Automatic transcription, there may be errors.

Host: Alessandro Oppo
Guest: Robert Bjarnason, Co-founder of the Citizen Foundation

Alessandro Oppo: Welcome to another episode of Democracy Innovator podcast. Our guest today is Robert Bjarnason, the co-founder of the Citizen Foundation and many other things, but I think you will tell us something about this in this episode. Thank you for being here.

Robert Bjarnason: Thanks for inviting me.

Alessandro Oppo: As a first question, what is the Citizen Foundation and how did you start it?

Robert Bjarnason: The Citizen Foundation is a nonprofit - we're actually two nonprofits. One is registered in Iceland and also one is American in the US. It started in 2008. I was actually living in London at the time when the financial crisis hit Iceland. Iceland was a pretty vulnerable victim, and over a period of time, all the news was about Iceland being bankrupt and so on.

You know, when going by taxi in London, talking to the taxi drivers - "Are you from Iceland? So unlucky, your country is bankrupt." Well, actually it didn't turn out really that bad in the end, but at the time it was an overwhelming feeling in Iceland where trust in government and parliament plummeted down to six percent. This is one of the oldest parliaments in the world, founded in the year 930, so like over a thousand years old. Over a thousand years of pretty good trust in the parliament, but it plummeted from sixty-seventy percent to six percent in two weeks.

Then it was just overwhelming - this idea that we weren't really in control over what's going on anymore. It was certainly in Iceland, but also in wider society in the rest of the world, at least the Western world like in the UK where I was. So we had this idea - this group of people, mainly my partner and I who started the first internet service company in Iceland in 1993, plus some other people from our network. We had this idea to start a civic tech nonprofit to help create technology to enable better decision making in society between governments and citizens.

We've always been using a lot of open source software and promoting and creating open source software. The basic idea was: how can we enable, by using technology, the internet, and AI and other things - how can we help governments and citizens make better decisions in an increasingly complicated world?

Alessandro Oppo: I was wondering, when you had the idea that technology could help people make better decisions - you mentioned 2008, but you also said that you opened an internet provider. Was that also connected to the idea that the internet could bring more discussions?

Robert Bjarnason: Yeah, absolutely. I was twenty-one when I started the internet company in '93. The internet existed since 1987 really - the text-based internet - but then the web came and we started this internet company. There was a lot of idealism about how this increased communication could give people a stronger voice in society, and even in a utopian way, how the internet and cyberspace could really transform social power structures and make society fairer and more equal.

So yes, I was definitely a part of that. That's also why we started that internet company - I'd been helping schools connect to the internet. I was at the Teachers University, and I downloaded the first web browser called Mosaic. I'd been on the text-based internet for years, and I tested it and was like, "Wow, this is the future."

We actually downloaded the first public Linux operating system called Slackware on floppy disks. That was all for free - Linux, you know. We set up telephone lines in my friend's living room. So I'd always been interested in using all this open source software as well.

So the idea in 2008 - opposed to this idea of the internet - I mean, the internet was not really delivering. Yes, it's given a lot of people voice, but when it comes to actually giving people a voice that can influence decision making, there was almost none.

But also, just in general, how open source community-based things can really work. Linux showed that the internet showed that you can actually have grassroots open source solutions that really make big changes in society. That's also the reason why we decided that since we're going to be working with citizen engagement to build up trust, we couldn't be a for-profit company. We had to be a nonprofit organization because otherwise we wouldn't have the right alignments.

Alessandro Oppo: In your opinion, is the internet still a space of freedom, or can it also be a space of surveillance in some way?

Robert Bjarnason: I think the internet is still a place of freedom in a way, but it's also basically a place of surveillance. There are different types of surveillance in different places of the world. In some places it's highly monitored and not really free at all for free thought. But I think the internet, like everything else, just reflects the society that it's embedded in.

Alessandro Oppo: So what are the main solutions that you propose to this distrust that is missing regarding, I would say, classic democratic systems - Western democratic systems?

Robert Bjarnason: When we started, we felt like, "Okay, so how can we use the internet?" We started thinking very quickly about e-democracy, collective intelligence. There were quite a few people - academics and others - trying different things.

The basic idea is that if you have a complicated decision, and in the modern world we have increasingly complicated decisions with competing demands - what is considered a good decision today maybe is not exactly the same as twenty years ago. The world is evolving, it's not static.

The idea is if you're doing a complicated decision or decision-making process that's going to affect a lot of people, then actually by reaching out at the right time when you're creating the policy to citizens to actually get information - this basic principle is that we're going to make better decisions with better information. I think few people can argue against that - it's even mathematically proven that more information is going to give you better decisions.

That's the core of the idea - the utilitarian part. But obviously there's also the democratic angle of it. What makes a good democratic decision is something that is going to be balancing the competing interests of the voters and the people who are living in a society that the rule or regulation or law or decision affects.

This can be small things like participatory budgeting - are we going to put a playground there or an outdoor hiking park? There are lots of good examples of participatory budgeting all around Europe and the world. But also policy like educational policy, traffic policies - all sorts of policies where you have many different stakeholders.

For example, the Scottish Parliament, where the platform and this idea is embedded into the committee system of the Scottish Parliament. They've been doing really groundbreaking work, leading the world in many ways when it comes to engagement. They have an engagement unit and they take it very seriously, very professionally. They offer like a service to the different committees to say, "Oh yeah, we have this issue, do you need information from the people, from the public, before you take this decision?" And then the committees say, "Oh yes, we need information on X, Y, and Z," and then that gets implemented.

It's like a way of actually delivering real value to the elected representatives in the committees by getting good information to them. This is used quite many times. It's just examples of how this kind of engagement can both lead to better decisions but also help empower people democratically because they can influence the decisions themselves by taking part.

Alessandro Oppo: I was thinking - so now we have representative democracy. Do you think that in the future with this kind of deliberative tools like participatory budgeting and all these new tools that maybe use AI, we could have some different kind of political system? Maybe without representatives, maybe still with some representatives or professional politicians?

Robert Bjarnason: Obviously anything is possible in the future, but I think democracy, law, and everything - how it all connects together - the norms that people are used to, there are all sorts of different things that need to change together for some really major changes to the system.

One of the things that we've been very much against in our work for seventeen years since we started this is direct democracy. We have taken clear stances that there are lots of decisions where direct democracy can't work today.

I think that's one of the things - this idea of proxy voting, for example, where you delegate your vote to certain parliamentarians. There are issues with direct democracy. We have referendums obviously, and Switzerland is obviously famous for that, but we have taken this stance against direct democracy for several reasons.

One is security - how secure are you doing it online? We're lucky here in Iceland, and also Estonia and some other Baltic countries, where a lot of people have electronic IDs that enable any type of direct democracy. But if you don't have reliable electronic IDs, you shouldn't be doing direct democracy where people vote on insecure platforms about something important, because that's going to totally undermine democracy. It's going to create distrust in democracy if people are voting on platforms that are not secure. That already limits lots of places from really doing any reliable direct democracy.

But we also draw the line at participatory budgeting things where people are voting for priorities. In Better Reykjavik - our main project that's been running in the city of Reykjavik for fifteen years now - people have a chance to put forward ideas about what to do in their neighborhoods. Seventeen hundred ideas came in a couple of years ago. Six months later, the city looked at them, professionals costed them, and then people can vote with their electronic IDs on what ideas are going to be implemented in the neighborhoods.

But if it was about policy, then we have a totally different dimension where the power actually lies. The thing is that the power in online things is controlled by who knows about it and how you control who knows about it. You do that through social media, and how do you control who sees what on social media? You pay for it.

So basically anything you do online for direct democracy - we're voting on this issue or that way - for us is not in line with our values. We always recommend against it because even if you have electronic IDs, if it was a controversial issue and it's all online, it's just too easy to manipulate. That's just the problem of referendums in general - referendums can go all sorts of different ways recently, as we've seen.

But having said that, I think what we can see in the future is that AI is going to help empower the organizational side of governments because often the reason why there isn't participation is because of lack of resources and lack of planning capabilities to actually incorporate the feedback. I think AI is really going to empower that part - it's going to increase the capacity of government to actually work through good decision making exponentially, not just in the next two years.

But then also with AI, it's going to help empower citizens. For example, with doing a new iteration of Better Reykjavik later this year - we have over forty thousand registered users - we're going to have an AI system that is going to allow people to have an AI agent that's basically going to be watching out for their interests in the city. Depending on the subjects they're interested in, it's going to be like your agent watching out for your interests.

I think both of those aspects - both in terms of more capacity for governments and also more help for citizens - that's going to really help with the quality and quantity of citizens working with government and AI for better decisions.

Alessandro Oppo: I was thinking about power and information. If I don't have information, I also don't have the power to choose to do A or B. Then there is this question about the lack of participation - if people don't have the information, they can't actually participate. If they had information about a certain person, maybe that person could see about changing something in the neighborhood where they live. I think this is a cultural problem related to education, because maybe in school we are not taught about how we should participate and how it is important.

Robert Bjarnason: Absolutely. One of the things that we've done over the years is that especially after mobile phones became popular like in 2013, 2014, 2015, when everybody had smartphones, was to lower the barrier to participation both through technology and design.

For example, when you are visiting a civic project on the Scottish Parliament or something, the first thing you see is a nice image, then there's a very short description of what to expect, and then you can see more through a "read more" button. It has a very simple user interface which we also have designed for deliberation.

Instead of when people see an idea asking them to comment on it, we ask them to help us out - help us find the pros and cons. We ask them to help us find the best arguments. I think that's an issue - people need enough information, but also how the information is presented to them and at what level you break down information.

For example, we did this with the government thinking about constitutional changes in Iceland. We worked with them on a Better Reykjavik platform where they basically had to break down the constitution into parts - like four different things they wanted to change. You could have put a big PDF with all the legal text and asked people to give comments, but we knew that would never work.

So what we did on our platform was create pages where you have the three top-level topics in very short text. You click on each one and then there are four things that the government wanted to do in the constitution connected to each topic - very clear, almost lowering the barriers so almost everybody could understand it. That simple change, compared to the PDF with the actual law proposal - that's like a barrier that ninety percent of people can't deal with. We lowered it so that ninety percent of people can process it because it's been broken down like that.

That's really important. But I think when it comes to the information part itself, I think AI is already helping people who are using ChatGPT or Claude or whatever. It's both doing searching for you, it knows your preferences, it's like on your side in a way. If you get a subscription to ChatGPT, at least now it has no advertisers, it's just you and the model, so in theory it's going to be on your side, knowing your preferences.

What we're doing with our AI agent for Better Reykjavik for your priorities is that it's not just looking out for your interests in the city, but we are actually going to build it in an open source way where you can use open source models, building it in a way where you can basically tell the model, "Those are my interests and the things in the city, my needs - I'm a parent, I have those children with those thinking abilities, I have a car," whatever. Then the AI will help you find the information.

I think about not having the information - if you go back thirty years ago, you had some people who would read all the newspapers every day and have all the information, and some people who didn't read papers at all and wouldn't be up to date on anything. But today it's totally different - information is coming from everywhere, some of it is true, some not. So having the information has changed a lot because there's just so much information.

I think AI is the solution to help us deal with that by understanding your priorities.

Alessandro Oppo: Is Your Priorities the main platform that you're working on? But before asking you about it, would you like to share something about your professional background - academic, but whatever?

Robert Bjarnason: Very short on my academic record, but I started - my father's an electrical engineer and my mother's a doctor. My father was early into computers, and I guess I was hyperactive, so he got me programming at nine. I had a computer called the Oric-1 microprocessor - it was like a training computer for kids. I got interested and really spent all my energy on it.

I sold my first piece of software in 1984 - a teletext information booking system that would scroll information. It was used by the national government for meteor strike information in 1984. I sold my first game program in 1987 in a programming language called Prolog - an expert system.

Then I took a break from computers. I was involved in seven feature films, spent a year as a substitute lighting director for the opera, club promotion, putting on raves, things like that. But then I got back into it. I was at the education network, which was a pioneering internet network for teachers, helping connect schools to the internet. That was when I saw at the Teachers University the Mosaic browser, and we started this ISP.

In '95, I started the first of two ISPs in Denmark. Then I was mostly traveling between there and San Francisco where I was working at a video game company. I met some great people, including my brother.

Then I got into really exciting technology back then - chat bots. In 2001, I met Lynne Hershman, who had just done a movie called Teknolust. She came to me and said, "Hey, can you put her on a Palm Pilot?" So we did that. It was actually an exhibition in San Francisco Museum of Modern Art for a year. People could beam the chat bot into their Palm Pilot using infrared light.

Then I moved to London after the dot-com crash, worked on video games, mobile phone games, and many other things. At thirty-six, I decided to take a break from the video game industry and start a fintech company working with hedge funds and trading platforms.

At that point in 2008, when this financial crisis happened, frankly if I had still been in the video game industry, I probably wouldn't have started the Citizen Foundation. But being in the financial world gave me the opportunity to make money so I could start the foundation myself. Also, this idea - if I can use technology and AI to help rich people make more money, why can't I also use technology to make society better and use technology for good?

Since then it's been the Citizen Foundation for seventeen years. We developed Your Priorities very early on. We also started a collective intelligence AI platform three years ago and started to work with Northeastern University's Center for Social Impact. We've been using AI and large language models since 2019, so I've been living in the AI world for a long time. The capabilities of what AI now has have been growing really fast the past five years, so we've been slowly transforming all our platforms to be AI-first.

Alessandro Oppo: I read that you were also started a pirate radio?

Robert Bjarnason: Yeah, that's because my father's a ham radio operator. When I was seven or eight, I convinced my father to let me use his FM transmitter with my cassette player and a little microphone. I had my cassettes - I thought I was doing like a favor with a five hundred meter or one kilometer radius where all my friends were listening in their cars. I was playing awesome music of the time. This was back in 1979 or something, a long time ago.

Alessandro Oppo: I was thinking - 2008 was a significant year because of the crisis, but Bitcoin was born in 2008. There is a lot of speculation behind Bitcoin, but also some ideology. What are your thoughts about Bitcoin or blockchain?

Robert Bjarnason: To be honest, I've never really got into it. I've stayed away from that. Every time people have asked me to support liquid democracy or whatever using blockchain, I've stayed away from that.

I like the ideology - this early internet, cyberspace stuff where we can actually create other power structures that are more fair. I like that, but it hasn't really happened like that. It hasn't really been of any use for at least what I'm doing.

I mean, I think it's a good idea, but I just haven't seen any applications that have convinced me. Because even though there's this decentralization thing - once again, it's proven that yes, in theory you can have it be untraceable and decentralized, but then humans are humans and they're going to use exchanges all over the place. There are ways of looking at and knowing what's going on. Even criminals, I guess, but it's not really fulfilling the original promise - or not the promise, but the original things people projected on it.

Alessandro Oppo: Humans are humans - I think that's a key phrase. The humans that are using Your Priorities - are they using it for which kind of uses? Can you share maybe some use cases?

Robert Bjarnason: So basically, most of the use is when there is a project going on. For example, right now there's a consultation with the Scottish Parliament about ADHD - Attention Deficit Hyperactivity Disorder. They started like a couple of weeks ago, already got like seven hundred responses. It's anonymous. Thousands of people have come to learn about it - they advertise it somewhere else and people come and participate.

In a city context, there's collecting feedback on a new traffic policy, collecting ideas. We have this feature called deliberation - a debate solution where we really manage to encourage people to find the best arguments.

Back in 2009 when we did our first pilot with the Shadow Parliament in Iceland, we used a regular commenting system like on YouTube or Facebook - just like a thread of comments. In the first evening of the platform about some policy, we had these people in a horrible personal argument on the first evening. We lost the platform users - they were going at each other personally. I was looking at it like, "Oh, this is not going to work."

But what we came up with - we got inspiration from parliamentary debates and other university debate structures. Also my experience in the video game industry is that when you are building a user experience like a website or an app, one secret is that you are really in control of the experience. When you have a computer game, the player thinks they're in control, and they are to an extent, but it's mainly the game developer who controls exactly what the experience is going to be. The same applies to any digital platform.

So I had this idea - lots of people play video games and think about how to run the game. Could it be an idea to mix video games with participation and debates, where people are actually debating about the problems of our society?

Alessandro Oppo: Could it be an idea to mix video games with participation and debates?

Robert Bjarnason: Absolutely. I think gamification - when you take small game elements - but we also have this idea for an actual game. Instead of just asking for comments, by having people help us find the best arguments in a project, we get people putting both pros and cons, where the same person writes both. There's almost never any arguments just by changing into this new structure of design. Also, you can't comment directly on somebody's point - you have to do a counterpoint. That's worked really well for our partners like the Scottish Parliament - they couldn't deal with it if it was just arguments all the time.

But gamification can go a long way. We actually have an idea we call the Community Challenge. We actually applied to Google.org, though we couldn't get that through. But basically it's an idea where governments or civil society can put forward challenges in the city or in a country, and then a team of human players with AI agents work together to solve the problem. They take different roles - it's like a role-playing game. There's a journalist role, an investigator role - different roles.

It's set up like a game, but it actually has real issues behind it. What's going to enable it to work is the AI agents. The agents are going to be able to keep things from going wrong, they're going to be able to do the complicated things, do lots of research, process information, and do all the fact-checking.

That's actually an idea we have on the table, but it would cost like a million dollars or something to develop, so one day maybe.

Alessandro Oppo: I was thinking related to institutions - a platform like the one you are describing, how easy or hard is it to explain to an institution, to a politician, this innovative platform?

Robert Bjarnason: I don't think it is hard to explain in terms of how it works. It's very simple to use. For example, in Iceland we have forty thousand registered users in Reykjavik, but totally ninety thousand registered users in the whole country because they have fifty-plus pilots using it. So like one-third of the country knows the platform, including the politicians.

Many people have used it. That's a bit of a special situation here in Iceland in terms of how many people have used this open source platform specifically. But it's really about does it serve a need? That's what we need to convince the politicians about.

But it's not enough to convince the politicians because you also have to set it up so the layers of bureaucracy that need to respond to it are going to work there. Any policy decision involves a very complicated interplay between different levels - on the city level but also on the parliamentary level in terms of lawyers, regulations, and different levels of bureaucracy. So you need to have a practical plan that there's going to make the participation work on the bureaucratic level.

One of the things that often kills most participation processes is budget - budget money. That's the reason why we started very early on with Your Priorities and our policy is that if you don't have money, you don't need to pay for it. It's a volunteer thing since we're a nonprofit. We've done many of the most successful Your Priorities projects where they didn't pay anything. We got around that "Oh no, we have to find money for this" - well, it's free.

That's one thing. For most modern politicians - smart people working in municipalities and governments - most of them already think it's a good idea to get more information to make better decisions. So it's usually not a lot of convincing to do or explanation of how it works, but it's just everything else - all the complications that result in participation not happening.

Alessandro Oppo: I'd like to share something about your organization in relation to your team - what kind of skills, and maybe how was it created? Where are you searching for people?

Robert Bjarnason: It started basically - it's always been like a small group of people. It's mostly been like five or something in different roles. For seven years, I took no salaries from the foundation. We've gotten grants from the EU and from all over the place, but it's always been like a passion project.

So several people have come and gone. Recently with me, we've had Joshua and my partner Robert, plus my son Alex and some others. It's been like a few people around - mostly friends and family type of thing. When you don't have dependable salaries to offer, that's usually how you get people to help, but that is also great in many cases.

It's obviously been a lot on the technology side - most of the technological development. We partner with, for example, Northeastern University Center for Social Impact. Beth Noveck has been leading those research processes. We partnered with her from 2013 to 2019. Also the Scottish Parliament and many others.

We are just bringing the technology component. That's our domain or scope, if you like. We don't offer the service of going somewhere and setting up a democracy project for somebody. We have partners for that from Democratic Society from Brussels and in-touch.org, European Citizen Action Service. That's how we've been - lean, family, friends oriented, but with a lot of partners that help us scale up the work.

Alessandro Oppo: I was curious about the projects that you are working on or that you worked on. If you would like to share something?

Robert Bjarnason: So really exciting project - I'm actually going to be presenting in San Francisco in August at the opening with this as part of the Future of Work program of the GitLab Foundation. GitLab is a software development security company, and they have this foundation. We were one of the projects that got funded last year.

This is with Northeastern University where we're using AI and agents to help states identify jobs that are requiring university degrees but really should be skill-based. This is the idea of the problem being that there are jobs that go unfilled often, and maybe some of them have too high education requirements. So we use AI to help states optimize that process.

This has been a really interesting project. I think that's where you can see in many places this idea of using AI for good. If nobody is funding it and it's not profitable enough to have that focus, it's really important for society to keep on funding nonprofit ideas.

Another interesting project from earlier in the year - we worked with Northeastern University and the state of New Jersey where they are basically doing research on the job market - what effect will AI generally have on the job market in New Jersey? They have this AI task force in New Jersey.

What we helped them with is that we first identified the root causes of the problems - what are the root causes of disruption of AI in New Jersey? We did this using our open source tools to do large-scale web research looking at thousands of documents, plus interviews, to look for everything recent people have been talking about regarding the effect that AI will have on jobs.

Then we had a curated list of two hundred AI impacts or potential root causes of problems that people might face. We put this into voting by the public using a tool called All Our Ideas, which we also developed, where we asked people what's going to be the greatest impact of the next five years of generative AI. Then people vote in a pairwise manner, and we got seventy thousand votes.

Some of the top items were things like privacy issues - people are concerned about privacy rights. Then we did a second stage where we actually used AI also to help us come up with solutions to some of the problems. It's a very classic example of many of the projects we've done in the past where we are mixing together AI research, then engagement of people at different stages, and then we also use AI to help us find solutions.

Alessandro Oppo: Other projects you're excited about - could be a project you worked on or maybe a project of someone else that is interesting?

Robert Bjarnason: I think in general there are many people looking at how to use AI for good. I think that sort of in general, the category of projects is that the promise of AI to do good for society is at least as powerful as the potential bad things AI can do for society.

The thing is that if we don't have the imagination to actually think about projects to do good, then the bad things will happen. I think we've missed this with social media. I was talking about this for years - how social media, which we saw quite early on, had the potential for some bad influences on society, on mental health and stuff. It just went on and on, and this idea came that we can regulate social media with rules.

Because we can't just stop the bad things happening with this technology, we've been promoting and are generally promoting that yes, obviously you want to stop the bad things, but you also have to do good things. You have to have positive ideas and things. You can't just try to stop the bad because you're going to fail quite badly. You also have to do positive things.

I think that's so important when it comes to democracy projects - we have this amazing opportunity now with AI. Just like with programming, I can do things now literally ten times faster programming-wise with agents than I could two years ago. Ten times faster - that is like ten of me two years ago. Think about that. If people are having interesting ideas for projects to do something for democracy, the barrier to entry is so much lower.

So maybe a bit of a long answer to your question, but the projects that I'm excited about are the new projects that listeners are going to be working on for this purpose.

Alessandro Oppo: To reply in some way to your question, I have this feeling that software now is not so important - I mean, it is important, but before you had to have a team of maybe ten developers, and now just one person could be enough. I can feel that in some years, maybe also my mother who is not a developer, maybe she will be able to talk to ChatGPT or Google something and create an application. So what is important now in the civic tech field? After it's not so important to build a platform that is well designed - I would say we humans should still be the designers, or could AI also have this role?

Robert Bjarnason: What AI does is that it empowers people with agency that need to want to do something. So if you have this need to want to do something good, now it's easier than ever in terms of just using AI to do things you want and empower you and a small team, one person, two people.

You're absolutely right - for sure. We are the same - we're a bit behind the curve on programming and stuff, but it's getting there soon. And that's also like with software - software is like a means to do something. The software is just there to achieve some sort of goal.

But I think we're already seeing that in the job market where with programmers in general, there aren't as many openings, but there was a lot of growth during COVID and a lot of people put everything into digital solutions. Now big companies have been laying people off like Microsoft let nine thousand people go.

There's this sort of interesting situation where people are getting let go, and you know, "Start your own startup" and stuff like that. That's great if they have that drive. But I think there's also a huge scope for people to do things in the civic tech world and civil society that aren't going to make big dollars but are going to make things better for people, make democracy better, or whatever. These can have huge rewards if you're going to do something like that and you're successful at it. Even if you're not successful, it is still highly rewarding.

Obviously pursuing startup money and everything - that's absolutely fine, I've been there myself. But I'm just saying with this age of AI and lower barriers to entry, I'm hoping also that how resources and money and everything is distributed is going to change.

But what matters is agency and creating real value, and there are so many things you can do with creating real value in civic tech and with democracy that could really make people's life better. Looking for meaning, looking for purpose - that's definitely for some people going to be a great path forward.

Alessandro Oppo: Do you have anything you're working on that you have some problem to fix, so you will need some help?

Robert Bjarnason: All the time. There are both software problems and other problems - all sorts of issues. One of the great things about this part of working in the nonprofit civic tech world is that there are often conferences, meetings, common projects, and so on where you get a sense of understanding the human aspects of all of it. If you're just with your head in the code, you also need to understand the human aspects of it.

But now with AI today, if I have a complicated programming problem, I have several options. I sometimes put the same question into ChatGPT, Gemini, and Claude - I call it like my AI council. Then I have a typical question. Also, ChatGPT's OpenAI's programming assistant - when I get bug reports or want a new feature, I used to put it in a task management system, but now most of them go straight into ChatGPT's agent. Instead of putting it on a list for later, I just put it straight in and say "fix this," and then it just does the feature or fixes the bug for me.

Yeah, we hope that continues.

Alessandro Oppo: You were talking before about the community of civic tech, people volunteering or working in the civic tech field. Do you have any message for them? Because sometimes there are people that have another side problem - finding funding for their project. They really believe in an idea that can make the world a better place, let's say.

Robert Bjarnason: My message is that there's been this transformation, especially when it comes to civic tech specifically - we have already been transformed in terms of how much capacity increase we can have on a small team using AI.

I also think on the funding side, in the next few years when people realize - when we see so much of our work... I consider myself a programmer. I've been a programmer since 1981. I have really worked hard at it, practiced a lot, become really good in my field. I've put in hundreds of thousands of hours of practicing. This is what humans do - we practice.

But now I have access to AI that is almost as smart as me. The next generation will be much smarter. That's sort of - I think my worth and value is a bit difficult to think about just because these neural networks are in a way taking my job.

But I think the value and what you can do in society is going to change totally when we see knowledge-type of work being so highly automated. There are going to be new things we can't imagine yet.

I think funding for making society better - I think that's definitely one of the positive potential outcomes we can get out of this. We basically have two scenarios: we have a scenario where we have thirty percent unemployment and corporations control everything, but we also potentially have a positive scenario where we have like a three-day work week and plenty of resources for doing projects connected to society because there's so much value created.

If you do anything that actually helps facilitate better things, that has value. There's so many things in society that are under-invested in because of how capitalism works in the world, mostly in the US. This is much better in traditional civil society being well-funded in Europe, not so much in Iceland.

But at least that's my belief - that we're going to start to really look at what's going on around this, what is important, and when we see things that used to be very expensive - like programming - plummet to almost zero, I think we're going to see more clearly what is important. I think there are a lot of opportunities for people to tackle civic tech, democracy, and so on.

Alessandro Oppo: Thank you, Robert. Would you like to say something more?

Robert Bjarnason: For me it was great to thank you for having me on here, and I'm sure we'll speak again.

Alessandro Oppo: Absolutely. Thank you.