value. It means that people can willingly share it out of altruism, but also out of showing status, out of creativity and things like that. But if we hold the deliberative quality bar to a very fine consensus, it’s almost impossible. It’s definitely impossible on the internet. If on the internet you want a very fine consensus, by definition, people with too much time on their hand win the arguments and everybody else gets burnt out. Even in face-to-face, it’s a very exhausting process.

But if you’re only aiming for rough consensus, meaning are there something that we can all live with, then you actually get to that point very quickly, either online or offline. And based on that, it creates this how might we questions that analyze this wicked problem, problem requiring coordinated action in a way that people very easily remember.

Like, when we’re doing marriage equality deliberation, we eventually coalesced saying that same-sex couples, when they wed, they wed as individuals with all the same rights and duties, but their families don’t wed, so the old style of familial kinship relationship is not disrupted. We say, “We marry the in — the by-laws, but not the in-laws.” Now, this is very easy to remember, and it’s a rough consensus. It’s definitely not something people sign, right?

But with this meme, each community can bring it back to their maxi public and remix this meme into a message that they understand, just as our YouTube counters, right? They, actually, all the people who vote suspect the other parties of, you know, meddling with the election, but they do listen to their party’s favorite YouTubers. But when all the YouTubers share the same rough consensus about counting elections, then it’s problem solved because people just listen to the YouTuber that is more politically aligned with them. But they already have the same rough consensus, as I mentioned, about the mini publics.

Now, the plural algorithms very rarely work, and not on the first try. When we roll out the mask rationing map, some pharmacies invented this method that people trade in their national health cards to some numbers, take a number system in the morning, and they process and swipe those national health cards during lunch break. And at evening, people will come back with numbers in exchange to a mask and a IC card. Now this, of course, saves queuing time, while a mask rationing map also saves queuing time because you’re not supposed to go to the pharmacy that have run out of mask.

But together, these two social innovations are like Mentos to Coca-Cola. They explode because people who use take a number system in the pharmacies, they don’t reflect on the map anymore, right? On the rationing map, they will look as if they sell nothing until lunch break, where they sell everything (laughter), which really looks like it’s rigged. And so the pharmacists get a lot of angry calls. And one of the nearby pharmacist even said a very large banner A4 paper that said, “Don’t trust the app! Exclamation mark.”

And so (laughter) I took a deep breath and walked in and asked the pharmacy this very important question, “If you are the digital minister, what would you do?” And then they just brainstorm in their own pharmacist group. And they got back to me saying, “What if you just invent a button, pressing a button let us disappear from the map?” And I’m like, “This is a really good idea, but we may take a week or two to change the code.” And they’re like, “You don’t actually need to change the code because we hacked the system.” Just like white hat hacking.

They invented it such a way that on the morning they would usually say, “I have received five hundred masks for rationing.” They discover if they input minus one thousand, then it gets into negative stock and it disappear from the map, because we can’t handle negative numbers. And that hack actually solved the problem. Soon as they run out of the numbers, they just go minus one thousand.

Of course, this is forking the digital service. This is essentially taking it into a very different direction. And now the pressure is on us. So we immediately institutionalize that. So after two, three weeks of iterations, we get into a point where the two social innovations finally work together. This is thanking to the weekly deployment cycle and also this key question, if you are the digital minister, what would you do?

And this also solves the free-riding problem because it’s at this core non-rival. You can have more than one hundred different visualization or co-creation systems. It doesn’t hurt to have another one. So that’s by definition takes it outside of the tragedy of a common case because people can fork, but they don’t have to start from scratch, those distribution methods. And indeed, the six thousand pharmacies, each one, I guess, serve as a sandbox of how to manage this co-creation together, and the best one, of course, just amplifies.

Now, the lack of accountability and sanction of election, I think it’s a core issue. Indeed, we’re piggybacking on existing representative democracy, as I said, the threat of, you know, just not voting in mayors that didn’t support open government. The same for presidents, legislators, and so on. And so I think for the foreseeable future, we’ll still coexist for a while, and the main selling point to that, I understand you already buy my, our ideas (laughter). But the main selling point is that for the representatives, it improved the signal-to-noise ratio. They no longer need to do all the discovery of the, you know, principal agent problem by themselves, solving that wicked problem by themself.

We do have people in the career public service who are very capable people, what we call the participation officers, and there’s one in each and every ministry in Taiwan, all thirty-two of them. The Ministry of Health one actually live with that cute dog. And so that explains the quick memetic generation because they just walk back home and take new photos.

And so for each new case, like how do we open up the mountains to mountaineering and hiking, then of course it’s by first self-selected. It’s the POs talking to people related to hiking. And we do mostly the discovery part, which is by Pol.is. But once we have discovered the common interests, the common values and so on, we enter this collaborative meeting, which is always face-to-face and almost always live streamed. And then we move to a more representative part, where there’s a traditional oversee cycle for budget implementations and things like that.

And because of this double diamond, we don’t confuse the discover and define, which is great for participatory democracy, and this develop and deliver, which is best left for career public service and the system integrators and the legislators. And the representative system can take care of the next diamond. Understanding the first diamond will cost them essentially nothing and save their time getting quality input from the stakeholder that’s involved. So we use the traditional deliberative methods and brainstorming and so on, moving from the challenge to the definition, but the implementation is still using classic methods for the time being.

And so I think this answers also your question about sortition-based civic assemblies. If people are motivated enough, we can substitute this self-selection with civic assemblies. Of course, we will take more time, I guess, maybe one or two week more to bring them up to speed because after all, they’re not self-selecting. They may know they have no hiking experience, for example. But I’m aware that we also have that for the Ministry of Culture, of Economy, and actually the National Palace Museum and the healthcare system, universal healthcare, they were based on civic assembly model.

So we basically augment that with a self-selected network input, but we don’t actually replace that. We augment this existing system of collaborative meetings, civic assemblies, but with the agenda set by crowdsourcing because in the agenda-setting phase, again, it’s non-rival. So I’ll, like, these are very simple, like one line — one-liner answers, and I look forward to explain that more.

Thank you.

Yes. Thanks, Audrey. That was fantastic. I suspect as we continue, Hélène, if you wanna also pull on some of these threads further, we can do that. But now we’ll turn it over to Ben for your thoughts and questions.

Great. Yeah. Thank you so much. Let me see if I can successfully share my screen. Great. So I will keep my comments pretty short, just to leave more room for discussion. But a thread I thought it’d be interesting to pull on a little bit is that it seems like, you know, sort of a unifying theme of a lot of the digital democracy projects happening in Taiwan, like vTaiwan, is that they’re aiming to look for opportunities to move beyond, you know, pure representative democracy. So no longer, as Audrey put it, just, you know, sending a bit of information every four years, but really actively engaging the public, and giving them a more active role in shaping legislation and shaping policy between election years, not just voting.

There’s also at the same time a really long intellectual tradition, which actually, I should say Hélène’s recent book does a really good job of summarizing, in some ways refuting. But there’s a long intellectual tradition defending limits to public participation or defending the idea of representative democracy against forms of democracy that involve more active participation. So I thought it’d be interesting to explore a little bit how some direct democracy initiatives like vTaiwan address or don’t address these sorts of classic concerns that political scientists and political theorists have, you know, brought up from time to time.

Before diving into that, I think it’s also just worth making the sort of, you know, very basic case for why we should have a strong prior that public participation is good. And really simplest argument for involving the public more in policymaking and legislation is that members of the public typically know a lot more about their own preferences and concerns than their representatives do. So we should have a strong inclination to think that if you involve the public more, and their preferences and concerns feed directly into decision-making processes, then we should often expect the outcomes to be policies that better reflect the public’s preferences relative to, you know, just sending some bits of information, you know, who did you vote for, and then your representatives, you know, try and figure out what would be good for you or what you would want.

There are also some other, you know, potentially pretty compelling arguments for why more participation tends to be good in democracy. So a lot of people have the intuition that there’s something intrinsically valuable about people participating in democracy and policies actually in some more direct way reflecting or flowing from the actions or the preferences of the people who are affected by the policies. Decisions are more likely to achieve public buy-in or be perceived as legitimate if the public participates more actively in producing policies. If you involve more people, you often get wisdom of the crowd effects. You can actually avoid corruption by having people keep an eye on the process or have the process be more open and public. There’s also potentially positive social effects of having people really actively have a role in their government, that can, you know, affect things like a sense of unity or, you know, a sense of understanding.

At the same time, though, there are also lots of classic arguments for limits on participation. I’ll divide these into two categories. One are pragmatic arguments. The idea behind these is basically that greater participation would in principle be valuable, but it’s just really hard to achieve outside of, you know, certain specific domains. So one argument is just it’s hard for large groups to meaningfully deliberate. If you take, you know, a group of ten people and you sit them down, you have them sort of share each other’s viewpoints in a respectful, good faith way, and try to come to consensus, there’s some intuition this is a naturally, you know, easier thing to do than having a million people do this.

It’s also easier to ask, you know, let’s say ten people, a hundred people to take the time to really sit down and learn about an issue and think through the objections and counter-objections and come to some sort of conclusion about it. You know, this is sort of in some sense the point of representatives is, you know, it’s their whole job to do this and it’s hard often to get, you know, people who have, you know, other full-time jobs and interests and concerns to take the time to do this. In some contexts it can also be hard just to get people motivated to participate if they, you know, don’t feel like their individual participation will make that large of a difference or it’s unappealing for various reasons.

Then on the other hand, there are desirability arguments. So there are also concerns that people often put forward that at least in certain domains, greater participation would actually lead to worse outcomes. It’s not just that participation is desirable but hard to achieve, but that you actually want to have certain limits on how much people participate between elections. There’s a really long history of people, from, you know, political scientists to, you know, ancient political philosophers, perceiving that there’s a need to insulate certain decisions from the public to differing degrees.

So there’s of course, you know, classic writing by people like Aristotle expressing concern about sort of unmitigated democracy. Of course people like James Madison who are involved in, you know, drafting the Constitution of the United States, you know, were very pointedly, you know, trying to frame what they were doing as building a republican government as opposed to a purely democratic government, based on the idea that there were certain risks with direct democracy or really heavy involvement that’s not — that doesn’t involve some degree of indirectness between the will of voters and what actually happens.

And there’s a range of concerns that people have brought forward. So a classic concern is the idea that voters often have limited expertise on certain issues or lack farsightedness, especially in the context of ancient philosophers or, you know, work from previous centuries is often a much more uncomfortable notion that elites are just somehow, you know, better suited to making, you know, certain decisions. This of course, you know, this sort of strain in the work of people like Plato. Another concern people bring up is a risk of politicizing certain issues in some sense that there’s certain decisions where, if you have a large public debate about them, there’s some risk that it’ll become polarized or in some other way decision-making process will break down.

And there’s classic concerns people brought up about tyrannical majorities. The idea that you naturally want to protect liberties and rights and if you lean too far into democracy, if you lean too far into giving the public direct involvement in decisions without various levels of indirectness, then there’s a higher chance of liberties being violated. A lot of these concerns have faded over time. So I think you’re gonna find very few people defending the sorts of critiques of democracy people like Plato had, and I think we rightly view a lot of them as, you know, based on, you know, an unjustifiably strong form of elitism.

At the same time though, a lot of these concerns, even if not voiced very frequently, are still reflected in the way that the governments of most democracies work. So I think the most obvious example of this is probably judicial independence. In most democracies, there’s a really strong principle that judges are appointed or elected for like quite long terms. And then the decisions they make, you can’t really be directly influenced by voters. So in the context of something like the US Supreme Court, obviously Supreme Court justices are appointed as opposed to directly chosen by voters and they have lifetime appointments. So they’re just basically, you know, fixed for life and they’re basically insulated from any sort of democratic accountability.

And these sorts of institutional designs are, you know, are sometimes questioned, but they’re often taken for granted in a way that I think suggests a lot of people at least implicitly have the notion that, you know, there’s some areas where it’s good to have a bit of insulation. There aren’t that many people who actively defend, for example, the idea of having a public directly vote on what’s constitutional or not constitutional or directly vote on the outcomes of Supreme Court cases.

There’s also some more recent, you know, arguments in this vein have had a more empirical character. So I think the most extreme version of this is a book that came out last year called 10% Less Democracy by an economist, Garett Jones, who basically argues that the sort of Madisonian, you know, idea of trying to place limits on democracy or limits on participation does actually have some merit to it. And there’s some examples that are brought up as potential evidence in favor of this position.

So another really classic case is that most economists tend to think it’s good for central banks to be independent. It’s good for elected officials to not be able to very directly influence monetary policy and it’s good for central bank officials to not be directly elected by voters. And yet the thought process here is that if you allow for more democratic control over central banks, then this is more likely to lead to short-sighted decisions about printing money, which are more likely to lead to runaway inflation. And there is some empirical evidence here that central banks that are independent do actually seem to manage inflation better, at least if we assume that inflation is a thing that you don’t want to have that much of.

It seems to be the case that appointed judges perform better on certain metrics than elected judges, including metrics of impartiality. For example, at the state level, judges who are appointed show less evidence of bias in terms of deciding in favor of people who are residents of the state or voters in the state versus out-of-state people. Politicians also seem to be more likely to vote against academic consensus when they’re close to reelection as opposed to when they’ve just been reelected or they’re about to retire. So politicians are more likely to, for example, oppose policies like rent control, which are sometimes popular but most economists think would have serious negative indirect effects, although it can of course be debated whether, you know, policies like rent control actually in fact are bad.

And so another just, you know, very salient example of this is recently during the impeachment hearing for Donald Trump, a number of Republican senators voted to impeach, and it’s really striking that this was made up of a couple of senators who are about to retire, then a few senators who have just been reelected and so won’t have to face reelection for six years, and two senators who are in very safe seats. And there’s some suggestion that, you know, more people would have voted for impeachment, but based on plenty of accurate beliefs about what happened in that context and, you know, their own consciences, but due to sort of, like, lack of, yeah, I guess, insulation from electoral politics, they were more inclined to vote in a direction, at least from a subjective standpoint, seems ill-advised.

There’s also some evidence as well that when media starts to cover certain political issues, at least in the United States, it can increase polarization, and it can make it actually harder for Congress to pass consensus-based legislation on it. So there’s at least the, I think the arguments which have been more recently made. And I think it’s sort of interesting to reflect on what these arguments imply for digital democracy.

It seems to my mind that some initiatives like vTaiwan and Join, which have just been discussed, seem to actually partly refute some of the pragmatic arguments against participation. So you know, one just, you know, very basic point is that these initiatives exist. They have in fact, you know, involved huge numbers of people in Taiwan, and they have actually seemed to produce outcomes which are actually good and sensible, like legislation regulating Uber in Taiwan in a way that’s like broadly amenable to large groups of people. And it does in fact seem like some of the tools that the vTaiwan initiative uses actually do resolve some of the classic issues with scaling up deliberation.

So just as some examples, which I think have already been mentioned, the initiative uses software to automatically map points of agreement and disagreement based on upvoting and downvoting patterns. It tries to figure out where people cluster or sort of the different dimensions of agreement and disagreement, which makes it really efficient to take large amounts of sort of initially unstructured inputs and then give people actually a lay of the land and actually see, you know, where do we agree and disagree. It removes the ability to reply directly to people, which reduces sort of trolling issues.

It essentially upvotes comments or questions, which there’s a broad consensus around. If there doesn’t seem to be a strong polarization around whether a question or suggestion is good, it gets highlighted more strongly, which reduces this risk of, you know, if people pay attention, they’ll get more polarized. And just the whole thing seems to make the process much more efficient.

It also seems to me as well, these classic sort of desirability arguments don’t bite very strongly. So again, the fact that these platforms really focus on discovering points of consensus seems to reduce risks around issues becoming politicized if the public gets really engaged in them. It also reduces this risk of tyrannical majorities. I think it also reduces this risk of extreme divergence from expert views. If it’s a point that you actually need to get, like pretty broad consensus around, I think there’s less reason to worry that, you know, something basically very, you know, wacky will come out of it.

I do still think, though, that it is sort of interesting to think about these classic arguments, you know, regardless of exactly the level of merit you think they have, and sort of look at, you know, vTaiwan and these other projects and sort of think about, you know, do these actually imply limits or do projects like vTaiwan actually show that there’s, you know, these serious flaws in these arguments which were harder to point out when the arguments abstract, but may be easier when you actually have concrete cases of things that work in spite of them.

And so I just have some questions for both Hélène and Audrey. So one is, you know, how seriously should we take these sorts of classic concerns about sort of downsides or limits to participation? So I know Hélène actually is, you know, her most recent book and her other work actually engages a lot with these questions, so I’m sure she has a lot of thoughts. Then, you know, more concretely, you know, what are the limits of initiatives like vTaiwan? Are there actually any domains where they might be ineffective or undesirable? You know, for example, would something like, you know, using vTaiwan to set monetary policy, would that actually be a bad idea, or is that actually, you know, something that could, you know, quite plausibly work?

And then just more generally, and sort of, you know, zooming out, what are the actual limits of participatory democracy? So regardless of, you know, what you think the limits are at the present day, with enough technological and institutional innovation, I’m curious especially whether Audrey thinks it might be possible to just almost completely move beyond representative democracy altogether, and sort of get the benefits of much higher participation without the, you know, some of the downsides or limits that people believe exist today. Wonderful. So, Audrey and Hélène, feel free to respond.

So I’ll defer to Hélène on the more theoretical points. I would just, again, in a reverse chronological order (laughter), answer this thing about our relationship to representative democracy. I don’t think about representative democracy that much, because I work on the new system. It may or may not make the old system obsolete.

But we do have a process, as I mentioned, the participation office or the PO process, and we have our monthly vote because our bandwidth to process such open deliberation meetings, including Pol.is and a Join platform, although it has already enabled, like, scaling this out a lot, we still have to survey the stakeholders, still have to do the research, still have to prepare this preparatory material, this informed handbook of deliberation and so on, like all the classic deliberative democracies do. And so because of that, we only have bandwidth in our national team of around 100 participation officer to take a case or two at most every couple of weeks. So on average, our bandwidth — we’re now almost on our 100th case now. So on average, we process around two to three cases per month.

So we vote on it. Each participation officer can every month bring a case that they think need cross-departmental deliberation, or people, after they join a petition on the Join platform and collecting 5,000 signatures or more, also automatically nominate for the participation officer voting. And we ask the participation officers to evaluate on three things, and this is a national principle on processing the collaborative topics and open government collaborative meetings at po.pdis.tw. And the three pillars is that it must have a broad, complex stakeholder groups with diverging views. And also it needs enthusiastic public participation, and also interdepartmental collaboration seems warranted.

And this part is, I think, the most important, because this is a very technical subject, and central bank very good example at that. Then the central bank participation officer would say this is monetary policy and there’s nothing interdepartmental about that. And that will actually just move it back to career public service. We don’t insist that they have to deliberate monetary policy. On the other hand, if they are now evaluating CBDCs, cryptocurrencies and things like that, then that is actually interdepartmental. I can easily think of five ministries related to it, and then that becomes more amenable for this kind of participatory democracy.

Again, if the stakeholders, they don’t have diverging views, then this is a waste of calories of us doing the discovery process. And also if the stakes are simple, and especially if it’s like a zero-sum deal, then we don’t usually use this process, because there’s very small chance that we can actually genuinely discover something that is of common value that people can move forward. And so a lot of the other points could be answered simply by saying we’re just setting the agenda. We’re not doing actually binding decisions. But that’s my, like, first, like, zeroth take, and would really want to hear Hélène’s take on it.

Can you hear me? Yeah. Okay, great. So I just wanted to consider the objections to participatory democracy, you know, on the ground that people are irrational, ignorant. I mean, it’s a very common objection, but to me it’s really premised on a vision of participation that takes place within an existing system which is designed almost to produce ignorance and apathy and lack of participation. So I don’t think you can reason — you can infer the competence or promise of participatory democracy based on the existing system where there’s very little participation.

And so when people think of including people, they think of including voters that are tribal, partisan, ignorant, misinformed, manipulated, et cetera, et cetera. Well, that’s not what we’re talking about. We’re talking about a different ecology, a different set of institutions where people are incentivized to act in a very different way than they are incentivized to act under, you know, the current electoral system, broadly speaking. So I’m less worried, and I think in fact it’s always this very theoretical objection that sort of dissolve when confronted with the empirical reality of success, you know, as demonstrated by Audrey’s work.

That said, the limits to participatory democracy, I’m willing to believe there are some. It’s just that we have so much more to discover before we have to ask this question. I feel that maybe it’s a little stifling to start with this question, I would say. That said, again, one limit could be the time that’s required of participants. It is time-consuming to be invested in these activities, and we don’t have an economy or an ecological sort of environment that necessarily gives people the time they need to do that.

I mean, they still do. You know, Wikipedia is basically born out of free labor. You know, and all these people who participated in the Finnish experiment or on the Uber deliberations, I mean, they did it. I don’t know what, you know, how, between the cracks, between preparing meals for the kids, between going to work, while waiting for the doctor. You know, who knows? But they do it. So we don’t wanna tax that too much, however. So we have to rethink perhaps compensation schemes. Maybe we need a universal basic income for this to be truly feasible and truly inclusive. So there are limits, but I don’t know if they’re inherent to participation so much as to the environment in which we place this participatory ideal. So yeah, I’ll stop here about the — one more thing maybe.

I’m not totally familiar with the book by — 10% Less Democracy, but I think the worry is about referenda and this kind of like purely aggregative judgment elicitation on very specific issues where the citizens do not control the agenda. But I think there are, you know, these methods of generating an agenda on the basis of a sort of a consensus that you sense and build, you know, through technologies, it’s very different. I mean, there’s still the possibility of capture. I’m not saying it’s not there, but it’s less the case than in the Californian system, where indeed the agenda is pretty much in the hands of whoever can buy the most signatures. So I think it’s a little different.

That said, I would be curious to hear Audrey talk about the danger of capture. For example, in the vTaiwan process around Uber, how did you make sure that, you know, Uber didn’t come in the way they kind of did in California, in the referendum on the gig economy, to push for their own interests against the interests of the other stakeholders? So how does your design prevent that kind of capture?

Right after the UberX case, we deliberate the Airbnb case, again in 2015. And Airbnb sent a newsletter email to each and every Taiwanese member saying, “Go to this Pol.is site and support our company platform.” So this empirically actually happened. And there’s two features in the vTaiwan process, and broadly the Pol.is process, that guards against this kind of attack.

One is very simple, in that the area, calculated by K-means clustering and shaped by principal component analysis, is actually not measuring the number of people. If you get 2,000 people voting exactly as I do, then you see an extra zero here on group D, but the area do not change. And it actually changes nothing about the binding agenda, because the binding agenda need to have cross-group supermajority, meaning that you actually — what matters is convincing people of different ideas, of different aisles, of different clusters. And so it doesn’t really pay to mobilize 5,000 people voting exactly the same. That’s the first thing.

And also the second thing is the crowdsourcing is actually a wiki survey. Most ways of capturing astroturfing and so on only works if the bit rate is low, like referenda, just one bit literally. And then it’s easy to distract. But according to our numbers, people who get motivated by Airbnb — lobbied, mobilized by Airbnb — they get to the Pol.is website, they see a bunch of really reasonable ideas about Airbnb, and only a third of them actually agree on all the core things petitioned by Airbnb. And the other two third discover more nuanced, more balanced way of expressing their preference, because the 99 statements are literally 99 dimensions, and they very quickly gravitates toward the dimension that are more dialectic, more nuanced, and not at all the Airbnb position.

And so increasing the bit rate, allowing people after voting for a while saying, “Okay, I don’t like any of these feelings. I will share my own feelings for other people to vote.” Again, just costing them, I guess, a minute, because it’s tweet length, literally tweet length. So however long you compose a tweet, that’s the time constraint (laughter), time burden put on you. And just like tweets being shared motivates people to share even more, more people are motivated to think of more nuanced statements, because they win that way, right? Because the only such statements that convince across the aisles did get into the final agenda that is binding in a multi-stakeholder face-to-face deliberation.

So to recap, the first thing is measuring diversity, the plurality, instead of the numbers, the headcount. The second is that a higher bit rate, a higher dimension for people’s preference to be expressed.

Yeah. If I may add, it’s exactly what we did as well in the Finnish experiment. We had a way to cluster, you know, groups, and to identify where they overlapped in suspicious ways. And so it was about the diversity rather than the num — the headcounts, indeed. Because especially in our issue, it was regulating snowmobiles, and so 80% of people on the platform were men, very libertarian, very anti-regulations. And so we had a — you know, if we had gone by headcounts, the views were really one-sided. But we had also some women, mothers of young teenagers, who are really worried about the speed limits and age restrictions. And so we were able to visualize the different groups, and so we were much more focused on diversity of arguments and perspective than headcount. Absolutely.

I’m now coming to think that the difference between, you know, the nightlife district of the internet, the antisocial, addictive, toxic, addictive drink, private bouncer part of social media (laughter), the antisocial part, and the digital public infrastructure, the pro-social part like Pol.is, I think is whether those credibility or importance are measured by artificial engagement scores, which is the norm if people want to sell advertisement or surveillance capitalism, or whether they’re actually measured by diversity and plurality.

If they get measured by plurality, it means that people spend time on the platform in order to empathize with each other more. And on the antisocial corner, it’s almost as if people engage to empathize less. (laughter) And that is what enable bad actor to gain money and views by publishing content to exploit the biases. They would change the scoring mechanism. There’s no such incentive anymore.

Yeah, and I think it’s also worth emphasizing, I think, because we get that objection a lot about the self-selection and the bias and all that. But they’re not deciding. So it’s really like, what’s really neat about what you’re doing, and what can be done in that space, is that you can really segment the process so that there are different steps. There’s ideation, there’s problem-solving, and then there’s decision. And when it comes to the decision phase, that’s probably where we still rely, as you said, on the traditional accountability mechanism of elected officials, who are going to have to endorse this input, which is so much richer than what they get through elections.

So but at the end of the day, there’s still this moment where they have to decide. My addendum would be to say, well, okay, but we don’t necessarily need to have them — I mean, it’s indeed debatable — but even that final step need not be in the hands of elected officials, if we can figuring out a way for randomly selected assemblies, for example, which to my mind are an authentic form of democratic representation, can also be made accountable in some non-electoral ways. So I think, you know, if you combine your ideas and (laughter) visions and some of like the theoretical constructs I propose, I think we have a completely different democracy. We’re outside the representative government model, which is — has shown its limits.

Yeah, I totally agree. The second diamond, it could be a sortition-based backend, it could be a liquid democracy-based backend, or it could be, you know, traditional voted-in electorates. But as long as we talk about rough consensus and the common how-might-we questions, then it’s compatible to each one of those backends of those second diamond. Yeah.

So just in terms of process, we are at the nominal ending of our event. But Audrey has mentioned that —

Yeah, I don’t really have anything after this.

— has a bit more time.

I can do this for hours.

(laughter) So I don’t know about Ben and Hélène’s availability, but I thought that conversation was very interesting, so maybe we can continue for a bit. And I’ll start by posing maybe a skeptical question to Audrey and Hélène. So, you know, I thought Ben laid out, you know, some of the concerns with direct democracy pretty effectively, and I’ll just flag, you know, it’s hard maybe, exposed to Western media and social media, to not be concerned about the, you know, any or various proposals of enabling the public to be more engaged in political discourse. It’s easy to imagine ways that that can go awry. But rather than debate it on theory, which is also useful, I am ultimately more persuaded by empirics, and so this is why I’m so excited about the achievements that have happened in Taiwan.

And then my question for, I guess, the panel is, what are the sort of lowest hanging fruit, the highest impact, but most likely to succeed deployments of civic technology of the kinds we’ve been discussing in, say, the United States or the UK or other countries that, you know, could be achievable and upon being achieved would represent a hard test of what’s possible? And Hélène, for example, to your kind of argument that we don’t know how capable, say, Americans are of a participatory democracy because they’re not given a chance.

You know, my skepticism is it may be that it takes years, decades of maturation and practice and sort of development of the culture and national consensus and national identity and sort of a range of things to have a civic community that’s able to do this well. But let’s turn this into an empirical test. What would be a sort of a high stakes rollout of civic technology where we could see how capable, say, Americans are of this kind of pluralistic direct participatory democracy?

Do you want a hard test and not an easy test?

Well, the hardest test that you think civic technology can pass. Right? ’Cause the harder the test, the more persuaded skeptics will be that this can actually work.

Okay. Okay. So I have some ideas. I wouldn’t recommend starting there because I actually think (laughter) that we wanna go step by step. But I think probably the hardest step in Europe, for example, would be around issues of immigration. Immigration, European identity, the place of, you know, Islam in Europe, it’s a huge issue right now that it’s like the third rail of politics, you know, and it’s going to be defining of the kind of communities we build. But we don’t let people talk about it. It’s too dangerous. There’s a total fear and distrust of ordinary citizens around those issues. Accusation of, you know, fascism, racism, Islamo-leftism, whatnot. In France, it is awful. So that would be the hardest test for me.

I don’t think I would recommend starting there, but I actually believe that, you know, if we road tested this format on other issues like we currently are in France, for example, with climate change, I think eventually we could get there and get to much more satisfactory solution. For example, around separatism, you know, between the Muslim community and the French community right now. So the only solution that the government has come up with is some kind of like redistricting of schools to make sure populations are mixed. So basically we’re forcing the kids who have asked, you know, for nothing to change school to be reallocated. Meanwhile, the parents don’t talk to each other.

Like, why don’t we start with a conversation between the parents about ways they could live better in, you know, together in neighborhoods that are 50% Muslim, but, and then asking people who are, you know, in a neighborhood that are very white, like, why is that? And having an open, trustful, considerate, respectful conversation about what it means to be French, what it is to live together, what it is to live in a plural society. I suspect similar issues around race in the US would be really complicated and issues of reparation, issues of, you know, like all that legacy is still so raw. So I think for me, this would be the hardest test. So I wouldn’t necessarily start there.

Hmm. I’ll continue this line of thought. I guess in many cases, the Sunflower Movement is a really, really good hard test, because the thing that we deliberated on was whether we deepen our service related ties to the Beijing regime, and this has like 20 different aspects, one of which is whether we allow Beijing-based companies to manufacture our then new 4G deployment telecommunication equipments. That was in 2014, but it’s very on topic now in other places as well for 5G deployments. And this has the nature of first, it’s specific enough that people who have good arguments on all the different sides can contribute.

But also this is of a shape that it’s unclear how traditional representation may work, just as there’s no, you know, teleworkers union, there’s no, (laughter) you know, association of people who set up companies in Cayman Islands and so on, which by the way, are the first two cases that vTaiwan processed. There is no easy way to discover in traditional representative like union association means exactly who are the stakeholders of, you know, PRC manufactured 4G components. But on the other hand, this is a very serious cultural issue as well. It pertains to trust and trustworthiness and so on. And but there’s also economic arguments.

I would say that a broad, like, wicked problem with no traditional representative solutions, but was specific enough that people could actually feel it, like lived-in experiences which allows the sharing of feelings, not jumping to solutions. I think these two criteria together, whatever fits the criteria in any given policy would be my suggestion, and that will actually make our implementations very easily translatable in circumstances, like I mentioned, because then people would just demand whatever existing data, remember the facts before the feelings, whatever existing data about the situation being made publicly addressable with its permalink, say, URLs and so on. My answer is also partly to answer Seb Krier’s highest rated top questions because, without answering questions, we might as well be feeding the people a pre-requirement.

And I would say also that one hard question that was a high priority for example, the French public was the environment and climate change, and that’s actually a hard test that was kind of, I mean, to my mind, successfully passed by the French Convention for Climate just recently. They came up with 149 very elaborate proposals, almost bills, you know, not technically because they constitutionally cannot pass bills, but they were like bills, on ways to curb greenhouse gas emissions, you know, by 40% in a socially fair way. And so they were able to do that even though we have a lot of climate skeptics in our population, even though they came from all kinds of horizons and party, you know, partisan beliefs and so it’s been done.

What’s not been so successful is the articulation of their recommendations and their work to the traditional representative system because what happened is that once their proposals went to parliament and to the government ministers and now parliament, they basically diluted and watered down the content of the law. So it’s really this question of how to reconcile the two logics, the new participatory logic and the established, you know, group interest-based logic, the lobbies, the — it’s much — it’s not quite, I mean, it seems to be quite successful in Taiwan, but in France we haven’t cracked that nut yet.

And I would just add, on the transcultural or global scale part, it just occurred to me that, for example, coronavirus is a really good example here because we did run a Pol.is-based, actually five Pol.is-based conversations at cohack.tw, which has participation from seven countries and many different teams. And the Pol.is rough consensus was centered around what’s the acceptable privacy and ethics boundary to the responses to pandemic, including, of course, digital surveillance, but also, you know, making the PPEs, protecting the vulnerable groups, making sure that ICU are at capacity, supporting frontline staff and essential workers.

And before the pandemic, we couldn’t even dream of having a five-country deliberation (laughter) of this kind because people have very different timescales and urgency. But because of pandemic, everybody has the same urgency, and then we can deliberate on how exactly is the norm when a digital tool and what’s the privacy acceptable trade-offs and boundaries so that we can actually innovate into, say, participatory self-surveillance to not be captured into surveillance state or capitalism in our response to the pandemic. So this is a very recent example that kind of transcends this cultural barrier because of a shared urgency.

Yeah. And I would say there’s a trade-off between, you know, we wanna go slow enough that we can convince everyone that these things work and we can transition smoothly from one system to the other. At the same time, on some issues, you might say there’s no time, and so we need to do this right away. And for example, I’m thinking this climate change, you know, stuff, there is, as I said, a global climate assembly in the works that’s going to meet starting in September, I believe, with a thousand people.

We need — this needs to be successful because ideally this would become part of the political institutional, institutionalized apparatus of a global governance if it proves successful, and then they would reiterate it every year until, you know, it’s smoothly running, well articulated to the other governance, you know, paths. So yeah, I think it’s hard. You wanna go find the right pace and meet the right trade-offs.

Yeah. So one maybe takeaway is, it’d be great to have a list of policy domains or areas where civic technologies could be rolled out, and you can score them by the importance or urgency, and then times the probability that the civic technologies will succeed or improve the deliberative process. And then, you know, sort by that and work your way down.

Yeah. I think there are also different issues where it’d be interesting to, you know, there’s like different potential pitfalls of it. So if I imagine, for example, something, you know, like vTaiwan being used in the United States, I imagine a bunch of failure modes for, let’s say, it being used to talk about reparations, (laughter) you know, being different than the potential failure modes for it being used to talk about something like the minimum wage, where in one case it’s, you know, an issue with huge value disagreements and huge disagreements about, you know, how you ought to interpret history and things like that.

And then something like the minimum wage, it’s, you know, this, you know, actually like quite technical issue where economists super disagree about, you know, like with a $15 minimum wage, you actually have employment effects. How large are the welfare impacts of employment effects? How large are welfare, you know? Like, and so there’s, yeah. And so there’s, you know, interesting things that you might have about how well can this actually integrate to technical academic disagreements. There’s concerns you might have about does it just break down if people have like very different values and it’s, you know, really polarized to begin with. And so I think different issues might be interesting test cases for different sorts of concerns.

Can I — actually, on this, I think that there’s a difference between issues where there’s an actual disagreement and one where there’s just a capture of the issue by interest groups that make it look as if there’s a disagreement when in fact the vast majority of people have already made up their mind. Gun regulation in the US, there’s a vast majority in favor of much more aggressive regulation, but it’s not happening because the system doesn’t allow for it. So here, that’s a low-hanging fruit for me. I’m sure if you did a vTaiwan type of process on this combined with a deliberative assembly, it’s done. It’s a done deal in three months, if Congress commits to passing whatever recommendation they make. Of course, which again is gonna be the key problem and the key bottleneck.

Then there’s the question of real disagreements, maybe on, you know, a reparations or something like that. At the same time, what we know from experiments in Ireland recently on abortion, very divisive issue, very, about very fundamental values. When confronted, when the problem is framed as a problem rather than as a value disagreement, even pro-life can converge on the idea that, well, you know what? When it comes to the law, we should decriminalize abortion. I’m still against it. I still think it’s a tragedy, it’s a crime, this and that, but the life of women, you know, is in the balance and I have to set aside some of my private beliefs to, you know, converge towards the disposition.

So if we can do that on a fundamental value like that or, you know, a set of questions like that, why can’t we do it on almost anything else? I just think that we have yet to touch the frontier of what’s possible. Yeah.

Okay. Well, let’s see if there’s any last questions. But otherwise we should probably excuse our many brains that have contributed here. So just scrolling through some of the top voted questions. I think most of these have been partly addressed. One... Oh, Audrey, I see you replied. Yes. So it’d be great to get the slides shared, which of course with openness as a principle (laughter) that would be the case.

Okay. Maybe I’ll just... These are sort of relatively narrow questions, but I’ll pose these. One is from Shivangi Rajora, which asks about the tension between the principle of something — decisions being fast or processes being fast and the processes being inclusive, so we’ll put that one. And then there’s a related sort of narrow question, which has already been slightly addressed, but the trade-off between, I guess, different processes and selection biases and representativeness. So to the extent that there’s a bit more you — either of you would like to say on that, that’d be great.

Yeah. When I said our process agile, it has a specific software development meaning. It doesn’t only mean that it’s fast, but it also means there’s a continuous integration of feedbacks. So when we say we fix the system, we deploy a new version every Thursday, it means that we go through this mini diamond, like people calling 1922, saying that there’s something wrong with mask distribution or the MPs making interpolations and so on.

So we do a approach a fast decision, quote-unquote, lowercase d decision, every week. But then that decision, lowercase d, is open to be forked by the social sector and also any good ideas gets amplified in real time and we schedule it for next week’s deployment and we continue to do that. It’s more than one year now, our central epidemic command center. So this continuous delivery, continuous integration is the backbone of being inclusive while being fast.

Yeah. On this, I would say that it sounds like a, you know, either/or type of question. Either you’re fast or you’re inclusive, but oftentimes processes that claim to be fast and sacrifice inclusiveness in the name of speed end up delivering terrible results. So it’s perhaps better to be a little slower and making sure everybody’s on board and in the end save time compared — because you will deliver better results in the end.

I see it a little bit in, well, I mean, the contrast between Taiwan and France on the management of the pandemic is very telling. Oh, we have a very, you know, active, vertical sort of government which took charge and like, you know, implemented, without any kind of deliberation, confinement after confinement, curfews, this and that, followed the experts very deferentially for a while. Well, in the end, we’re doing much worse than Taiwan did. So how is that, you know, how is — how was speed so great in our case?

And maybe we had some other issues. I’m sure it’s not all things equal otherwise, the comparison, but still, I suspect that had we been a lot more deliberative, slow and inclusive in the beginning, had we sort of taken the temperature of various communities, listened to different groups instead of just following experts, I think we’d be in a better shape now. Same on the vaccination campaign.

So I actually have one more question that hasn’t really been asked. So, you know, we are the Center for the Governance of AI. We’re often looking especially towards AI technologies. So I, you know, I don’t know how much anyone on this seminar has thoughts on it, but what are your views about what machine language understanding could do to civic technologies, or, you know, some of you mentioned AI facilitators for deliberation.

So is this, you know, is this sort of over-hyped, that it’s unlikely that we’re going to have meaningfully useful tools anytime soon, or do you see in the coming years the rollout of AI tools that are sufficiently competent with language understanding that they can really amplify the benefits here?

In my experience, anything that we can very easily explain to six years olds or I guess ELI5, five years old, is good for facilitation. And I mean it by, for example, Pol.is is a kind of AI, but it’s not deep learning. It’s just principal component analysis. And you can actually explain K-means clustering and principal component analysis quite easily, as opposed to, you know, the latest convolutional or transformer models of deep learning. So that’s already a quick — a easy win.

The other thing is about the modality for engagement. For there’s some people who prefer text-based conversation. There’s some people who prefer a visual, like mind map and so on. There’s people who couldn’t really think through without a back and forth conversation and so on. And machine translation as well. We run the cohack.tw with the machine translation powering Mandarin English bi-directional conversations as well.

But all these functions are very, very simple to explain. They’re very, very simple to correct if there’s any biases. It doesn’t really substitute judgment. So as long as the machine learning is for collaborative learning, assisted collaborative intelligence, I think is a boon. But as long as we spend more time explaining it than the time is saved by automating away, then I don’t think it’s a good idea.

Cool. So my own view is that I think it’s gonna be unavoidable to use some kind of automated facilitators at some point because if we’re gonna have conversation at the planet scale, the cost of micro-facilitating each group of six or seven people is — it’s just too much, so. And this task doesn’t seem too hard to automate, I think so. I imagine this would be a great way to augment democracy in the future.

And then the way I see AI, I mean, again, I’m not an expert at all, but it to me it’s like an improved mirror that we can hold to ourselves to explore our collective psyche a lot better and, you know, our history, the facts about our culture in relation to other cultures. I think it’s an amazing tool that as long as we keep it as a tool is, you know, same way that experts should be on tap and not on top, I think AI should be available, but not in control of our lives. So I think this is the fate of every technology. From that point of view, I don’t think it’s very different.

Yeah. I’d also — it’s the not something which will be available in the next decade, but the long run sci-fi thing I suppose I’m excited about is if, you know, we ever get to the point where you can actually really communicate, you know, your values to machine learning systems. You can really have machine learning systems have a good understanding of what your preferences are. That seems like something that could really seriously overcome a lot of the issues with participatory democracy, like especially the time cost.

Like if you can in some way automate the process of giving feedback about, “Oh, I would like this, or I wouldn’t like this,” then that also seems like, you know, something we’re not about to have, but something in the long run we might look forward to that could really allow for much more, you know, meaningful representation of viewpoints in the political process.

Mm-hmm. I think that’s great as long as it’s democratized. And I don’t mean it as cheap and accessible like the 21st century use of democratization. I mean the old meaning of democratization, meaning citizens’ control, and also co-governance.

Great. Well, with that, I think it’s a excellent place to conclude. I just wanna thank our panelists again. Thank you Audrey, Hélène, Ben, for this wonderful conversation. Really exciting ideas. I do think this is the frontier of democracy that’s being theoretically and empirically explored, and I’m extremely excited to see the continued empirical evidence of your innovations rolling in, so we can, you know, see what’s possible for the future of democracy. Thank you.

Thank you, Allan. Thank you.

Thank you, all.

Yeah. Thanks so much.

Thank you. Live long and prosper. Bye.

Same to you. (laughter)

Live long and prosperity, yes.