The the pronunciation is good. Thank you for having me on your podcast.

Yes. SwarmCheck, is a technological solution of Optimum Pareto Foundation with a mission of improving rational public discourse. And technology, in our opinion, is most effective way to do it. So we created SwarmCheck. It's mainly argument mapping software, for collective decision making, Delphi processes, and, improving collective intelligence. So, it can be applied in variety of, uses and and topics because argumentation, as its core mechanics is very universal. Right? We we see argumentation everywhere. So with correct technology that supports collective argumentation, we can improve not only the quality of the discussion, deliberation, and collecting data in the process, but, improve decision making as well by having more perspectives, more, critical voices, and better quality of the decision as a whole. So, yeah, that that's that's it. Probably, you have more more questions. I will elaborate.

Sure. We have many use cases. We, completed over 40 projects, with, this software. So I will give you some some range of of topics that we tackled in the past. So maybe I'll start with the the the current one we are, excited about. There is a method of, achieving numerical results for your questions called the Delphi method. So it's the expert deliberation anonymous, which is, iterative process of producing numbers, estimations about some future event or some risks, some prioritization. And with anonymous argumentation about those initial estimates, we can, achieve consensus of the group and be quite sure that, our final results, achieved by this process is, good for decision making or or, strategic decision making or or, risk planning or, even scientific publications. So for example, pharmaceutical company approach us to conduct Delphi process with, clinicians to estimate, the risks of illness contracted with some virus. After the the the treatment, there is a period in which the the the their solution is is, given to to the patients. But after the this process can be some can can appear some some additional, symptoms. So for people with certain, criteria, we can estimate how risky is for them to basically stop the treatment. Right? And with those, data, we can show, okay, experts collectively have consensus that the treatment, needs extension for certain criteria with this amount of risk to to to to to stop them. And with this data, we can have scientific publication that, can convince, for example, decision makers in in government to finance the the the extension of of the treatment. Another case, more on the side of policy making, is, basically creating collectively, new strategies, new new policies in, local, governments. And so, one case was the renewal of, ten year educational policy in the city of Poznan. So the the challenge was, okay. What do we want to achieve in in next ten years? What are the most important issues? Because education is big thing, and there are many people who are interested in in in the outcomes. There are students, of course. There are teachers. There are school directors, public officials, NGO workers. There are academics who know a lot about the educations and and effectiveness of of certain policy interventions and so on. So there there are people with different point of view on the same topic, and with different interests, and we have to somehow, make them, come up with the the the solution for for those policies. So we use SwarmCheck to to map all the augmentation about the, key proposals to basically come up with, solutions that are anonymously produced and represent the collective intelligence of the group. Maybe here is the good time to explain what is argument map for people who, hear about this for the first time. Absolutely. So, basically, if if you, if you have a book, right, you have line of, lines of of text. You you start, reading it from the, well, usually, from from the left, top corner and and go to the right and and go to the bottom and so on. So it gives you only linear progression of some narrative. Right? You you you basically have a story that have beginning, middle, and end. But, we can extract all the useful information, so the specific claims, and map their relation to each other. Basically, arguments, are those those claims that that relate to each other by giving each other support, or they contract each other or give or show disagreement. So when we extract those claims, we we can generate a graph, a a a map of individual claims and their relationships. So right now, we, don't have, this linear text. We have something that that is a network, network of reasoning, of argumentation, of ideas, and we can clearly show how they relate to each other. This gives, gives us a lot more information about the reasoning and, show the subject we are discussing more objectively because there is no story about this. You can basically travel as you like, through the connections on on the graph and read what are the, agreements, what are the voices of support, how people explain their their support, what are the justifications. And on the other hand, the the are, disagreements, contradictions, critical voices, and those in turn can have their own supports and disagreements as well. So, basically, you build block by block this, graph of collected reasoning. And when you do this anonymously with software that that can integrate, different, points of view, you you have outcome that is not controlled by anyone, but only represent the collective knowledge, of the discussion gathered. And then you can, conduct additional analysis of this of this graph. So, for example, you can analyze how, certain claims are network, how many supports are there, if the supports are well sourced, in in so, at the end of the of the process, when we have graph of all arguments and and their relationships to to to each other, we can analyze it from the standpoint of, okay, which claims are more, the have the best supports, what are the sources if we use, for example, scientific data to support some claims, what are the contradictions, disagreements, critical voices, and every level of this graph can have the same things. So, basically, every, when when we start with the first premise, we we we can add to to each, to to to to this, artifact and other, arguments. So let's say, I I I click on the on the premise, every, student should wear school uniform, something like that. And, the the system asks me why one should think so. And I give my, my premise, my my reasoning. Let's say I I think that, uniformity, allows poorer students to to, not feel excluded. And this my my explanation is a claim of of itself. Right? So people can agree with it and give their reasoning. So maybe there is some scientific study that supports my my claim. Or maybe there is some criticism that even though the the uniforms may provide that to to some extent, there are different aspects and and so on. Right? That that undermine the the, the the idea of of feeling well for for those type of students. So so, in this way, we we have graph of all reasoning that is, collected through through the discussion and can be, additionally supported by the reasoning source from, literature, from research, from other sources, even from other discussions because the the claims are reusable in our system is very important aspect that it can connect, discussions that are held in different place, different time. And after we achieve this argument map, we can run analysis on it. So, we can see, okay, what is the line of reasoning that supports, our main claim, main idea? What are the, risks? We can see on the graph, basically, the, how strong are the branches that leads to to some, so some outcome. We can see which claims are, the best networked in the graph. So, it can indicate that this, claim is very important for people because they addressed it a lot. We can see which claim, don't have any support, and those may be just some fringe ideas or just the the group don't have any way to to support it or contradict it or is not very interested in them. We can see that some, initial, line of reasoning was very well supported, but at at certain points is undercut by by very good, counterargument. So, this is something that so we we have, in in the end, something that would be very hard to analyze by a person because all those connections you have to have in your own head. Right? And, you have to remember, okay, how the the the claim argument sentence that that is, spoken at the end of the meeting relates to to something that was said at the beginning. Right? It's it's it's very hard to to have it in mind, even, if you use, something like large language model, artificial intelligence to to analyze this transcript of of the meeting. It will be hard for artificial intelligence to to have those connections in, in its, let's say, attention. So argument maps give very good and reliable and explainable reasoning about the issue that is created by collective intelligence. And this is, why I think it it's, it is one of the best tools to support collective decision making.

Sure. So, the how many people are there? It it depends on, of course, on on the issue. We had, groups as as small as, let's say, four people to discuss something, but as large as, 40 or 80 people. So, of course, it is the the let's say, individual session of the discussion, the the optimal number of people is around, let's say, 10 people. But we have we can have many sessions in parallel or in in sequence because we can when somebody reuses some claim from the past, some some argument from the past, the system will join those graphs and, we we we can have something that is, bigger than any, anything that one group can come up with because it's now the result of the discussion of of two groups. So, yeah, it it depends. It it can be scaled up to basically global civilization if we imagine so. The the the the core insight in this process is that argumentation in in public discourse, on in scientific discourse is not infinite in in every topic. In every topic, we we hear finite number of, arguments. And when we try to, put them onto graph, ontology, we can see that, okay, here are the the same repeating points. We don't have to map them over and over. We can reuse them from the past and see how they were addressed. This is very important to, counteract misconceptions, disinformation, and, basically, mistakes and errors one can make in decision making process. Right? So it's quite important, for us to, allow groups not to repeat mistakes over and over as it is the case in public discourse right now, but we want to, use this collective intelligence to only build on top of previously gathered knowledge. Yeah. So the the the idea is that, it's potentially feasible to to basically map out the whole public discourse, by collecting or important arguments. Important in the sense that they are reasonable in this discussion. We have to exclude spam and something that is not arguments and so on. Right? But we we don't, in any way, intervene in the merit of the, claims because this is what arguments are for. If somebody thinks it's, unjustified, one can give counterargument. Right? And this can allows us to avoid something like censorship or focusing only on something that moderator moderator think is important. So, yeah, it usually it's not needed to have the the global civilization to to to map out the the whole discussion, the the most important argument. But the group of, let's say, 10 to to 40 people is usually more than enough to to have, to represent all the viewpoints that are appearing in public discourse.

Yeah. So the the ultimate method is something that, exists from the fifties. It was developed by Rand Corporation as a mean to, enhance decision making in, very important, strategic decision making. And, it was improved and developed, through the years. There are many versions of of the Delphi process. The main problem that we help to to to solve of the Delphi method is the time consuming part of the of the process. So, basically, every expert that is invited must be must remain remain anonymous, and at the other side, have to present, their estimates in every round for for each, question and have to present reasoning remaining anonymous. So classically, you know, the the the surveys were sent to the expert. The the they were collected back. Somebody made calculations. Okay. This is the the mean of your estimate. This is the standard deviation. So the measure of lack of consensus if the the standard deviation is higher or the, how much consensus is achieved if it goes to, to zero. So, basically, how how how much variety there is in in the estimates of experts. And then in order to to get close to zero, so to achieve conscious, consensus, experts need to present the reasoning that they have to present. Okay. I'm trying to convince you that my point of view is, correct. So I I let's say my estimate was lower than the mean of the group. So, I I can give some counterarguments for the position that should be high. Right? And then we can go into details. So, basically, one expert present the line of reasoning, another expert can address it, can present their own arguments. And, it in, collective argument mapping, it's it's very easy. It's anonymous already. Everyone can see the same map. We are on the same page. We speak the same language. It's, everything can be done quickly and, with preservation of best practices to to to achieve collective intelligence. But when we don't have the tool like this, it's easy for moderator or or for person who collects the information to, to present it in a way that that will break the anonymity. Let's say, experts write short essays. Okay. My reasoning was blah blah blah. And somebody can okay. I I I know that this style of writing, so so this is this expert, the the anonymity is, is broken. Or maybe, some experts are very to the point and write only one paragraph and other experts will write long essay. Right? How we can compare those? And, of course, the the problem of plain text, with the linear narrative still can, misguide other experts. But with argument maps, we only have reasoning, and we can objectively see, okay, all the arguments that were deemed important by experts are here. We can look at them and then, rethink our initial, estimate and give another one, which will lead us to to consensus. And from our experiments, our projects, we can see that, the standard deviation that that is measured on each round, the the measure of consensus, is shifting significantly after the the round of argument mapping, to to the consensus. So I think it's improvement on on that part. And, of course, we don't need to, create summarization of of, statistical analysis for for the estimates. We don't need to, make summaries of of, every essay that expertise everything is real time showing on on on software. So, it can be done really quickly. So the process that could take months can be done in a in one day or one sitting, basically. It depends, of course, how many questions are there and so on. But, this is the the the improvements are in in time and, quality. Because the the important thing at the end of the process is we don't only have results with summary, statistical summary, and, what what is the mean that the the averages, on the on the after the last round and and the measure of consensus, the standard deviation, we additionally have the reasoning that led experts to this conclusion. So, there is additional safeguard that this this this, estimation is correct. Or even if some future event will, give us new information, we can see, okay, if this, new information can be added to the map, we can see how can it affect, initial, sorry, final conclusion. So, it's easy to to to use this knowledge for the future, goals.

Yes. Exactly. And, we can go to the nuances. So that that's the very important aspect from the standpoint of reducing polarization because especially in political, topics, people tend to align with some, ideological lines, party lines, and so on. But when we go into the detail to the nuances, people are much less ideological. And, they can, from my experience, discuss very reasonably, particular claims. So maybe initially, there are some disagreements, but when we go to those, premises of the premises, we can see that people are quite rational. And when they see, viewpoints of others, the the perspectives they didn't come up, themselves, they're actually learning to to think more critically about the issue. And with those the this additional knowledge, they're in real time adjusting, their their own perspective. So, yeah, I I think it's quite important that this process is not only interesting and and good in terms of producing good data, but it's good for the users themselves. They they are educate being being educated on the topic in real time by by engaging with argument mapping. And, not only that, the the critical thinking ability of people who engage in argument mapping increases, radically. So, there were some studies that measured the the standard deviation of improvement of critical thinking, of students who who did semester of, on argument mapping in in academic setting. And in comparison to first year of college in general, the the critical thinking ability of, students who used argument mapping course, was three times better than, the than the baseline of the first year of college. What's more import more more interesting, people who attend, critical thinking and philosophy mixed course. Yes. In in critical thinking, tests. Up basically, the the standardized test of of critical thinking. And in, there there were three situations, that that were, studied in different studies. But when we compare the results, the the first year of college, is, three times worse in terms of increased stability of critical thinking than course of argument mapping. And, critical thinking course with mixed with philosophy, gives worse results than argument mapping, about two times worse. So, from the techniques I know that increase critical thinking ability, with some scientific study backings, I I don't know the better tool for increasing, individual ability, of critical thinking. Yeah.

Sure. So I I was born, in Poland. I still live here. I was born in in Strahovice. I I moved to Krakow, to study law at the Ibelonian University. But instead of being a lawyer or a judge or something like that, I I was mostly interested in argumentation, about the, legal theory, about the philosophical aspects of the law. Because maybe what's my characteristic is I'm quite attracted to to philosophy, and I think it's very useful, for solving, everyday problems, counterintuitively. Because when people people think about philosophy, it's something that is detached from reality. And, it's not entirely true. Maybe some academic philosophers are quite detached from reality and and solving real life problems. But good thinking about problem solving is something that always, helps you to to achieve better results. And, from the perspective of of society and, basically talking to each other, having to to to live together, we are creating laws and rules to to to how how to behave, how to conduct ourselves. And, those rules can be better or worse. Right? They can improve, some type of value that we care about, or, can make things worse. Sometimes they are trade off. For example, on one hand, we we have freedom. On the other hand, we we have security. Oftentimes, it's one of the cost of of the other. But, let's ask the the about the conditions in, let's say, political dissidents in in Russian prison. Right? Russia is is a neighbor of of Poland, so we we have a lot of bad history with each other. So so we, we we the current geopolitical situation try to, you know, distance ourselves for for from their political regime. But that is precisely why because in political, the political dissidents in Russian prison, they they don't have either freedom or, safety. Right? So it showcase the the the the example can can showcase, to to us that there are some way to to optimize society to basically have the most freedom, the most security at the same time. And then, of course, there are some trade offs after afterwards. But at least, we can search for those optimal solutions that that nobody is is worse off, and the the the society as a whole is is better off. This is, optimum Pareto principle and, well, not on the principle, the optimum Pareto analysis of the the the, social values that that, gave the the name of the Optimum Pareto Foundation, which me and my colleagues in in the, Jagiellonian University, founded basically at our at the end of our studies, because the the the idea that we can use deliberation, we can use, good design principles about creating policies, creating laws, could be applied to decision making, to have better, policies on on municipal level, to on national level, maybe on international level. And it it it's something that that can, show us that that the abstract reasoning about the argumentation theory, about the philosophy, and about the values can have real impact on a decision that that affects us on on everyday lives. Because, if society makes good decisions, everyone is better off. So, yeah, that that that was the motivation to to start the the Optimum Pareto Foundation. The the common interest, let's say, about the, dialogue and deliberation and alternative means of of solving conflicts. And and on top of that, from more than ten years ago, I I started to be interested in artificial intelligence. And, this technology in my mind is something that is one of the greatest, potential for for humanity and one of the greatest risks. Because, basically, how we are different than other species on on this planet is by our mostly collective intelligence. And this is something that we are adding to to our collective intelligence of of humanity. This technology that improves and, let's say, accelerates some aspects of intelligence. And, when I see that the current most powerful systems are black boxes, So we don't really know how they work. We even the the, developers that that created the tools cannot explain how the certain decision was made. It's something that that worries me. Right? Because, how we know if the the solution is correct, if it's if if it's not something that, leads to some sort of deception, what type of values are are maximized? Are we agreeing with those values? How we can assure that the individual inputs, our values, the things we care about is something that artificial intelligence that that conducts those calculation in the black box cares about us as well. Right? We are we we cannot be sure. So this is something that that is a worry of mine. But on the other hand, when we use artificial intelligence to to reduce the errors that we make, to to help us select, information that is, well sourced. And it is not disinformation, created by internal, political, actors or external, like, in the case of of Russia, one of the biggest, perpetrator of this information, currently. Yeah. It's just too much for one person individually to have good information diet and and check everything and, you know, follow all the fact checkers and then the fact check fact checkers. This is this is too much. But with the collective intelligence and the aid of artificial intelligence too, not to, to replace our thinking, but to enhance our thinking and, enhance our collective intelligence. This, this could be one of the the greatest benefit for for for humanity, in my opinion. But the important thing is the the the replacement part. When artificial intelligence replaces our thinking, what's left for us? Right? The the decisions are just flowing, over our head, flying over over our head. We we are not subjects in the, deliberation in in public discourse. We basically are objects of, manipulation. We are just data points that are used to optimize some, you know, financial results of some company. And, yeah, the the this situation when we, as humans, lose our, personhood in decision making is something that is, very worrying for me in emergence of of artificial intelligence and and rapid growth of it. But this is something that we expected for a long time, so we prepared swarm check and argument mapping as a tool not to be, in competition with artificial intelligence, but, the tool that preserves good deliberation, preserves the ability to to voice our values, our reasoning, our point of view, and b, include that in decision making that can be incorporated in artificial intelligence as well. So, yeah, the the question was about my experience, and I quickly moved to ideas. So so, maybe to to summarize it, I'm a person that that is moved by the ideas. And then I when I see something that is important, I try to act on it and, do something useful. So maybe yeah. That's, that's something that that Don't worry. It's to do my work.

Yeah. A couple of things. So I I was interested in, sociology of the law as well. So, basically, how the law can shape personhood. So, basically, the law says the president can do this, this, and this, and when you are a prisoner, you can do this only. Right? What what so the the the idea that law is a instrument that that gives power, I I think it's attractive just for people in general. The the the power is attractive for for people. But, when I observe the process of democratic elections, it's very strange to me that people are not questioned that much the the whole process that we develop because most of people are not satisfied with the results. Of course, currently, there is some process that that hijack the the collective discourse and and we see on the Internet that so many people support somebody and so on so on. But people are generally, more rational about their, yeah, I mean, in general, about the overview of of democracy. They don't think it's the best system to, select most, optimal person to to hold the position to to, create the best possible laws. Most people think about democracy as the the least, good system, that actually, you know, produce some amount of freedom, stability, and we have to deal with those, stupid politicians as a, some sort of unfortunate externalities. Right? But it strikes me as something something weird. Right? That we we can have better decision makings, making systems locally when we, for example, discuss something on a seminar about the law. So, the the the discussions were very smart, thoughtful, empathetic, and, they incorporated many points of view. But if the same topic is discussed, you know, in on parliament, it it it started to to to resemble, you know, a circus, something that nobody watches as intellectual, activity, but maybe for entertainment to to see what how stupid one politician is or what outrageous other politician said. And we are actually dissatisfied with with this, type of deliberation that is that is seen, on in in parliaments. Right? So the the question was for me, okay. How how we can, take the thoughtful deliberation and actually good knowledge about the from from people who, study certain topics and move it a little bit, towards, and extract some some reasoning, some arguments, and and move it towards, decision making in in parliaments that that where the decision about our everyday lives are are made. Right? Even on on local level, how we can access the the voices of the citizens to, govern our city better. And the the the problems are very human that we have even politicians are are humans. They have, cognitive capacity of of of human. It's very normal thing. So so one cannot put everything, on in their head. So, this is this is why I I started to go into artificial intelligence field because, in artificial intelligence, there are solutions for for for those problems. Well, for the problems of having a lot big amount of data and and making decisions about that. But the the there are some additional issues and, basically, solving issues about the issues about the issues, led me to to the combination of, artificial intelligence and and and law and reasoning about, norms, about values, about policies. And from this, I just started to, you know, going to to conferences, going to libraries. When you have problem to solve, it's much easy easier to educate yourselves on on those topics because you you care about solving it. So, in my line of study, I I try to follow those interests more than, you know, the classical, let's say, syllabus of of of the of some courses. And, yeah, I I think that this approach can lead to more interdisciplin interdisciplinary view of the of the problem. And then when you see the problem, you know, in in more places, in different areas, then you can see, okay. I nobody's really solving it because, everybody is in their own silos, intellectually, let's say. They're not combining necessary knowledge to to to solve this problem. And you can see it, you know, in in everyday discussion when, let's say you you start just polite discussion with somebody and and all of a sudden you are shouting at each other. So from argumentation, it lead to argument in a sense of of, you know, conflict. It that those situations strike me as, as something that that usually can be prevented can be prevented with better better dialogues, but but better, phrasing of of some words, better listening. But it's actually hard when so many arguments, so many claims, fly around in the discussion. We don't have cognitive capacity to to store them all and see all the connections. So we, use our emotions to to move our, ourselves in the discussions, in the dialogues. And I think it's unnecessary. Of course, it's it's good when emotions animate you to to do something good. But, especially when you watch, you know, Internet and and how the public discourse devolve over there to for to to shouting matches and algorithmically enhanced, outrage. Right? It it's something that's, that that that that that makes you sad, that makes you, stressed. And, seeing all those instances of basically deliberation and and dialogue, that goes bad, one can, you know, enhance their own opinion that, okay, if we just make slight improvements about the way we communicate with each other, we can have better policies, better decisions, less mistakes, and we can all be better off. Right? So, yeah, all those aspects, led me to, to this path more and more. And, yeah, there is, of course, drawback of, you know, argumentation being this all encompassing tool that we use every day. So, people don't really well, some people just get it from from the get go. Okay. Argumentation mapping is useful because you can see it. It it can function as cognitive scaffolding for you. You can see bigger picture. You can you can critically analyze the topic. You can use artificial intelligence to to to enhance, the the analysis of the discussion and decision making. It's all great. But for some people, it's like water for, for fish in the sea. Right? It it's just all over there, so you don't really think about improving it. And it's like, it's like air. It's like something that is just there. It's like a state of nature, but, of course, it's not. The language is something that, was constructed by our cultures for a long time, and and the the the way we talk to each other is very is is is changing still, but but change a lot through the through the history. And when we started to, be more cautious about the the the language, the, present the the reasoning as something that that can be examined, this, led to beginning of of philosophy. Right? We we started to exchange ideas. We started to think critically about the world. We start to think about big things like, what what is real, what is not real, how, why are we conscious, what what is the purpose of life, what is moral, what is not moral, how we can best, arrange society for the benefit of all, can we know the truth or not? Those are very fundamental questions that people living, you know, 3,000 ago started to to think about that, basically helped our civilization grow, exponentially from that time because the philosophy led to to to science. Science led to, technology and our modern world that that is based on it. And it it when we go back to the beginnings of of philosophy, of critical thinking about, you know, the the, Plato dialogues, the the the world is vastly different, but the problems at the the bottom, the core problems remains the same. But when when we look at how the the knowledge progresses, we can see that some ideas that people believe in the past were just wrong. People didn't have good, reasoning behind it, don't didn't have good arguments. And, step by step, collectively, we we developed, science and academic institutions and so on that, gives us much better understanding of of the world we live in. And I think the same goes with, ethics and and moral philosophy and and sociology and and and law. But it's very hard currently to be, knowledgeable about all of those topics. Right? But to certain extent, we we need to be knowledgeable about all of it, to to make good decisions, to, not not to lead ourselves, you know, off the cliff, because the the it's quite clear that the changes in the world are are very fast. Right? But when we don't have good, sense making tools to assess the changes, to to basically navigate those complex problems with, critical thinking and clarity and, interdisciplinary knowledge, it it's inevitable that we will make huge mistakes that are very costly and, we we we wouldn't like that. Yeah. So if, if we can build on top of the knowledge, created by generations, as Isaac Newton said that if he saw further, it's because he stood on the shoulders of of giants. I think that this is exactly the the thing that we want to capture using this technology. That public discourse can be something, that is this, shoulders of of giants. That, a lot of things and thoughts and private conversation and public conversations, are very valuable in terms of the their content, the the the knowledge, the the reasoning, the arguments that we, that we use. But the, unfortunate part is that collectively as a society, we have amnesia. We just talk the same things over and over. We have the same arguments, the same conflicts over and over. And, I think right now as a civilization, we are stuck. We are stuck on this crazy loop of, the same arguments over and over that are very chaotic, and, it gives us very bad energy to to conduct ourselves. So the the the vision that attracts me is that using, the knowledge of of people, using their personal experiences, their, maybe professional experiences, maybe their academic, expertise, in a way that contributes to our collective knowledge as a process in which we are still remain as citizens, as people who are engaged in public debate, not be replaced by by politicians, not be replaced by social media algorithms, not be replaced by large language model or any kind of artificial intelligence, but as a as a part of, you know, community that can add, something useful for the public discourse. But we we need something that will, shepherd the public discourse to, remember those arguments, to to use them in the future when the same topic arises again, to to to move past the the shallow conversations and shallow, conflicts to actually, maybe resolve some of them, maybe to to, map out the better understanding of some important decisions, maybe to end conflicts on on bigger scale on, in terms of economics, in terms of geopolitics, in terms of technology development. Because, otherwise, we are creating society that can lead into some, brave new world territory if we are lucky. But, if we are unlucky, some Orwellian territory. So, yeah, this is basically our our future we are talking about, and we are just don't have any means of collectively navigate through the spaces of of possibilities about our future. Right? So, yeah, that's maybe the end of my long speech.

And the the this this problem is very big, in terms of, large discussions. But it's very easy when a large discussion is, split into, smallest possible pieces. Basically, claims and arguments. When we analyze one claim, it's much easier to to check if the phrasing, the definitions are understood by, the parties engaged in discussion, if they are if they use the same language or maybe there is some equivocation. So the same terms used as something different. So, we we started to build our mechanics exactly with that, point of view that, when when we, break out the big problem into smaller pieces, it enhances our ability to to to solve them. And it's there are finite ways to, to and final techniques to use to to, go from problem, from miscommunication to better communication. The the there are, argument fallacies that can that we can check. There are certain aspects of the phrasing that can make some something more more clear. There are some ways to to to paraphrase the the same sentence, that that would use language that is understood by others. And and sometimes just focusing on on select key issues is good enough. One don't need to know all and to analyze everything. Right? If we are certain that the process of giving one argument to the the map, to the discourse is is okay, and improving on this argument and making, going from from being in, in error to to to be less wrong, is something that is scalable. And this, can lead to to having a collective discussion on the scale that is currently, not possible without the technology.

Yeah. So we started in, well, ten years well, eleven years ago in May, with the Open America Foundation. We started as a group of friends from university. But later on, when we decided that, okay, the SwarmCheck idea is the the technology and educational projects that are connected to argument mapping is something that we we want to pursue. We started to build bigger teams. We started to hire philosophers. We started to hire, developers, designers. And, basically, we nearly, grew exponentially for several several years. Until, two years ago, we got 40 people, working in different projects, for in in policy making, in using, swarm check, in r and d project to combat disinformation, so on. But, unfortunately, the the the grow was halted by conflict with the public institution, that funded one of the our recent project, and we had to reduce our team. So, currently, we are only in we we have a team of six people. We we still maintain development, let's say, this service aspect of of our, our entrepreneurship. We, conduct services for for the municipalities and for the pharmaceutical companies and and for everybody who wants to, improve their decision making, can conduct their studies and so on. But, yeah, we we are right now in recovery mode, but because, yeah, we we we took a gamble to rely on public institution. It should be something that, you know, every citizen, should, well, could rely on in in normal circumstances, but, sometimes public institutions are, faulty and there there are many corrupt corruption scandals regarding this particular institution. So our project, took a hit, concerning this the situation. But and we we were left in very difficult situation because we had to rebuild our software that were, you know, in one third of the r and d project. But we managed to to overcome that and more difficulties, through through two years. And, right now, we are in, as I said, six member team. And this time, I think we will, move a little bit slower in terms of, building team, but we still, want, you know, big things in terms of, projects and, outcomes, especially, I think, combination of arguments, mapping technology and, our, ways of building it into expert systems in combination with language models is something that, that is very very promising in in terms of many fields like, of course, safety and ethical development of artificial intelligence. But for the perspective of, let's say, investors and and public institutions, legal tech is something that we look into right now. And, with with our team, we built a POC of of the system that uses argument mapping to have, explainable, legal reasoning about the, basically, any subject. This is something that we are pursuing right now. So we in our team, we, as I said, have developers, but, let's say philosophers slash lawyers people who are interested in in those type of, areas and have expertise in it. And, yeah, that that's the current situation. So I guess, we are looking for for collaborations because, you know, in in in the past, when we had much bigger team, we we could conduct many more projects at the same time. Right now, we are taking things one step at a time, but we're still maintaining good quality of the services, and we still are developing the the product.

Yeah. That that that's a good question. So the state of the software is that we can, fully operate, conducting, Delphi studies, social consultations, deliberative processes, and so on. And to that extent, we basically have everything we need. But when we look how the artificial intelligence is, being developed and the ethical issues concerning the transparency being, leaving humans out of the decision making loop and hallucination problems and, basically, errors that enhance the the, some human very human way of of thinking, taking shortcuts, heuristics, and so on, we can see that our, technology have so much more potential. So on on one hand, we are, sufficiently developed technologically to to conduct, projects, that utilize collective intelligence, like like like those stealthy studies I mentioned. Yeah. The the the platform have this potential of combining artificial intelligence and, collective intelligence that that we would like to, just have more resources to to focus on and develop. Because, while we can provide much value in terms of improving collective intelligence, I think that the the future, is relying on on coming up with the strategies of incorporating collective intelligence into, thinking of artificial intelligence. So this is why we we want to focus on on legal tech, and, we want to focus on stuff like decentralized science, but we we cannot have everything at at the same time. So, if we could have collaborations with people who are interested in legal tech, in decentralized science, and in artificial intelligence in a sense of, explainable AI, ethical AI, and, build building workflows that deal with augmentation. We basically have two, r and d projects that are written there just waiting for to to to be financed about using workflows and agents that help improve collective intelligence as contributors. So, basically, as small scale moderators that, could, suggest sources that could use argument mining for, collecting additional data for a discussion from, let's say, scientific literature or maybe, to analyze something from legal perspectives to, give criticism to your idea, to, check if the, phrasing is, something that is, confusing for others or can be phrased better that can, join discussions that initially maybe were not joined because, the phrasing of some claim, were not similar enough for the system to detect it as the same. There there are many small, aspects of the process of of building collective intelligence that can be improved by using large language models. And people who are interested, especially technically, in those areas, you know, can, contact me, because, yeah, we we need to to to collaborate on on this this front. Maybe people who, wants to, jointly, apply for some projects for for for grants that deals with creating data that is combination of artificially created data and human collect created data that is governed by by humans. And, of course, when somebody is interested in in just using our software as a for for their own benefit. It's it's something that is always good for us, like, decision making, like Delphi, processes, like, public consultations, improving, internal deliberation in organization. Yeah. Those are people who would would like to collaborate, at the time. And, hopefully, we can with some some initial push, we we we can go back on on track on improving this combination of artificial and collective intelligence to something that that is for benefit of all, that is of benefit to to the people who are, collaborating with us on this solution. But the the the end goal is something that I I I think is is, just public good, something that, reduces the this this problem of of trade offs. Right? The this trade off of, security and and and freedom. The the the problem of, you know, having to manually engage in in in very minutiae of of, public discourse and being, cut it out from it completely by artificial intelligence and so on. It's something that we can we can combine without the decrease in, any of of of this extreme. We I think that our approach, can provide the the golden ratio, of being a subject in in in public, public life, and decision making of organizations and so on, in collective intelligence, but without the excessive requirements of, knowledge and and and, critical thinking and everything that that is necessary to, you know, to to to not to make any mistakes, to not not make any errors. This is something that collective intelligence should take care of. So, yeah, that if this vision is something interested, interesting for for listeners, you can contact me. My email is on SwarmCheck website.

Yeah. Initially, I had to, have a side job just to start the company. We didn't have any external funding and anything like that. We just used our own time, to develop some prototypes, to to, develop workflows. Our first argument map was on, just whiteboard and, some corks, cork table, with with the pins. So, basically, argument mapping is something that can be done manually, but, of course, you cannot cannot compute manual results that well. So we we started with with those type of projects. We started to to to to get funding from educational projects because, as I mentioned, argument mapping is very useful for developing critical thinking skills, and this is something that can be utilized very well in this type of projects. And argument helping builds interesting databases. So one of our first projects besides, these educational projects, whereabouts what what are the arguments on, skills, that will be needed in the workplace of the future. Right? So we we discussed with many experts, and and we created argument maps about those. And we, put it into one ebook, for for people to to see and educate themselves. So with projects like that, we went to bigger and bigger things. One of our biggest, projects were, educational project, think like a scientist, in which we showed on argument maps ways of analyzing the information from methodological standpoint. So, basically, we we we saw that, popular science sometimes is only about the results of of science. Right? So this is like a big telescope and look at the pretty pictures and scientists found out that chocolate is good for your brain or bad for your brain and stuff like that. But we want people to to to see, especially, young people and students, to to to see, okay, how, scientists know that certain claim is true and how they can know that certain claims is false. So we go we went into those, argument maps about the methodology in social sciences, in in, more, STEM fields, about the process of science, about the citation, about the peer review. And there is to to to surprise so many much to to improve there, much to criticize. Even the peer review process is something that, you know, some people are not very happy with because, the the review can be done by people who are not that that very well versing the subject. Of course, they they should be from the the same discipline and so on, but, they provide some criticism that one cannot easily disagree with. Right? They they you cannot have a deliberation, in in the process of of peer review. And, the different reviewers can disagree with each other. What what to do then? And, what what is what is the, mechanism that compels reviewer to to have, like, a final ground truth knowledge. Right? There is no such mechanism. So, of course, it's not saying that peer review should be abolished or anything like that, but it shows in that in certain ways, peer review could be improved even with argument mapping. Anonymous argument mapping could improve peer review quite extensively. And we showed how can scientific paper can be transported to, argument mapping realm. How, claims can be discussed, how sources can showcase that, okay, this this data is, not only good because it's in the paper, but certain aspect of the experiment produce the data that supports some claim. Right? So so from this, like, heuristics and broad view that, okay, this is true because the scientific paper said so, we can go into the details about the methodology, about the ways to to to be sure on or unsure unsure about the results, about quantifying the uncertainty, about the scientific results, about the replication of of the study. So those things, let us to to to to create this this project. And, for many people, it it helped to to to develop, critical thinking skills that we measured at the end of of the project. And, now we have a data that showcase that, okay, using argument mapping and having discussions about the methodology is a sure way to to increase, effectively your critical thinking. On top of that, we we conducted research on the user experience of argument mapping because, interestingly, the same information in plain text or in graph form or in dynamic argument map. So, basically, the map that starts with one claim, when you click, it show one argument, the next argument, the next argument. So sequential view of the of the map. The same information, makes people, retain information better or worse. So the worst is, plain text, and the best is a dynamic view of argument map. Why is that? It's hard to say, but there are some, hypothesis. The the one I I like is, I think is is have a lot of validity is the idea of cognitive scaffolding. So, basically, you, your attention, your your cognitive powers are not wasted on, maintaining the the the connection of of of data and, adding information one one by one builds this this, collective, sorry, the the the knowledge about the subject in in a way of, like like you build the the structure for the building, the the the scaffolding. Right? So you can navigate the scaffolding easily when you have the the means to do it, when you when we engage visually, in the logical, relationship of the of the knowledge you are acquiring. So that that was very interesting result, result for us. And, yeah, the the after the the educational projects, we we started to, engage, companies that maybe they they would like to use, argument mapping in their own processes. Unfortunately, many managers don't really think about, you know, making the best decision. They think about their job security, and, they're they can have, I think misguided, view that, you know, if they are not the sole decision makers, that they will, be seen as unnecessary by the by the company. And there is a lot of, reluctance, in private sector to to use decision making processes. So, in this branch, we we went to okay. When you need knowledge and and, you you know, your your job depends on it that this knowledge is is correct. So this led us to r and d, teams, to, pharmaceutical companies that they they need, studies to to, for example, Delphi studies to, extract knowledge from from experts and and be sure that the process is right. So so the this was easier for for, for us to breakthrough in terms of, private sector. But, at the same time, we started education project projects. We started, with cooperating with public sector as well. And public sector was more, enthusiastic about our approach because they are compelled by the law to conduct public consultations on on many policies. So, sometimes they do care about the innovation in these areas, but sometimes, you know, whatever you, do is fine by us. Just, produce the results for us. If formal aspects are are good, that's, that we are fine. But it allows us to to to actually, you know, test, this approach. And, as I mentioned, in case of educational policies, in terms of climate policies in in local level, in terms of public consultation, and, actually, you know, including the the people's arguments and opinions, to to some final policy, is something that that we see, that people who are part of this processes can see benefit. That they, they're presently surprised that the whole process of deliberation feels like a game to them. It's like something that is entertaining, in addition to to to be useful to to, to see other perspectives, to to voice their own, opinions. And people really like the the idea that it's not only opinions, it's not only, my way or the highway. It's, it's a knowledge that that can be, criticized. It it it can include some opinions that later on in the discussion were proven to be wrong. Because in everyday discussions in in in, let's say, face to face participation, mediated discussions and stuff like that, it's, too much relying on moderator's ability to, translate everything correctly and remember everything and summary everything in a way that is not exclude excluding someone else. And to to to if the conflicts arise, then the conflict is rather, hashed than than, you know, result of productively. But people actually, appreciate constructive criticism. But it's hard for people to to give constructive criticism, in everyday setting because this is not skill we are born with. Right? So, we we still do some, public, consultations and and policy making. But as I mentioned before, before, right now, we are most, mostly focused on our, Delphi processes and, and our legal tech, applications. Because in our mind, is in our opinion, it's it's something that can, can be the close closest to, the the the fastest application that utilizes artificial intelligence and collective intelligence, to, to the market's, need of having, a quick contract analysis, having, improve some legal reasoning in some court cases and so on. But on the other hand, it's not far enough from our r and d that can, improve artificial intelligence and collective intelligence in in areas like combating this disinformation or having crowdsource science and and creating big databases of connected reasoning in in a manner manner that is similar to Wikipedia, but maybe on more, of Wikipedia of connected reasoning, of connected arguments. So so this is the pathway that we are still on, but, we we, yeah, we we are focusing on on the legal tech and and developing methods at at the at the time.

Yeah. Yeah. We we are very interested in, DAOs and technology like that, for decentralized, organizations and decentralized decision making, and decentralized science as well, as I mentioned. But in terms of what what is the future of law, it's it's something that, have a spectrum of of possibilities. Right? So, on one hand, we can just see at the past, and how before, something that that we see as a given. So the rule of law, the judicial system that is independent of executive system and the legislative branches, the the things we get for granted, we take for granted, maybe, will, devolve. Right? Maybe, more authoritarian, system of, governance will, out compete. Maybe not necessarily outcompete, maybe just they will collapse from, internal, strife and and, you know, the the way currently politicians are getting power in democratic systems. There there are a lot of worrying signals, that, you know, democracy is something that may be a phase in in the history of of civilization. We wouldn't, like to see that. We would like to strengthen the the best aspects of democracy, but it means that democracy should evolve. It should respond to to the current, problems. Because alternative of democracy and the rule of law is just the dictatorship. Dictatorship as somebody dictates you what is the law and must obey or be put to death or something like that. Right? So going into the the happier, outcome, I think there there are still some important, challenges. On one hand, we can imagine let's let's start with positive future. We can imagine that, public discourse itself can be a governing force for the best law that we have. It's technically feasible that the discussion, like, we have right now and millions of discussions that people have, you know, in public sphere, not not in in their private lives, but when they want to engage in something publicly. We can use those discussions to, extract important arguments, reasoning, value based reasoning, and add it to to collective discussions. And those discussions can, you know, influence how the, rules are applied to to the whole society. Basically, we can directly and indirectly influence the, the rules by talking about them. Right? That would be something very sci fi, but totally technically feasible. So there there is one of the idea of that. So so, basically, we are governed by by our collective intelligence, in terms of country or maybe humanity as a whole. I don't know. Some some people see is it's as something positive. Some some people fear more more, nefarious things like new words or something like that. Of course, I I think all of it is possible. Right? We can if we can have global government that is positive for people. We can have global government that is that is bad for people. We can have individual nation state that are good for for citizens. We can have nation states states that are in war and, they're oppressing their own citizens. So having, you know, ability to use current technologies or, and future technologies to, surveillance, citizens, to to control them, to, remove their own agency and and personhood. And, you know, basically have the pretenses and and and the show of democracy instead of democracy as as we see in many authoritarian countries right now. Yeah? It's it's interesting that most of them conduct some, some, democratic elections just for the perform performance of it. That is real possibility as well. Right? And in those cases, law can be used as a system of oppression, as a system that, coordinates a part of, executive of of, you know, government, police, to to control, citizens, to to, check for dissidents, to to to to shape the society, as the most powerful people wants. Right? So so, we we the the the one, option is is quite good. Right? We we can just basically improve rationality of of law and and have a say in in our future and, use collective intelligence to to to navigate, our collective decision making, navigate, to to the future scenarios that that are possible to have good sense making and and so on. But on the other hand, we can just, you know, leave all all of that. We can, reduce our agency. We can reduce, ability to, make decisions on our own behalf, reduce voice of of criticism of the power of the government. And, yeah, I have only the the spectacle of of democracy. I think everything in this case is is possible and many scenarios in between. So, for me, it's it's still undecided. Right? What what would be the role of artificial intelligence? On on the one hand, it can speed up some some processes. It can, basically allows us for for better deliberation, for for better decision making. But on the other hand, it can be used as a surveillance tool tool. It can be used, as a means of thinking for us, not with us. It it can maybe, in the future develop as a new agent that that is, you know, the new force of of decision making. So, I think it's there are no physical objections to to to have, like, superintelligence, strong artificial intelligence that is as smart in, accordance to to humanity as we are in accordance to chimpanzees. Right? So when we are the second most intelligent species on on the planet, we we we have the same ability to influence the future of this planet as the second as the current second most intelligent species have. So that that that's worrying. Right? And I think only alternative to it is collective intelligence. There is no other thing. We we cannot being stupid there, we cannot control something smarter than us. I think that many many smart people think that it is it is possible to some extent, but, I would present maybe some some argument maps. So so so the, the the risk of artificial intelligence that is smarter than us and is basically pulling all the shots about the future that our planet is moving into and realizing their own goals, It it is quite possible in I don't know about near time future, but even in medium term, it's it's not something that would that I would exclude. Maybe even if, maybe even not if not not in near term future. So, the the risk of being the second most intelligent species on the planet that does not have shared means of decision making. So basically using collective intelligence with the artificial intelligence as the, method of of making decisions about our future, the future of civilization is something that that is, one of the biggest risk. If we don't if the ability of artificial intelligence increases and our ability of incorporating this intelligence into our collective intelligence is not increasing, we will be left behind. And, I don't know if if it was recorded, but I compared the the most intelligent species right now to the second most intelligent species, chimpanzees, and their ability to influence the future. Right? So if we, as a humanity, will go down as a second most intelligent most intelligent species, I don't think it would be wise to to to not to give ourselves the ability to to be incorporated in the decision making process of the future artificial intelligences, and the the only way through this process is collective intelligence. The the the interesting thing that that, showcases that is that this is possible, is the interesting aspect of technology that you don't really need to understand everything to use technology. Right? You you don't have to know how the, electricity is made, how the light bulb is created, how the, electricity is, moving, you know, through the city, the the the collective, electric grid, you know, is is operated to just push a button and and you have a light. Right? The the there is some power into in in the setup of, of knowledge that we can use. And I think that the the ability to reduce knowledge in augmentation have the same property. Right? So we we still can be, on equal footing even, with, intelligences greater than ours to be, part of this collective negotiation, this collective augmentation. Because if our voices, voice is added to the voices of others and and the argumentation of others, it creates something that is more than than just our one idea. Right? It is the complex of idea. All the ideas that were discussed in this, that that were discussed in in this, in this topic that can influence the the decision making processes of even super intelligent beings. Right? Because once the the knowledge is produced, it can be, applied many, many times for our our benefit. So the philosophical discussions, the discussion about values, discussion about why it's good to do something that that motivates, and frames the decision making are so crucial. And we don't really have them because, the Internet is full of the the surface level discussion. And the data that is used to train the models is mostly on the surface level discussions. But we need the ability to join the, value based reasoning to, argumentation that that can go deep into the nuances and represent many perspectives in into this this type of discussion to preserve our our human values into the future, into the data that is used by stronger and stronger artificial intelligence.

Yeah. The the the the first thing is just, you know, if the things I I spoke about are in of interest to you, just contact me. We can see how we can collaborate. You know, the the the scale of collaboration can be discussed because we are quite flexible, and we have experience of incorporating many people, to our organization. And another thing is that, I think that it's it's very noble to to to work in, civil tech, because it's, it's something that is much needed and it's not very easy to, to survive in this field I would say the the fact that we after this big, collapse of of one of the our biggest projects, we still managed to to survive with additional project problems like that is due to the, tremendous amount of work by our team. And, I think that many, civil tech organizations and and people who are interested, manage to struggle with similar issues or just, you know, just having to explain those complex ideas to about new ways of governing and or the technical and UX difficulties that that stems from that, is a thing that oftentimes do not result produce result quickly. So, yeah, in terms of having impact, which is, I I believe, something that many people care about mostly, it it it is good that we have, initiatives that, you know, can, we we can network be network, like, with the meta gov initiative. The the the podcast as yours is is very good example of something that, can, be a beacon for many people who are interested in the in the subject. And, I I would say just, you know, don't only look around, just give it a go. It's not that difficult to to, to be engaging in just one small project that that have a beginning and end and try it out and experiment a lot, and share the results with with the community, especially, I think good examples of applied civil tech are very encouraging because they are not only showing the community that those things can be done, they show, decision makers that this is something that is happening. Others are using it and having good results, really goes long way. So, yeah, I I would encourage you to to just keep persisting. And, yeah, if you have ability and time, engage with the community, engage with me if you want. I I oftentimes spend my own free time about to discuss the the those issues and and those those projects with people. So, yeah. Keep on. It's it's one of the most important thing, one can work in. So, given our current state of the world and, you know, the trajectories, for for democratic governments and democracy in the future and emergence of artificial intelligence, we are on the crossroads, and the future is uncertain, but we can push it a little bit into the directions that we all would like to see. So that that that would be my message.

Yeah. I think that the last thing you you the the the aspect of of, you know, worse is something that inspires me to to to to mention something. Because when you see ordinary people that, have to be in some worse situation and to to to be soldiers and so on, Governments have to come up with the ways of, you know, convincing, you that you are able to to to kill another. It's not it is not natural for for humans, for most humans, excluding maybe some psychopaths to, take, a person's life. Right? So so this is just something that, are, of course, powerful forces that shape us into, these situations in which we kill each other. But there there are so many cases that we can show that when we are able to to talk to each other, even our enemies, right, the the ability to to resolve conflicts are are immense if the, you know, powerful governments are not backing again. And I think that we as a society need something that will protect us from the abuses of power, the abuses of tyranny, abuses of people who, invade other countries, abuses of power of of people who, you know, want further their own interest. Because if we are able to talk to each other directly without our governments, but, coordinate as a person to person as as a, you know, ordinary humans that want to live and and have a good life and so on. We we have so much thing in common. And, the only, way that and in the past, there was this vision of the Internet. Right? That when the Internet will emerge, we'll just have this connection. We we we don't have it, but it doesn't mean that it it is not possible. We just have to have right tools for for this type of communication. And, yeah, we just don't give up. Don't don't think that, social media as it is is the only way to to to communicate. The social media didn't exist two decades ago. Right? It it it everything changes, so fast that, it it's important to to remember that that we today, we are building the future. So don't stop your imagination, on what's possible.

Also Yeah. We have to be sneaky about this.

Thank you, Alex. Thank you for having me.