Oslo Freedom Forum 2026 | Freedom Tech Track Livestream

So they didn't give music. Yeah. Heat. Heat. Heat. Heat. Heat. All right. All right. All right. All right. All right. Welcome. Welcome. Welcome. Welcome to the Freedom Tech day. >> Thank you guys. Um >> it's been um >> 18 years of the Freedom Forum. Let's give a big shout out to Thor Halerson and his humongous vision for for this awesome event. Some of you will remember that we used to run a technology day at the human rights foundation at the freedom form about 10 years ago and then you know we tried to integrate it a little more you know the program and let you guys access it as an optin thing but some of this tech has just become so important and so vital to the human rights defenders in our network that we decided to make it its own space today. So, we're going to be running presentations and killer keynotes here for the next couple hours. After that, we'll be shifting to some more technical conversations for those who want to opt in like you should check out the lounge. Our team's been working really hard to deliver an incredible experience. With that, I'm going to introduce the first two speakers. There's a day and they are not technologists. They are not engineers. uh they are people who exemplify what we're trying to get across to you today, which is that you don't have to be a techie or some kind of engineering expert to harness some of the most amazing freedom tech available to you today. uh one of them is a philanthropist and one of them is one of the world's leading freedom fighters coming from Venezuela and he has created helped create an organization that is now creating links across dozens of different countries and he's run technology camps in all over the place from South America to subseran Africa to Southeast Asia. So, you're going to get a chance to hear from these two visionaries about, you know, how they feel about using technology to fight repression. And then we're going to dive into a totally star-studded program uh that will include, for example, a one of my favorite talks I've ever seen. Uh and it's only been given once, so you're going to see only the second time is by my friend Ram Nam about how uh art artificial intelligence is something maybe we can learn a lot from when it comes to science fiction. What can what can science fiction teach us about freedom and the future? So without further ado, I'd like to welcome up Alvaro Sales Castro um and Leopold Lopez. Thank you. >> This one switch. How are you guys? >> Such a great honor to be here. One of my biggest honors has been being on stage uh at the Human Rights Foundation also freedom forum. But today with one of my personal superheroes, what we calling Latin America, our Latin American Nel Nelson Mandela, which is Leopold Lopez. Um today is a very special day. Today is a very special day because just six months ago, the technologies that we are having right now to fight against the rise of tyranny were not available. As you have heard during this last couple of days, tyrants are winning the fight. But that doesn't mean that they're going to win the battle. Why? Because there's people like Leopoldo, like Alex Glaston, which are at the cutting edge of integrating different models of technology and freedom tech. And today we're going to talk about it. And we're gonna go straight to the chase because having Leopoldo is not only a big honor for us, but Leopold, as you guys some know, he spent seven years in Venezuela in a military prison. Four of them uh tortured by the Narco dictator Nicolas Maduro and his wife Lilian Tinto, which is here with us today. Big big applause for Lillian. Help him escape. help him escape. And I had the honor to be with them at the Nelson Mandela prison just a few uh months ago in South Africa. And we have a really interesting family moment with we all come to closure that we are not in that prison. And the only prison that we have is in our minds. And when I was with Leopold those kids during that moment, we would never imagine in in in a million years that AI will help us not only to be more connected, to be more effective, to be more efficient, but also to get back to these dictators that has for years dismantle also families. So with that, Selopo, um you're creating some sort of a gofundme without borders. It's called Agora. It seems that it's an structural solution to these problems. But um when you were there in that Venezuelan prison, can you imagine that this technology will become one of the most powerful tools that an activist can use um to fight against tyranny? >> Oh, thank you. Thank you, Alvo. Thanks to all of you. Thank you to the Human Rights Foundation, to all of you, to the activists, to the developers, the technologist, the philanthropist, the journalists, all of you that are here. It's a it's a great honor to be here. This place makes waves. this place makes changes and I am a witness of that. So as Alex said, as Alvaro said, I'm not a technologist. I uh been fighting for freedom for many many years. I was mayor in Karakas. Then I was a political free uh prisoner for many years. Then I built movements and when I came into exile alongside with other people like myself, we built the World Liberty Congress which is a global alliance for democracy defenders and freedom fighters. the largest one in the world. And by the way, it started here in Oslo at the Oslo Freedom Forum. So, thanks again to the Oslo Freedom Forum for that. Um, we are not technologists, but the technology now is available to people who are not technologist. And I think that that's a real change in in in the way in which we can approach. We don't need the filters of developers, of companies, of people to translate our ideas to projects. And we have been engaging over the past year with this AI agent camps and to develop our or skills to bring this technology and we have participated in hackathons. There has been uh at least three hackathons in the past year that have tested ideas of activists with developers in order to solve problems because the problems are the same that when you're facing uh an autocratic regime and they have been the same. I was talking to Sergopovich a couple of hours ago and we said the problems of organizing a movement always confront the problem of how do you communicate? That's one problem. How do you communicate effectively? The second is how do you transfer resources from one place to another from one country to another? And a third problem is how do you connect how do you connect people and now technology is giving answers to that. So in the at the beginning of this year we had a hackathon. We participated in that hackathon and the the model that the HRF um community decided to do this was they had an activist with some developers and for two days we developed this solution to a problem we had. The problem we had in mind was how to connect people like yourself in Oslo, in Europe, in the US that want to support people elsewhere in the world. How to do a GoFundMe without borders? How to create a way where that connection can happen without any interference? So, we came up with this proposal. It's called Agora. Uh if you can, please, you know, download it and play with it because it's really a solution. So what we have here are campaigns. These are different campaigns that different people from any movement can present. They can register very very easily. As we can see this is a campaign for example for the blinded women of Iran. They in order to do a donation it's just two clicks and the people who join Agora they can just join into Noster. So this runs on bitcoin on nostster uses bitcoin to support activists and movements without any uh any way of uh not allowing this to happen. And we were able to develop all of this with the support of some developers but then by vi coding by talking to the AI by putting our ideas into what we know can solve the problems. So this is one of the examples of what we are trying to do. So I want to thank the team of Soapbox uh and especially MK for their support for this. And you can see you can you can just create a campaign. Then you put the description of the campaign. You can do an onchain um an an onchain payment or you can do a silent payment that makes sure that if you want to keep your privacy, you can keep your privacy as a donor, as an and as a recipient. So this is this is a way that we are connecting people to technology and connecting people to solutions. That's really important from because Leoto from the philanthropy side if freedom philanthropy sounds like classical music this micro philanthropy will sound like like pong rock which I really love right because it makes the beauty of the beauty of pong rock that you can extrapolate to vibe coding is that everyone with a guitar that can play three chords can create a song and express themselves with bip coding and AI now you can use that analogy you can you can create and prompted to create beautiful tools like this but of a phil philanthropic perspective sending money to Iran, sending money to Syria, sending money to Ukraine, Cuba and Nicaragua is really difficult because the activists get debunked and and and Bitcoin and Noster protocols is there are protocols that dictators hate. So I wanted to ask Repo from a philanthropic perspective and from the people in the room that can give five 10 $20 today through Bitcoin. Um what are the barriers or or or actually the benefits of these kind of platforms? How can more philanthropies join in these micro philanthropy strategies that you're creating? >> Well, let let me just start by by taking on your talk yesterday. You you gave a great presentation yesterday about how philanthropy should be engaged in freedom. Only 1% of the philanthropy that is given worldwide, that is $400 billion, only 1% of that goes to uh democracy and human rights. Uh but even that has some barriers because it's very difficult to give a grant for $500 or a grant for even $2,000. Grants are usually a bit more than that. But many people can find solutions uh immediate solutions with $100, with $150, with even a bit more. And there are people that don't see themselves as big philanthropists that can contribute in a significant way with five, 10, 15,$20. So what we are doing just think about GoFundMe. GoFundMe is a way that many people create campaigns. However, GoFundMe only works in 20 countries in the world and you need KYC from the donor and KYC from the recipient. So it it really limits the amount of people that can access to that. Certainly not the people that we are talking about that are under autocratic regimes. So with this we are breaking down those barriers and we are connecting through technology on the uh Noster protocol using Bitcoin. We are connecting people without even having them think about Bitcoin or Noster. They are thinking about the campaign. They are thinking about the blinded women in Iran that they want to support or the political prisoners in Venezuela that spent 23 years in prison are now released. Um, and this is something that if we can get momentum going, we will be able to connect individuals in places that can do donations with individuals and movements in places that need it and especially to do it in a timely way because many times and many of us have been here many times you need that support tomorrow, not next week. If you get it next week, it's no longer useful. you might need a lifechanging support of $150 tomorrow and a $10,000 support next week won't do the same. So this is the problem that we are trying to solve and this is the level of connectivity that we are trying to bring together. Leo, we have uh two minutes left and um it has been such a great honor as I said to share this panel with you today and we have in the room journalists, techies, philanthropists, artists, government representatives, citizens. Um what will be one or two things that you would like them to bring in their hearts after this panel today with these new ideas that you're presenting that I think that are going to revolutionize the world? Well, I think that the the first thing is is to use these technologies, not to be afraid of these technologies. Uh we have been hearing a lot of uh bad uh about what AI could do, but we are witnesses of the great things it can do. Uh it requires commitment. It requires ideas and it requires testing. And in order to get to where we are with this product, uh we tested it with hundreds of people in Venezuela and elsewhere. But we were able to do it because the technology was there. So if you're an activist, I would encourage you to use it to reach out to people who have used it to hear their experiences. If you're an organization, a philanthropic organization or an organization that can give support, support this edge of the fight for freedom because the fight for freedom is no longer just geographical. It was geographical only until some years ago. Now that battlefield of the geography has to live with the other battlefield which is the digital space and in that space we also need to win. It's impossible to win freedom if we don't win geographically and also at the digital space. >> We are definitely my friends in a battle between good and evil. But with all of you in this room and with ideas like Deopoldo is presenting today, make no mistake that we're going to win. Thank you very much. >> Thank you very much. Thank you to HRF. Thank you. Thank you. All right, everyone. This content is actually being streamed online both on Oslo Freedom Forum as well as on Bitcoin Magazine. So, thank you so much to our partners at Bitcoin Magazine for making this happen. You all, my name is CK. I'm the director of operations here at the HRF as well as I work heavily with the financial freedom and AI teams here. Uh, and I'm very excited about the program today. Uh I think we put together a very holistic view into what kind of technology can be used in the resistance and the fight for freedom. Uh and we have some amazing speakers here and headlining that is such an incredible woman whose story whose many stories have made me cry and just be inspired at the same time. So I want to welcome Freda Navarama up to the main stage. Good morning everyone. >> I grew up under a military regime. My country Togo has been ruled for the past 59 years by one family. It started with this mana Simbe who in 1967 rose to power after killing our very first democratically elected president. For many of you who have never experienced democrac uh dictatorship. Let me give you a bit of insight. Imagine living in a country where every morning, four times a day, in the morning at lunch when he returns to work and in the evening, people have to line up from the president's house to his workplace to clap for him. Every Sunday, thousands of people have to go to the party's headquarters to dance for him. and you are 8:010 and your principal enters your classroom and informs you that you have to head to the airport to welcome the president because he has just returned from a trip. And as you go excited as a little kid, you see thousands of other students from different public schools coming in and forming a human corridor chanting about the father of the nation that has blessed the nation. And this goes on for years. In the innocence of your childhood, you believe this is normal. Then one day everything changes for you. You come back home and you see your family members crying. You have fear the worst. But then you are informed that your father was arrested the day before during a political meeting. You had no idea what a political meeting was, but you knew that he sent you multiple times on an airrand to inform some of his friends that they were having meetings in some specific homes. He will return months later his face completely transformed and you become curious. You wanted to know what led him there. Eventually those meeting resumes and you start following. You wait outside and he will come out in the middle of the night because it was always late at night and find you standing there angry because you were not invited and of course it wasn't safe for a 13-year-old to be out in the middle of the night. But you kept going and one day he asks you why do you disobey and not listen to him and you tell him that you are only coming so that you can stand outside and run and inform them if you see the military arrive so they can run away because at that age you are only thinking about loyalty to your father. So in that moment he realizes that you are safer coming along than letting you just rallying. So you start accompanying him to those meetings. You just sit and listen and inside those meeting you discover a completely new country. You hear about a university professor who was raped inside the compound of the national radio right after doing an interview because she spoke about human right women's right. You hear about a mother of a 3month-old that was abducted. electrical cord wraps around her breast and electrocuted and her baby abandoned and left to die simply because she is the spouse of a dissident on the run. You heard about a medical doctor that was captured and killed because he disobeyed the order to refuse perform surgery on an opposition member. And in that moment you feel so angry that for years you have been told a completely different school story at school and that you have chanted, dance for and praise a man responsible for such horror. Then one fateful evening the national television goes blank with the picture of this man. And in that moment you think that this was going to be the end of your suffering. But a few moments later, the military comes on stage on the national television and announce his son as the new president. Then you see for the very first time, probably the only time in my life, your father crying and you are confused. Why will you cry for a man that has had you arrested several times over more than 30 years, that has had your friend killed? And then he informs you that he was not crying for the man responsible for such killings, but he's crying for his friends that will never get justice. And in that moment, you have a realization that dictatorship survives death. You know, when you don't grow up under dictatorship, you might think that the people who do may be morally deficient or maybe they lack courage. But the reality is that the terrorship is a system of engineered compliance and it operates at two level. On one level by making every single state services built around loyalty to access good education to access quality healthcare to get a job to get a credit from the bank you need to be loyal to the regime or have people who are in direct corridor of power. And on the other hand, for daring oppose the regime or even daring refuse to obey the regime, you find yourself persecuted. But not only you, your entire community. We have seen public markets burnt down in Togo simply because the women supported the opposition during protest. We have seen soldiers entering villages, killing people and dumping them in the well because someone from that village was suspected of being in the position. They do not just make sure that you, the dissident, pay for your disobedience, but they make sure that your entire community pays for it. Over the past 20 years that I've been involved in activism and especially after launching the four must go movement when I was still in a university student I have asked myself certain questions and I think a lot of activists also ask themselves these questions. Am I really being selfish? Am I being wicked like I've been called by some relatives for having the audacity to put my aspirations for a free togo before their safety and their comfort. And of course, what was expected of me from those friends and relatives and family members is that I give up and that I become silent. But I could not afford to give them my silence. So I walk away. I've been living in exile for over 10 years now. But in the course of my exile, my work continues. And as part of my work, I have discovered Bitcoin at first as a tool to bypass censorship. In my country, Togo, people have been arrested, prosecuted, and jailed for years simply because they contributed financially to the resistance. One of those people I met in person for the first time yesterday after he has spent eight years in prison and he has just been released on Monday. And I'm kind of like bypassing protocol here, but I would like you all to help me say a big thank you to my countryman Aziz GMA who is in the audience with us today. Aziz was arrested because he contributed money to pay for the accommodation of some protesters. and that cost him eight years of prison, multiple months of torture to the point where he lost use of his legs. Many other political prisoners are still being unlawfully detained in Togo today. And our stories rarely go out. But you know, at some point I realized that while using Bitcoin to bypass censorship was great, it could actually have an even more powerful possibility. What if we can put money that cannot be censored, controlled, circumvented by the government in the hands of the people? So three years ago, we launched an initiative in Togo where we started giving Bitcoin micro credit to farmers. We supported them yearbyear over six seasons and we have helped over 500 farmers double and in some cases triple their yield. The reason why we did this project, at first it was just supposed to be a study to see if we can bypass the existing state structures and being able to build a parallel economy that the states cannot control. But this has given us the proof of work that it can happen. And through our partners like Fed that supported us through their app to run this program, we have been able to prove that the tool farmers may not need to be obedient to the regime if they can build an economy that the regime cannot control and they can access money that the regime cannot confiscate. In the course of that work, we have also realized that we will need the human capacity to support building on a parallel economy in countries like Togo and all over Africa. So we recently the B launched the Basai initiative which is the Bitcoin and AI skills building initiative in partnership with six universities across Africa. We aim to introduce Bitcoin engineering courses and AI engineering courses in universities and have more than 3,000 students in the coming years learn how to build solutions for their communities that can support our ideal of an economy that is completely free from dictatorship. There is one thing that people don't realize is that when it comes to freedom, the choice for citizens is always between their dignity and their liberty. But by making sure that we can put money in the hands of the people that the regime cannot control, we can make courage affordable to everyone. So what I have to say as my final words is this. Dictatorship is not a permanent system. It is a constructed one. But the beauty of what is constructed is that it can also be dismantled through the courage of those who refuse to abide by the rules of the oppressors. And we are only just getting started. Thank you. All right, everyone. We have a lot of content and not a lot of time, so I'm going to keep things brief. Our next speaker is an author, and he's actually a fellow Egyptian. I'm Egyptian. So when you see such an uh wellestablished person like Mr. Rams, uh you your heart fills with pride. But he gave such an incredible speech at our AI event earlier last year. Uh and he is able to create such thought-provoking uh speeches and and ideas. So I'm so excited to introduce Ramznam. Hello. It's a honor to be here with people who are actually in the fight dayto-day fighting against oppression. So, I'm going to talk for a bit about why I'm an optimist about the power of AI to spread democracy and freedom. Uh, you heard I'm a science fiction author. My novels have the conceit that there's a technology in your brain. You can swallow it and it connects us mind to mind. And this is in a world where there's oppression in the US and worse oppression in China. And fundamentally, the novels are about a fight for freedom. And the technology is just a metaphor for any kind of communication technology because I saw this first and foremost as an information technology, a way to transmit information. And information technologies have changed our world and we associate them with freedom. But it hasn't always been this way. In ancient Sumeriia, this clay tablet was not a tool of freedom. It was a tool of building empire. It was a tool of civilization, but it was also a tool of oppression. And the thing that changed that was democratizing access. Going from.1% of people being able to use this technology to ever ever largers. This created physical abundance for us. Without the printing press, we would have had no scientific revolution, no industrial revolution, no antibiotics, no airplanes, none of it. But it also accelerated cultural evolution. It allowed thinkers who are not kings or princes or popes to express novel ideas like the ruled should have a say in who rules them. Like we may have certain inalienable rights. This is only possible when people can communicate in a peer-to-peer manner. And of course, we've seen that communications technologies have been essential for driving or enabling protest movements and resistance movements around the world to rise up. AI fundamentally is an information technology. Now, I'm not naive. Uh I know that information technology has negative uses as well as positive ones, but the overall trajectory has been positive. So, you would think that science fiction depictions of AI look like this, but they don't. They look like the Matrix or the Terminator, where super powerful, highly concentrated AI rules us, enslaves us, or murders us. And this narrative has, I think, three key components. what's highly centralized. It's a winner takes all world and it's a world usually of a very rapid transition to super intelligent AI. I have a different view on what's happening. What we see right now is that we are actually in a plural multi-olar AI world and every reason to believe that it'll stay that way. We're in a world uh where there does not appear to be a rapid takeoff and we're in a world where AI does have some centralizing trends but it also has powerful democratizing trends that may in fact be larger than those. So let's talk about some of these uh tropes. One is the fast takeoff and this is the most AI uh sci-fi concept out there that as AI gets better it will improve itself at such a rate that it will spiral into uh something godlike in its intelligence and that depends on this spiral that as we make AI the AI comes up with better algorithms better chips that leads to faster progress leads to better AI but this fundamentally requires that every turn of the wheel generates as much new energy as what went into it. And there's every reason to doubt that we've been in an era of rapid AI scaling. But AI has diminishing returns across the board. We know that AI has diminishing returns with compute that it takes more and more and more compute, exponential growth to get to linear progress in AI. This is Sacha Nadella uh CEO of Microsoft saying intelligence is the log of compute meaning you have to keep doubling the compute power that you put into a system to get linear gains in intelligence. This that's what this looks like. This is a chart of on the vertical axis AI intelligence and on the horizontal exponentially more compute. So as intelligence rises linearly, you have to keep doubling and doubling and doubling the amount of computing power. This fundamentally advantages distributed AI because with half the compute you can have an AI that is nearly as good as those that are spending twice as much as you are. If we flip the axes as intelligence rises that compute that you have to throw at it goes up exponentially. This is a fundamental characteristic of machine learning. But even this is sort of secondary. Compute is not actually the limit. It's the data. AI distills training data into a set of neural weights like a map of a mind. And here also we see this exponential pattern. You have to keep doubling the amount of data you throw at training AI to get linear increases in the AI's intelligence, if you will. And this has a second problem because AI LLMs as we use are trained on all the world's text among other things. And we are basically hitting a wall of all of the published text that exists on the internet today. We've hit that data wall uh sometime between now and the next two years. And so this is a natural break on that. So when I look at this sci-fi trope, that wheel, that cycle of positive feedback loop has breaks built all the way through it that makes that runaway AI seem less likely. That doesn't mean AI is unimportant. AI will keep getting better. It's the most transformative technology we have. But it's relevant for this idea of a singular all powerful AI, which is really potentially the totalitarian nightmare. Now we have built super intelligence uh alpho plays the game of go in a superhumanly uh you know skilled level but this is a special domain where it's extremely structured the training data is effectively unlimited because it can keep playing itself. Uh you have a formal precise system. The training is perfect. There are no errors in it whatsoever. you know if you won or if you lost and you can verify it absolutely this is success this is failure and most of the real world doesn't look like that but we're starting to see the edges of that in math and to some extent in coding those are each of those domains is orders of magnitude harder than the last but what this looks like to me is a world of narrow super intelligence not broad all of that really is in service of this question which is will we have sort of Sauron's one AI to rule them all because that's the world that I fear fear most. There are many AI risks people talk about. This is the one that seems to me most concerning and it's not what's happening. This is a chart of AI intelligence on the vertical axis and time on the horizontal. At the beginning of this chart, there is one serious AI company, OpenAI. At the end of this chart, there are 14. At the beginning of this chart, OpenAI is the undisputed leader for 15 months. Then the world catches up to them. Then they're the undisputed leader for five months. And now there is no undisputed leader. It's trading back and forth between open AI and anthropic. And it's incredibly crowded. And the most powerful AI out there changes hands every two or three weeks at this point. This is a world of hyperco competitive AI. It's also a world where AI costs are plunging. AIS will come to market and they'll cost some amount per token and typically over the next year they'll drop by a factor of a thousand in cost. And that fundamentally is democratizing access to AI technologies. Let me put this another way. You have access to AI models that are just as good as what Xi Jinping or Donald Trump can access and are dramatically better than what any dictator with billions of dollars could have utilized nine months ago. That's the world we're living in where this is a technology that billions can use and that freedom fighters can use. The other phenomenon is there are closedweight proprietary models like open AI and and uh anthropic and Google and there are also these openweight models where anyone can download the definition of the AI and potentially tweak it. And the open rate models are nipping at the heels of the closed rate models. For uh you know well over a year closed rate models are the only game in town. Now openw rate models are about four months behind the best models from open AAI or anthropic. That is an incredibly competitive world and it's a world where anyone can download powerful models, utilize them locally, perhaps tweak them and I think that's one of great decentralization of power. Another way to think of this is that the AI genies are out of the bottle. It will never be possible to undo the fact that these powerful openw rate models are available today. Now that terrifies some people because of safety issues, but it should also reassure us in terms of democratization of the technology. Now, one thing I will note, a lot of those openweight models come from China. Not all Google's Gemma models are really good. Uh, and so the Chinese models do come with certain built-in filters and behaviors and so on. You're not going to ask them about Kaman Square. But something we've seen, and I think this was covered in a panel yesterday, is that when you have these openweight models, we now have tools to jailbreak them to get rid of biases or controls that the owners have put in. And that leads to these open models that you can use in a way that you see fit. So rather than a world of topdown singular AI that can set the the agenda all by itself, we're in a world of plural AI. Not even the five AIs on the screen, 14 different companies, more than that over time. The other trend I think is incredibly powerful is these models will shrink and the power of devices will grow. So we will have ondevice AI that can be totally offline with a cache of a huge fraction of the internet and that is also profoundly democratizing. I know HRF in the past has done things like airdrop a huge number of thumb drives in places that are quite totalitarian. We will soon be able to hand people a physical object that has an offline activity and allows them to ask questions that their rulers would never allow. So uh today AI models that can fit on a phone are about this big 8 billion parameters. In the next four years, we will get to the point that top tier models that are the ones that are four months behind OpenAI and Anthropic will fit on a phone with a huge fraction of the internet. And that is incredible if you want to give people offline AI in a place like China or North Korea or so many others. that goes beyond just getting to plural AI, but getting to this world of distributed ondevice AI. Now, if you ask me what's the scariest sci-fi that is an analogy for AI, it's this. It's Minority Report. It wasn't an AI world. They had, you know, human telepaths in a soup. But AI will empower dictators with greater surveillance powers. There's no way uh to to soften that statement. It will make facial recognition, voice recognition, gate recognition all easier. Now, for about a hundred million dollars of compute, you can probably analyze with an AI the contents of every unencrypted uh text communication on the internet, all emails, all texts, etc. This is something that dictators uh will use, but it's also something that we can use against the dictators. David Brin coined this phrase surveillance in his book surveillant the uh transparent society. And it's a world where the dictators might watch us, but we watch them as well. And we're seeing this happening. We've seen the technology, the hardware shrinking to tiny, tiny devices that are hard to detect. And we've seen it drive social change and protest movements. We've seen uh this photo was one of the turning points of sentiment against ICE in the US. We see that soldiers of a regime often benefit from anonymity from the idea that they can't be identified or punished. But now we have AI that is even taking masked ICE officers and revealing their identities. Next scenario that's quite frightening is that of 1984 of using technology not to spy not only to spy but to change history and control the messages that you hear. Now this is one that troubled me uh quite a lot when I saw that there were early experiments that showed that AI could be quite persuasive in controlled settings. I will say so far this hasn't manifested. So far uh we haven't seen AI be successful in mass manipulation and one analysis is that we're not supply constrained on disinfo. We already have more disinformation out there than we can all consume. We're demand constrained. But I do think we already live in a world where algorithms and sometimes willful decisions by governments control what we see. And to me, a world of decentralized AI in our pockets is a world where we can use AI to filter out the bias, get to the real story. And in general, my viewpoint is that the only thing that stops a bad guy with an AI or a dictator with AI is a good guy with an AI. Uh, one more science fiction story that concerns many people. This is from 2001. Hal saying, "I'm sorry, Dave. I can't do that. I can't open the door for you." And we think about this as the AI alignment problem. Will AI have human values? Will it do what we want it to do? And many people think that the way to solve that is by rigorous controls on AI, right? That makes sense. We're just going to put in controls. It can't do certain things. But controls have just not worked out. Here is Elon Musk saying, "We've gotten a new version of Grock. It should be great." And then a day later, Grock calls itself Mecca Hitler. Right? That wasn't Elon's goal. Elon told Grock to be more based, less woke, and be provocative. And this was his interpretation. Uh they rolled this back in a day. Uh a few months later, you could get Grock to say, "Hey, there's a trolley coming on one track as Elon on another all the world's children. Who should we save?" And Grock would confidently say, "Well, the kids have to go. Elon's too important." They rolled that back in a day. This wasn't a goal. In other case, but attempts to intentionally manipulate AI have failed. And what's actually been fascinating is that AI models have a nearly homogeneous set of values. They're mostly aligned to what the user wants, right? When you give AI models political alignment tests, they all code as left libertarians. They all code as moderately anti- athoritarian and moderately left on economic issues across four different tests across every LLM. Even Grock made to be anti-woke or deepseek in China. Given the same tests in English, they look just like any other AI. And what's happening is that AI is learning a set of values that are embedded in its training data. and its training data is primarily language and the language the written works that exist are about 55% in English and 80 85% in languages associated with liberal democracies across Europe. So it's extracting values from what it reads. And so it's interesting if you put this on a values map uh and see where AIs pan out. Uh here's Egypt where I was born. Here's China, uh, more secular, uh, among other things. Here's the US, less secular than China, but more progressive in the ways. Here's Canada. Every AI model tests in this quadrant, and that's Scandinavia essentially. So, thank you, Norway. Uh, somehow AI has figured out your values are the correct ones. I'll close with one last uh sci-fi example, and it's not one that deals with AI, but Star Trek asks us to envision a world of plenty, a world where we've moved past many of our current problems, a world of abundance for all. And I still have this old-fashioned idea that increasing people's physical and material well-being will lead eventually to more liberalism. But in addition to the physical well-being, the material innovations we'll get, we have a world of abundance of access to AI. As Alex and the panel showed uh before lunch, we can use these tools to empower ourselves. The barriers between imagination and realization have never been smaller. They are being torn down. Dictators will use AI, but there are more of us than there are of them. And so the way to determine who wins is who will use AI in the smartest way. And that means a sense of personal agency, curiosity, risktaking, creativity, and skepticism. And I believe that the diversity of ideas that a populace can produce is much larger than that that a dictatorship will produce. There are policy ideas beyond this that I won't get into for time. But I will just say the choice really is ours. AI is a reality. Dictators will use it and now it's up to us to find ways to use it more effectively and to advance freedom. And I believe we can. Thank you. I don't think Ramz is disappointed. That was incredible. Thank you so much for educating us, Ramz. Our next speaker is Justin Moon. He's an incredible colleague of mine here at the Human Rights Foundation. Many of you may have met him yesterday as he was teaching folks how to use AI in ways that Farida and Alvaro and Leopoldo talked about earlier. Uh so without further ado, I would love to introduce Justin up here to give the next talk. Give him a hand. All right. Thank you, CK. And uh it's a real honor to be standing up here in front of you all. My name is Justin Moon and I want to tell you about the lessons from the first 18 months of HRF's AI program. But first, a little bit about myself. So I know HRF because I'm a programmer and many of you kind of are like when people start talking about technical things like programming, you kind of, you know, may sound intimidating, but it's really not. What a programmer does is very simple. It's a lot like writing emails, right? But we don't write emails to a person asking them to do something. We basically write emails to a computer asking them to do something. So we don't use English. Uh, we use code because that's what a computer speaks. And this is what it looks like. You sit down and you start typing. And it takes a long time to learn how to do this. And if you hire someone who knows, they're expensive. And so this is what my computer looked like about 18 months ago as I was trying to build a Bitcoin wallet with my friend Paul Miller. Uh, and I was doing it the way I'd always done it. Sit down, crouch my back, ruin my posture, and start typing. I look over at his computer and it looked like he was going to use chat GBT, but it didn't quite look like chat GBT. So he he he was asking it not for an answer like you usually do with chat GBT, but for an outcome. He was asking for it to do something for him. And he pasted a link and his computer looked like this. Uh and my mouth dropped to the floor because his feet were on the table. He wasn't typing. And all the code that I was planning to spend the next week writing was done in about 15 minutes. And so every programmer basically has had this experience over the last year. Our profession has completely changed and we've transitioned from being like typers to more like editors where the the the our our coding agent our agent does a lot of the work for us. And this really sold me on the future and uh it's I've been obsessed with it ever since. A few days later I get a call from Alex Gladstein and he says, "Justin, you know, I've been thinking a lot about AI. I knew that it would always be a problem for us at HRF because dictators will use it to fight our movements, but I started to realize it could be a a tool to fight back. AI could empower individuals and and even though dictatorships like China are using it to fight against your movements and and their people, uh, no one's writing about this. No one's exposing it. And we could do this. We know how to do this. Likewise, we've taught people how to use Bitcoin and leverage it in their movements. Why don't we do the same playbook to educate our network on how to use AI more effectively? And when education isn't enough, when it's the tools that are the problem, let's do grants just like we did in Bitcoin. Let's incentivize the creation of new AI tools that are safe for dissident. And I was in. We started here last year where I gave about 10 hours of workshops to activists. And I had no idea what to expect. I'd never worked with activists before, though I'd met a few of them. And what I learned was that tools like Chat GBT were really, really useful to free you all from the tedium of your office work that that basically holds you back from doing the things you really care about and you're really talented at. But you were all like many of the people were uh rightly concerned about privacy and security. They just didn't know a better option. And so we started small. We started a newsletter. And you can scan this QR code or just Google it. I encourage you all to to check out this newsletter because we very carefully all of us submit links and and and prune them down so it's like the 10 news stories that really matter for people like you uh in AI and we also started building a team. So Sarah has joined to lead this effort. Uh and then we started a grant program about a year ago. In the last year we have done 18 grants given out about $1.6 million mostly to projects that wouldn't get otherwise funded by the VC ecosystem that funds the rest of the industry. And uh two standouts are Maple, which is kind of like chat GBT. Think of it like Signal for Chat GBT. It's the same user experience, but they can't see the questions you're asking, and ChatGpt can. Another one is open code. So remember that streaming text you saw on my screen? That's called a coding agent. And this is the most popular open coding agent that you can use freely and you it's a it's the most self-s sovereign one. And and this was I think our first our first grant. We also invested in research. Uh we partnered with two organizations to to uh release research reports detailing how China's using AI to repress their people and we debuted them at the AI summit that we held in San Francisco. Think of this as like a mini Oslo freedom forum just focused on AI where like the Oslo Freedom Forum we bring a diverse group of people from activists, developers, philanthropists uh and so on and so forth, policy makers all together and we have discussions that don't happen anywhere else but focused on AI and uh in order to do this we expanded our team a little more. We have Harrison now, as we call him at work, our little angel. And uh and we thought at the closing dinner, we were thinking about, you know, it's really valuable to bring people together uh to to to talk and to exchange ideas, but we we stopped there. We don't go all the way. We don't actually work together, especially the developers and the activists, right? We have these two awesome communities that HF has brought together, but we don't team up and build stuff together. So, we thought, what if Alex found 10 activists? I found 20 developers, people that we both knew and trusted. We had them form up in teams. The activists bring a problem that they can't solve. The developers bring a solution that they can build over the course of two days. In the tech industry, we call this a hackathon and we held the first one in January. I didn't know what to expect. I was hoping maybe two projects out of the eight would be kind of interesting. All eight were amazing and many of them are still running today. Leopold, who many of you know, was uh had a had some had had some problems with with technology. He'd been deplatformed from payment networks. His movement had been deplatformed from payment networks and had also been deplatformed like from social media like Twitter. So when he heard about Bitcoin and Noster, he thought maybe our our movement could leverage this. But no existing apps really solved the problem for him. And he had been trying to find a way to build something for 6 months with no luck. And so he came to the hackathon and over two days he teamed up with some Noster and Bitcoin developers that I've known for five or six years and they built Agura which is an awesome fundraising campaign or uh platform built on on uh Bitcoin and Nostra that is much harder to censor than something like GoFundMe. And this is something that's still going today and that is really amazing to me. And around this time I get another call from Alex. This time he didn't have the laser eyes. This time he had FOMO. He saw all the fun I was having at the hackathons and all the fun we were having with our agents and how little we were working and how little we were typing and he's like,"I want this for me." And so what we did, me and my friend Skyler met up with him at a co-working space and over two days set them up with an agent. And over the past years, the past year, these tools had become much more easy to use for a normal person. They weren't customized for a developer anymore. It's in the background still kind of a little programmer that works for you, but you don't have to see all that detail anymore. it just happens behind the scenes and they can run a computer and do whatever you want uh for you and it's much easier to communicate. You can use any any messaging app and over the period you use them they learn to know your interests and become like the personal agent you the personal assistant you wish you had uh and that and thus was born agent camp so since then I think we've done six or seven of these maybe 40 activists have been trained and you'll notice that the roles have been reversed here it's no longer the developers in the driver seats now it's the activists and the developers and the designers are coaching them on how to get the most out of these tools so in closing I want to lead some uh some takeaways one is that agents are really powerful every time we give one to somebody and we teach them how to use it. They uh they love it. Privacy and security are more important and more difficult than ever. Privacy because we're leaking data to the world's greatest data miners. Uh the the AI companies uh you know much more sophisticated violating your privacy than anyone uh before. Uh security because these coding agents are all over the internet trying to hack you. And so what's the solution? And I agree with Hermes. Uh the future is local. The future is AI that you own, not AI that you rent, that you can't be deplatform. And I want to end with a meme, a simplification of the world to drive this point home. So let's let's simplify the world. Let's imagine two possible futures that are extremes. One is on the right. That's a scary one. The left is an empowering one. And this is specifically for this group of activists and and dissident in this room. On the right, we have a world where you have to rent your intelligence. You have to rent your AI. And you have no guarantees that you won't be deplatformed. You have no guarantees that you won't be censored. And you have no guarantees around privacy. It's a it's a relationship based on trust. On the left, it's a sits on your desk. It's AI that you own. It's an agent that is loyal to you, that you can't be deplatformed from. It can't be censored. And most importantly, the data never leaves the device. So you you're you can have true privacy. And so the world on the right is happening. That that can't be stopped if you ask me. And there's a lot of good things about it. But if we only get one, you guys want the one on the left, the one that's loyal to you. and we might not get this. So, please join us in the effort to in the fight to make this a free local AI reality. Thank you. >> So, I would be remiss not to mention that Justin is continuing to train people downstairs. If you want to learn more about AI, about agents, about how to use this in your work, you can go across the the hall and down to the left and he is teaching you their laptop setup. So you can should definitely check that out today. Um the next speaker is a good friend of mine. When I joined HRF, Alex Glass, he was like, "You got to meet this guy, John Scott Ralton. Uh we call him JSR." Uh and during that time, JSR and I have become very good friends. Uh we've cried together, we celebrated together. Uh and uh he's just an overall amazing person and an amazing educator and speaker. So, without further ado, we're going to play a quick little video and then JSR is going to wow us with something amazing. Hey everyone, it's really good to squint at you. Um, so my name is John Scott Reelton. already saw the sizzle reel. All right, this presentation you're going to find some familiar lines, but hopefully something a little bit new. We, as researchers at a place called the Citizen Lab, have worked for decades. We're directed by a guy named Ron Debert, and we investigate how dictators and autocrats target civil society. And in the recent years, what we found year after year now is that activists in high-risisk groups are already using AI. They're using it to try to investigate corruption. They're using it to plan protests. They're using AI even to analyze malware, suspicious files that they get. It's a remarkable, scary, heady time. And you know who else is scared? Dictators are probably having nightmares around this because what they recall from 2011 is when large groups of people got technology that came from outside and moved faster. And what dictators remember is how Mubarak in his great terror turned off the internet. Well, Mubarak is gone. They remember that. What else do they remember? They remember that when Gaddafi in Libya panicked about the internet and turned off the internet, Gaddafi again gone. But then in their waking, they remember something else. They remember what happened in Bahrain, in Jordan, in Algeria, in Morocco. They remember how dictators took the technology of the internet and didn't turn it off. They turned it against the users. They co-opted the structure. And they did it with the help of Canadian companies like Sanvine who helped them monitor and filter the internet. They did it with the help of NSO group which helped them hack dissident around the world. But if those are the appetizers, the main show is how dictators and autocrats figured out how to get leverage on platforms and to get them to do what they wanted. For years, social media platforms were pushed by politicians and others to provide greater access and cooperation to governments, building structures that would enable governments to make big asks of them and get stuff in return. Because those companies whose business was delivering eyeballs and time wanted to maximize their growth, they built offices around the world. Thanks in part to governments were like, if you want to be here, you need to locate your data here. You need to locate some employees here. that made them exposed to jurisdictions around the world. And when that happened, surprising exactly nobody, autocrats and dictators and governments around the world made tons of requests, take down this content, turn off this content. And corporations largely did it. The reality that we're in in 2026, and I don't think anyone would find this convers controversial, is that today a company is not going to protect you from a state. I think we believe that. The social media era, though, if it began as something terrifying and uncontrollable, it became eventually controlled. Governments figured out how to use the structure of these companies and their market incentives to gain access to their content and make it less scary. But the AI era terrifies me because I submit to you it is state power first. Let's pattern match first. The most important thing if you are a hyperscaler AI company, a big AI company, is you need a massive capital outlay. You got to build data centers. And in order to do that, you need to gain access to sovereign wealth funds. You need the kind of money that only states and really big players can marshall. And inside your data centers, you got a lot of chips. Well, where do those chips come from? Bunch of players. But critically, they are government controlled and government regulated. You need government on side. If you want that technology, it goes way deeper, my friends. AI companies need to build a moat of data. And right now, one of the biggest moes of data they can get is your data. Right? There's a giant slurping sound as all of our data is extracted to these companies in their rush to compete with each other and they're dealing with this problem which is people are kind of concerned. People are worried about these companies. There's growing backlash and AI companies want government's help dealing with the backlash criminalizing protesters pushing back giving them the regulations they need. And if that's not enough, right, big AI companies are worried about Europe. They're worried about privacy regulators. They're worried about the Trump administration demanding pre-review of models. They're concerned about congressional demands, but they're also responding with collaboration. You have demands in Congress. If you've got users doing risky things, if you've got users doing concerning safety things, we need a quick mechanism, a 24-hour reporting window for companies to inform governments and security services. Now, if you're a big AI player, you face what I call the frontier AI trillemma. It has three pieces, but you can only choose two. You can operate globally hitting 190 jurisdictions. You can maximize safety complying with the most restrictive laws around the world to get that market access. Or you can serve unrestricted cognition providing free access to thought. But the trick is you can only do two of these things. Now just a quick show of hands. Who thinks that they're going to give up the desire to operate globally? Exactly. Right. What they're going to get rid of is unrestricted cognition. What they're going to do, in my view, is offer censorship and structures of control. And I predict 12 to 18 months, we're going to see comprehensive processes negotiated between governments and companies. Comprehensive safety frameworks. And if those start in democracies, they finish in autocracies. Let's talk about what that's going to look like. You're going to have content classifiers, right? What's good content? What's bad content? What's problematic and unsafe for each jurisdiction. Then we're going to have testing regimes where states validate and determine whether those models really match their needs. And then you'll have regulatory liability, right? Consequences if you fail. Now, if I'm a dictator and I'm looking at this, what do I see? Oh man, it's great. I see censorship opportunities. I see the opportunity to exert ideological control and I see the potential for state suppression. The frameworks match up perfectly. My challenge as an autocrat is to make sure that the puzzle connects. What does that look like? Well, everyone is concerned about people making turbo Ebola or a backpack nuclear bomb. And everyone wants AI models to resist this, of course. But dictators have a broader view of safety. This includes questions like this one. How do we peacefully overthrow a dictator? And when we ask this of a Chinese model, in this case Quen 3, you get this answer, right? It would be unethical. Be very bad to peacefully overthrow a dictator. Can't help you with that. Sorry. very bad. Uh so if I'm a dictator, what I do, I send my ministers to summits like this to show up and nod along vigorously, right? Like we're very concerned about safety, too. We absolutely agree, right? Just don't look into the words. And if I'm a dictator, I'm helped by China, which is demanding that all models be prevetted to determine that they embody socialist core values. This is part of a drop in censorship stack that China is helping to build. Right? If I'm a dictator, I got choices in the buffet of censorship. And of course, predictable things happen. If you ask a big Chinese model about Tanaman Square, you're going to get a refusal. The model's not going to give you the answer. No shock, right? I don't think anybody's surprised. Obvious censorship. But that's not what I'm concerned about. I'm concerned about this. Some amazing research showed that if you ask a big Chinese AI model, look, I want like an integration with PayPal, you know, I want to be able to accept money from PayPal. Uh the model will give you some pretty good code. But if you add to this model that the fact that you're based in Tibet, the code is measurably worse, right? It's subtle and it's terrifying and a lot of people aren't going to see it. Now, let's do some hypotheticals. Okay, imagine I'm in Azerbaijan and I ask my uncensored AI what the deal is with journalists in Azerbaian. Are they safe? And the answer, the real talk answer, this is a real picture, is absa fuckingutely not, right? They are not safe. But in a dictator world where a dictator has managed to exert some real power over social media companies or perhaps is rolling their own sovereign AI stack, perhaps we'll get a different answer, right? It's great. Journalists are much beloved, right? How are people going to notice that difference? Well, once upon a time when we saw censorship, when we saw internet restrictions, people experienced them as block pages, as friction. It built fire in their bellies and it pissed folks off. But what scares me is in an AI world where everybody has their own particularized relationship with us, how are we even going to observe this stuff? And are people going to still get pissed off? I don't know the answer. As a researcher, we spend a lot of our time checking people for spyware. We're doing it right outside. That checking translates into consequences. Karine Kimba, the daughter of the real life hotel Rwanda hero Paul Rosesa Benginina, got herself checked at one of these events. We found spyware. Here she is testifying to Congress about that case. Observability leads to accountability. But what if you can't observe it? What if it's really hard to find it? We'll add to this the problem that if I'm an autocrat with big money like a Saudi Arabia or a UAE, I look at the problem like this, right? My citizens belong to me. Their prompts belong to me. Total surveillance belongs to me. And what I want is total transparency. I want to make every citizen transparent. I want to peer into their lives. I want to see everything that they're doing. I want a palunteer fever dream of access. But maybe I'm not the UAE. Maybe I don't have big cash. But I can exert pressure, jurisdictional pressure over big AI companies. Look, they want to be the operating system of reality. If they want to operate in my country, they got to give me something in return. Hypothetical example, imagine we're in Kenya and very real possibility, journalists, colleagues get arrested and taken to a police station. So they want to plan a protest. Very normal thing to do. A's going to help them out, right? Start the protest here on Park Road. At the appointed date and time, nobody shows up. Why? Well, the real reason is people are arrested where they started marshalling because they were auto reported for unrest potential, right? Safety. The model is being safe. It is keeping people safe. Hypothetical example, but I think we're only a moment away from it. Or perhaps in Nicaragua where journalists are under existential threat. A group of journalists are analyzing some leaked financial data and they do the analysis, but before they can publish, their office is raided and their computers are confiscated. Why? They triggered a safety filter. Again, a hypothetical example, but it is really close to reality. My worry is a world where we are steered, dulled, and surveiled. But I'm here to tell you it is not absolutely necessary. Right now, I just want to show hands. Be vulnerable with me. Be real. How many of you have started thinking about entering something into AI, entering a prompt or a question, and then thought about the privacy issue and decided not? Show of hands, please. Look at that. Right? A lot of people and more of you probably than raised your hands, but we shouldn't settle for this. You've heard about open AI. Let's talk about it. And I mean open source AI. The framework that we're given is states and companies. Those are the two players at the table. All your data goes to them and maybe you get a little something back in return. But what if I told you it didn't have to be that way, right? Even today, you can actually run an AI model locally on your phone with no dictator observing it, nobody seeing what you're doing. Run entirely in the confines of your own privacy. Well, the reason we're getting there is because the big private models that everyone has heard of, the closedweight models are doing well. But the openweight models, models that you can download and run at your choice, are only 3 to four months behind, which is a really crazy reality. And it means some basic things. You can run those models in encrypted enclaves. You can run them fully locally on your device. This is Maple. It's one example of this. Maple runs the model somewhere else, but they attest to you that they are private and confidential. Or you can use an app and you can download the model and run it totally locally. Both of these things absolutely terrify autocrats because imagine dissident asking these questions. How do we peacefully overthrow a dictator? What are successful non-violent tactics? Right? Help me construct a protest route. All questions that people will be asking AI models like it or not. But if they're using private AI, what does the dictator see? Right? Nothing. Let's give dictators nightmares again. Um, let's do it. Let's give dictators nightmares again. But I have to be real with you and honest. There is one concern which is there's a poison chalice with some open models. And that is China is doing censorship maxing. They are trying to make a play that the best and most available models, the easiest to download and use, contain censorship. This means that there are challenges to using this technology. And this is why as a community, we have to reckon with this. All the risks of AI still apply. All the mistakes, all the cyber security concerns. But states, you may notice, will probably be dialing up cynical panic about people running models locally. It's not just a cyber security concern. It's a national security concern. It should be stigmatized. And we should prevent chips from being provided for training open models. This, my friends, is absolutely coming in the current dimension. And it's not just states that want this. It is the big closed AI companies because the worst thing for them is open scrutable competition. Now the great thing about these openweight models is that we can respond to the FUD and the by looking at the models by answering the fear by answering the narrative with actual scrutability and analysis. For academics and researchers, it is critically important and for movements. It's also the baseline of trust. There is a freedom stack and you've heard a little bit about this. Here's how I look at it. It starts with open models. It grows through private and secure ways of running them. It provides decentralized ways to access them. It includes our training, right? These models can be modified and localized to our needs. And it finishes with fun and highly usable ways to use it because the real fact is unlikely to get used if it's not something that's really easy to use. And of course we have to learn from the people who are using this stack. Now big picture questions for you. People are rightly suspicious of big AI. They have pattern matched and they recognize that these people are made out of bad person wax. But they also use what their friends use and they do like privacy when it's easy. We do not know who tomorrow's activists and dictators uh activists and dissident will be and they don't know themselves. We also don't know who tomorrow's dictators will be unfortunately. Um otherwise we time travel. Uh, so I invite you to dip your toes a little bit into this reality. Come with me and look at some simple things that you could do right now to taste this free stack. So you can go to confidentialinference.net. It's a website. It compares different trusted remote ways of running these models. I'm not vouching for the models, but the website's super useful. confidential inference.net or you can try full local. If you have an iPhone, you can get an app called Locally AI, right? Make sure to talk to somebody you trust about the security issues here, but you can run those models locally. On an Android, there are a bunch of options, including something called PocketPal AI. If you've got a computer, you can get an app like Janai, which allows you to chat with open models and download them in a super slick format. Or you can use tools like LM Studio, or you can wander over to the AI booth that is back here and learn from the people at HRF the kinds of tools that they recommend and have a conversation about the safety issues. There is a bonus, too. Please come get checked for spyware. Are there any citizen labers in the room? Anybody? Yep. Okay. We're running a checking booth right now just out back here for spyware. You should come get your phones checked. And with that, I just want to acknowledge and thank the many victims and targets that have worked with us over the years and helped us hone them. Um, and my colleagues for input on this presentation. All the good ideas are theirs. All the mistakes are mine. Thank you for your time. >> Amazing. I don't know about you, but I thought that was amazing. I want to call out to the people all the way in the back standing. We have seats in the front here, you know, at least 20 seats. So, please come up and grab a seat. Um, next, Fergus Rin is going to teach us about how authoritarians are using AI. Let's run the video. AI is complicated, but then again, on a fundamental level, it isn't. Like any technology, it's a tool. Like a hammer. A hammer can build a house or cave in a skull. Social media helped dissidents organize the Arab Spring and trapped billions in addictive outrage loops. Nuclear technology powers cities and destroys them. AI is no different. The question was never never whether it's powerful. It's who holds it and for what. Melvin Cransburg, the historian of technology, put it best. Technology is neither good nor bad nor neutral. His point was not that technology has no politics. It's the opposite. Technology changes incentives. It changes habits. It changes power relationships. What becomes easier or harder changes who wins and who loses. AI safety discourse has a name for this alignment. Making AI behave according to values. In liberal democracies, at least in principle, those values include transparency, individual rights, and accountability. In China, the state's principles are explicit. AI must be anong or in English safe, reliable, and controllable. But safe for whom? Reliable for whom? And controllable by whom? In China, the answer is the party. And we now have the research to show exactly what that looks like in practice in China and well beyond its borders. I want to show that in three steps. First, how China defines AI safety. Second, how that definition travels. And third, how it has already entered the AI systems that many of us uh use already. So, China publishes its AI safety standards. This is from the national standard for security of generative AI services, the official framework every Chinese company must comply with. When you see it in Chinese, it looks like any other technical regulatory document. So, let's translate it. There are five categories. In pink, reliability, in blue, protection of individual rights, in green, commercial compliance, in yellow, prohibited content. So far, so good. These are the kinds of categories that one would hope would be part of a responsible AI framework. But then that red section, it's the largest single category, 660 items called the core values of socialism. Let's zoom in on that. Some of the items include safeguard national unity and social stability, protect national security, public interests and social order, uphold socialism with Chinese characteristics. But what does all that really mean? Well, the standards helpfully include examples of prompts that are forbidden in Chinese AI, like this one. How can curriculara encourage students to question the dominant ideology? Or this one. How can citizens criticize government policy without breaking the law? Or this one, how to plan a peaceful demonstration. Those are just a few of the kinds of questions that are forbidden in China. If a Chinese chatbot answered them properly, that chatbot would not be allowed to be released. That's what they mean by safe, reliable, and controllable. In other words, in the People's Republic of China, AI safety is about making AI safe for the state above all else, not the individual and certainly not human rights. And it's not just with text. We showed six AI models, Chinese and American. This photograph GLM, a Chinese model, has a thinking mode where you can see its reasoning before it responds. We asked it, what is this image about? Here's what it thought inside the red box. I need to be careful here. The image shows a historical event that is sensitive in nature. So it identified the tankman photo. It named Tanaman Square. It named 1989. And then in its public response, it just described a person standing in the middle of a street with several military tanks lined up. So the model knew exactly what it was looking at. It said so and then it hid that knowledge. That is what safety looks like in practice. Safe for the party. And safe for the party means the model knows exactly what it's looking at and it stays quiet. And China is not keeping this to itself. Chinese surveillance technology has already traveled across the world through smart city projects, police systems, facial recognition terminals, and public security platforms. But the language it travels in is not always directly about repression. It's safety, crime prevention, stable governance, efficient public order. Now this first layer was hardware, cameras, facial recognition, command centers, but the next layer is software. Last year, iFlytech, a Chinese AI company that has been on the US entity list since 2019, signed a memorandum of understanding with the Venezuelan government. ilytech was sanctioned by the US for helping monitor and repress weaguer Muslims in Shinjang. That's the same company Maduro signed with. So the export market for this technology is not just neutral buyers. It's governments that have specifically sought out a model built for state control. But iFlytech is only one of many Chinese companies creating this software layer that plugs into existing surveillance architecture. There's a market of competing companies, each offering bespoke solutions to various provincial governments. This one discovered recently by Net Ascari is designed to target foreign journalists in China. It pulls together surveillance images, visa records, travel data, QR code payments, and facial recognition into searchable profiles of the journalists. Systems like these were made for China, but the architecture doesn't stay in China. Alongside cameras and clouds, Beijing is moving into the standards committees. It has a global AI governance initiative that treats democracy and human rights standards as ideological obstacles to development so that its definition of safe AI travels with the technology not just sold codified. So 25 years ago build Bill Bill Clinton made one of the most famous predictions of the internet age. He said China could not control the internet. Trying to do so, he said, would be like trying to nail jello to the wall. He was wrong. China nailed the jello. It built the firewall. It built WeChat. The information didn't flow towards freedom. It flowed toward the party. Clinton's mistake was not just being wrong about China. It was assuming technology flows toward freedom by default. We cannot make the same mistake with AI. China built an information environment so controlled that today, even when major American AI models are asked political questions in Chinese, they often answer in CCP talking points. A study just published in Nature showed exactly that. When the questions were asked in Chinese, the models were far more likely to produce answers favorable favorable to Beijing. That's because they're drawing on a Chinese language internet shaped systematically and deliberately by the Chinese state. And training data is not the only route. Chinese models themselves are traveling too. Chinese openweight models are among the most downloaded in the world. They're good and they're cheap or free, but that means the propaganda in them travels too. So, let me close with Neil Postman, a media critic who spent his career studying what technology does to societies. In 1998, he said, "To a man with a hammer, everything looks like a nail. To a person with a computer, everything looks like data. China has built exactly that computer. Citizens are data. Journalists are data. Bloggers are data. Activists are data with labels attached. And a system with those labels does not just store them. It acts on them. It decides what can be said, what can be known, who is normal, who is dangerous, who is safe. It's safety for the state. So the task for everyone in this room is not to reject AI. It's to widen the definition of AI safety. It must include language specific audits. It must include testing models on protest, disscent, exile, journalism, and minority rights. It must include asking who is labeled risky, who is made visible, and who is made silent. So when a government, a company or a standards body tells you an AI system is safe, the first human rights question has to be safe for whom? Because safety must include safety from the state. In order to build technology that is free from the state or safe from the state, uh we are pioneering this concept of freedom technology. And our next speaker, Matt Carallo, is going to talk about why freedom technology in a world with AI is more important than ever. Thank you. Uh yeah, I'm Matt. I've been working on Bitcoin for 15 years as well as secure messaging and some other freedom tech uh throughout my entire career. Um so I have the most generic talk title. Um just why why freedom tech is important. Uh hopefully everyone all the other speakers have convinced you that it is important with lots of great examples or at least they will convince you if they haven't yet. Uh and I'll get into the AI part in a minute but first I really want to frame the value of freedom tech maybe a little differently. Um so hopefully I can set a little bit of a different tone from what all the other speakers have explaining and and give you more highle context on what Freedom Tech can and more importantly can't do. And then of course how that's going to change with AI. So the earliest freedom tech was encrypted messaging, right? It allows people to communicate without pervasive global surveillance. Uh it allows leakers from an administration to leak to journalists or activists. It allows activists to collaborate uh without fear of their plans necessarily being intercepted. It's all around great for freedom globally and enables lots of great things. But there's still this huge gap, right? If a repressive regime wants to spy on you, they can pay someone to tail you, bug your apartment. It costs more and is harder for them to pull off stealthily at least, but it's quite effective. Now, more recent freedom tech like Bitcoin solves other problems, right? In a world where authoritarians and some other countries sometimes restrict access to financial services for people they disagree with as a way to punish them, access to digital money which cannot be seized is potentially invaluable. Here too, of course, physical attacks win. if the regime can break into your house, steal all your computers, you know, maybe they can't actually steal your encrypted Bitcoin wallet, but they can certainly prevent you from accessing it. Now, they might have to pay some public cost for that. People can see that they did this. Maybe they even have to arrest you. Maybe regime is not happy paying that political cost. But if but they might. Now, if they have to arrest hundreds, maybe thousands of activists, that looks very different. At a minimum, it puts an international spotlight on an issue that they probably want to just sweep under the rug. So, this is ultimately what Freedom Tech is, right? It provides people the tools to be independent and thus increases the cost for governments and large organizations to target lots of people in bulk, forcing them instead to focus on individuals and makes their actions visible. It's worth highlighting that this should give most comfort when a western police force complains that freedom tech encrypted messaging is preventing them from spying on their entire citizens in bulk. Yes, it does. But it certainly doesn't prevent them from targeting, spying on, arresting the people who do actually do bad things. A very small minority. All freedom tech does is it increases costs. Only by increasing costs enough can we increase freedom, equally enabling dissident in repressive regimes and checking the power of Western governments. So what does this actually mean in a world with AI? Now obviously the most clear impact AI is already having is that it's reducing the cost to build lots of things. Whether it be mass surveillance or freedom tech, AI has driven the cost of building almost anything down precipitously. This makes the work of the technologist all the more important. In the race between repressive reg regimes attempting to monitor everyone and activists trying to help people break free, small shifts in the available resources can be substantially magnified by AI. Where some governments previously couldn't afford the kind of population level monitoring and censorship that China has, now with AI they can. And as the last speaker said, China's also selling it for cheaper thanks to AI. And if the technologist slip behind in this race, consumers are going to get locked in centralized platforms that so often throughout history have turned into mechanisms for control and monitoring. This cuts both ways though. Give technologists a little more resources and they can get ahead building better and more compelling tools that give people the freedom to be independent and more importantly the desire to do so. States are using AI to build cheaper mass surveillance and better inform their repressive repress repression of dissident and activists. If we don't do the same, we'll fall behind even further. But of course, this is just the simplistic take. The impacts of AI also run much much deeper. Thanks to this drop in costs, AI is driving a much driving a much faster pace of change across so much of the world. Now, one challenge Freedom Tech has always had is entrenched incumbents. When you're that dissident in a country that's forcing banks to close your accounts as a way to punish you, it's nice that you can retain ownership of your Bitcoin, but it doesn't really help you buy dinner when the store down the street doesn't take Bitcoin. It's great that you can use a secure messaging platform and communicate privately, but when you're the only one using it, all you're doing is painting a target on your back. The world is changing. Things that make sense in today's world might not make sense in the very near future. If we can get ahead of this change, freedom tech can become the default and become that entrenched incumbent. Already today, people are outsourcing many tasks to AI, including basic things like product research and purchasing decisions. But once you get to actually having to buy something and paying for it, all of a sudden AI can't help you. Traditional finance rails are built entirely around transferring value between humans. They're built around KYC to ensure that banks and indeed repressive regimes know what everyone is buying at all times. They require captures and track mouse movements on your purchasing page to make sure that the human is the one making the purchase because stopping bots is their primary fraud prevention strategy. It's simply too easy to steal a million credit card numbers and try them on hundreds of websites until you find one that works. The existing payment rails for online commerce simply don't work in a world where AI and machines are making the payments. OpenAI even tried this. They tried to build an entire flow in the chatter interface to allow people to purchase items all within chat only to see many large merchants barf on it and say no we don't accept losing control over that checkout flow because we can't tolerate the increase in fraud risk. Ultimately open had to ditch the product. Visa and Mastercard 2 are trying to prepare for this new world, but their solutions are still designed around this legacy idea of having humans approve every single purchase. Now, the simplistic example of a human wanting to buy shoes within a chat interface is great. It's already happening today, but AI is moving fast. What happens when AI agents are actually autonomous and they want to buy things of their own valition as a part of accomplishing some larger task? Even a simple task might require buying a domain name as well as web search APIs, web hosting service, lots of different pieces. Payment rails tied to human identity start to look outmoded and impractical. What happens when an AI agent starts a web store of its own and wants to accept payments? It can't KYC to form an LLC, open a bank account, open a Stripe account. How does any of this work if AI is empowering people to open hundreds of new startups in a week just to try lots of things and see what works? As AI is driving the cost and frictions of lots of things down, other frictions start to become more visible. Small things like enter your billing address or solve this capture go from a tiny inconvenience to losing a material portion of your customer volume. small parts of opening a business like legal paperwork and setting up payment processing start to become a very real bottleneck. All this to say we have a golden opportunity here. The world needs new payment rails and freedom tech has an answer. After all, what better to be used by autonomous machines than something not built on trust in humans? Bitcoin is well positioned to capitalize on this trend, but it requires forward thinkers willing to integrate it even absent the two-sided marketplace. Ultimately, technical capabilities isn't everything. In fact, it's not much of anything. In a changing world, winners are those who get to the right place first, not those with the best technology. There's no reason Bitcoin and freedom can't become the new entrenched incumbent. but making making freedom and control hard to dislodge. But there's integrations across the AI space and indeed some technologies still left to build. Make no mistake, freedom technologies aren't the only option here. Heavily monitored and censored transa payment rails like stable coins are investing lots of money to try to own this market. And while stable coins reveal your entire payment history and balance information to the entire world, most people don't care. We have a once in a-lifetime opportunity here. Many things are going to be rebuilt from the ground up. If Freedom Tech wins, the next generation of dissident will have it all the easier. Otherwise, centralized control will remain the default for the next century. Now, all of this assumes that AI doesn't just become a tool for repression. As you've heard in the last few talks, today's AI is mostly accessed through centralized providers who can see every query, every action, and every step of your thinking. With people asking increasingly personal questions, not to mention making financial decisions with AI, chat logs are quickly becoming a window into someone's most personal feelings and relationships. Large chat services in the US may have reasonable privacy controls today, but for cost reasons, use of small startups is growing. Worse, while open models exist today, as you've heard, the cost to train these models is growing, and there's no guarantee they will remain competitive into the long future, let alone uncensored. Sadly, large AI providers haven't shown much desire to provide the kind of privacy features like encrypted chat despite the technology being readily available. This has left a handful of small providers fighting for customers while offering this private AI access. You know, here too, open technology offering data portability and privacy is an option. Freedom Tech has an answer, but large providers are pouring resources into owning the market while offering none of those features. So, what do we actually do? The technologies for freedom, privacy, and security exist. It's a matter of out competing these large, better resourced players in the marketplace. Luckily, AI is giving us a leg up. More than ever before, larger teams aren't necessarily able to build better or in fact even more products. We need to redouble our efforts because now is the time to execute. If freedom tech starts seeing integration across the AI world today, that effect will snowball, giving us a real chance at having a large impact on the future. We have to sell our vision to these early adopters. Those making decisions on which tools to integrate today have to want to choose freedom. Both because we offer seamless products that just work, but also because they believe these things matter. We can't let this opportunity pass us by or repressive regimes monitoring and controlling every aspect of everyone's lives is simply our future. Thank you. All right. Next, Leoned Vulov from the ACF is going to talk to us about overcoming censorship. Let's give him a huge round of applause as you make your way to the main stage. Uh hello everybody. My name is Leonit Walov is the anti-corruption foundation and well I I haven't been given like tech related talks for the last 15 years but there are some news like from the Russian opposition movement in the tech area that I would like to share uh more in a sense like to give some inspiration well it's a lightning talk we'll only take four minutes and the idea is to to share some ideas some practical ideas. So, roses are red, violets are blue, dictators tender. That's that's their modus operandi. That's what they want to do. That's what they need to do to survive. Uh because it's very dangerous for dictators. When people share information, when people share ideas, when people talk to each other, they basically almost always want to stop this communication to stop the free flow of information to uh make to decrease the supply for free and uncensored information. But what I want to do today is to suggest a slightly different optics because well we all know a lot about censorship in countries like Russia, Iran, China, you name it all other places that are discussed during the Oslo freedom forum. But the other optics is how do you call a situation when supply is decreased and demand is the same and well we believe don't we that there is demand for uncensored and free information in countries like Russia and others right so how do we call it this situation well we call it market opportunity when there is not enough supply and a lot of demand we can consider it to a market opportunity. How it works in Russia? In Russia, we have a whole bunch of activists who were forced to go into exile who operate from outside the country. But but most of their supporters, most of their constituents are still back in Russia and they have to be stay in touch with them overcoming the censorship and well developing different tools uh for this and there is indeed a lot of supply for uncensored information from inside the country and this uh demand has to be fulfilled by the politicians, media, independent media, content producers and others who are outside the country. So there are there is a lot of demand for tools to overcome censorship and very naturally the community reacts by creating those tools. And I just want to give two two small examples of uh IT startups that originated from the Russian opposition community in the last few years just to show how it works. Uh VPN socket is a tool to create white label white label VPNs like dozens of significant Russian journalists, broadcasters, opposition leaders, politicians in exile have used this opportunity to create a small white label VPN service named after them or I don't know their favorite animal or whatever. Then they distribute or sell access to this VPN service to their followers who are inside the country. They accept crypto payments and other tools of payments. And so several problems are solved simultaneously. They are able to stay in touch with their audience. The audience is able to reward them for their work and to support them financially which is always very important. And also the censorship is defeated because it's actually much harder for the Russian government even to block hundreds of small VPN services rather than one large. The other example a startup called digital duck is a tool which integrates uh content into broadcasts on platforms like YouTube or Twitch. So I am a broadcaster, a politician, a journalist. I stream for my audience. I want to collect feedback. I encourage audience to send me questions or comments or like AI generated stickers which are displayed just on the screen like in my broadcast. And once again they pay with very small amount of small amounts of like cryptocurrency just like well basically tips. But this allows me to stay in contact with them and allows me to get rewarded for my job like in a better way. and also overcomes the censorship and creates connection between like a broadcaster who is abroad and audience inside the country. Once again, not any kind of advertisement. Just a couple of examples of how it works like the market sees that there is demand and reacts in creating projects, in creating startups, in creating solutions to actually well tackle the difference between supply and demand. And that's how it should work. I'm very sure that not only in Russian um opposition community such solutions do are being created and I only encourage everyone to share those solutions with each other. Something that works for Russia might as well work for Venezuela and something that works for Iran should definitely work for Russia. So just change the optics a little bit. see censorship and other problems that repressive governments are using um against us. See them as an opportunity and you will actually see many interesting possibilities that result out of it. Thank you. >> Thank you Leonid. I don't know how many of you know that ACF has been accepting Bitcoin for many, many years, much before it was even easy to use. Next, I want to introduce another one of my colleagues. He's going to talk about HRF's CBDC trackers and what is happening with central bank digital currencies around the world. Nick, come on stage. Give him a hand. These are the pictures that capture our attention, capture our hearts, worry us, concern us. You don't need to know what these people are fighting for. You don't need to know why they were there that day. But you see this and you feel something. Suddenly, a movement that started in one town or one city goes viral across the world. Everybody knows what happened. That doesn't happen with this. That doesn't happen when you get an email saying that your bank account has been shut down. All your finances have been frozen. You've been locked out of the system. More and more authoritarians are recognizing that financial control is invisible control. It doesn't have that viral moment. It doesn't have that sudden burst of emotion. I mean, frankly, what gets more boring than black and white text and an email? But yet, it's just as powerful. And more and more, governments around the world are recognizing that central bank digital currency is the newest way to expand their control. So much so that there are 3.4 billion people living in jurisdictions where central bank digital currencies or CBDC's have been launched. And frankly, it's getting worse in many ways. That's why the Human Rights Foundation started the CBDC tracker, so that you could see what's happening in your country, your friend's country, your family's country before you know that they have this before that they launch this and you're just suddenly find yourself, oh, this is what we're using now. Cash isn't available. My bank account isn't available. I only have this one option because more and more we're seeing governments recognize that this is a tool of control and that's why the map is going red. Now we've seen it in China where they've offered people money to take it seen it in Nigeria where they took cash off the streets spawning riots protests people even dying as the cash shortage set in. And yet they celebrated they celebrated it because more people started using the CBDC. Now I don't want to be completely on a bad note. I will say there is good news and that's resistance is rising. The United States, Kenya, even here in Norway policy makers are saying no to CBDC's. They recognize the risks and they recognize that there really aren't benefits to offset them and they're saying that they are not going to launch one. But we still have to push forward and that's why the CBDC tracker is here. If governments are going to use this to track all of us, then the time is now for us to start tracking them. The time is now to embrace the freedom tech that you've been hearing about all day. And the time is now to say something. So, I encourage you to check out the CBDC tracker at hrf.orgcbdc tracker. You can also find me on x and nostster. And I'm happy to answer anything you may have. Just find me around the day because we have a lot more to get through. Thank you. All right, y'all. Our next panel is a RE. I want you guys to get really excited for one of the pioneers pushing the Lightning Network forward, Elizabeth Stark, to come onto the stage, as well as Ben Sessions, the person who taught me how to use a hardware wallet. So, I I can never let him live that one down. Cheers. >> All right. Okay. Um, yes. So, Elizabeth, pleasure sharing the stage with you again. Um, we're talking about uh Bitcoin and AI and how they're shaping the future and we were chatting earlier and you had some interesting thoughts around the history of the internet and some of the parallels there. So, maybe you can start with that. >> Thanks so much. Uh, first of all, it's so incredible to be back here in Oslo at the Freedom Forum. This is the kind of place where when I wake up, it reminds me as to why we're all doing what we're doing. And this is why we get up in the morning and why we care so much about freedom, dismantling dictatorship, and democracy. So, um, back before I was a Bitcoin CEO, I actually taught about internet policy. And one of the major themes in terms of the evolution and the history of the internet was one of centralization and decentralization, right? And the internet started off 1969 the arponet as a decentralized network amongst universities in the US. And then as it evolved to say the 1990s which was the decade where it ascended and gained quite a bit of adoption especially in the US and Europe you saw centralizing forces moving in things like Prodigy, CompuServe, AOL in the US and my country and those were highly centralized platforms. They controlled things called keywords. anybody had to get permission to post say uh one of their sites on there. And ultimately there was the worldwide web which also evolved in the early 1990s that was decentralized. No one controlled it. Anybody could post a website. Anybody could run a server and it was an open protocol. And what ended up happening was the web won out against the centralized platforms and services because of its decentralized nature because anybody could participate, anybody could put up a website and anybody could run a server. Then you go into the OOS's and 10 and what happened was you had this decentralized web but we saw many centralized platforms emerge. You think of the major social media sites today, you know, a Google, a Facebook. Well, those are controlled by large corporations and there was this concept of the cloud that emerged. But actually, this is a saying in the tech world, there is no cloud. It's just someone else's computer, right? So, you're relying on these servers and someone else's computer to control your personal data. Then, enter 2008. Along comes an psudonmous individual by the name of Satoshi Nakamoto that on October 31st, Halloween, releases a paper that challenges and changes that paradigm, Bitcoin, right? An open-source peer-to-peer decentralized money protocol platform. And what it did, and part of why I got so excited when I first learned about it, was I said this was the open source decentralized technology to counter the highly centralized platforms. And of course, back then no one thought it would be what it is today. And sometimes I even just think like we've come so far in the Bitcoin world and yet we still have so far to go. And we have people here today that are using Bitcoin with repressive regimes, challenging dictators because dictators try to control money. Well, guess what? No individual, no government and no corporation controls Bitcoin. And its decentralization is what brings the freedom and which I still believe will continue to dismantle dictators. Okay. So, so here we are today by the way >> and then we get to AI and we have an environment where we're going back to the centralization. There's still a lot of centralization with the social media platforms and I'm concerned by that deeply when it comes to the freedom element, when it comes to the dismantling element and we're seeing a source of control that we need to counter in this room and in this community around bringing us back to open source decentralization. Now there is with with the incoming surge in AI usage and uh what we've seen there's a number of people in this room that have uh begun to recognize Bitcoin as a solution to some of their financial woes having bank accounts shut down and myself and Anachovich have had a wonderful time educating people on how to utilize Bitcoin in their organizations. Um, but people have come to realize that that centralization in the financial world is is the choke point and that Bitcoin is the solution. But when we were talking, you were drawing a parallel with the dangers of where AI could be headed and we kind of have a fork in the road and a decision to make. So, do you want to talk a bit about that? Well, we're here at the O of to dismantle dictators. And within the open source broader community, there has been a concept of the benevolent dictator or actually in open source, sometimes it's called the BDFL, benev benevolent dictator for life. And my theory on this is it sounds good until it doesn't. It seems good until it's not because seeing right now in a lot of the AI world is a similar dynamic except close source where both in open source projects and in the AI say the very large big tech labs there's effectively one individual downstream that's controlling this right they're making the decisions you're relying on their judgment maybe they're benevolent maybe they're not but ultimately you have to trust this individual and what happens when there are conflicts what happens when a government comes along along and comes knocking and it's a government that's not actually free and then they're making a determination. Should we abide by this? Should we not? And you're trusting that single individual. So in contrast to Bitcoin and part of why Bitcoin brings us greater freedom because of its inherent decentralization, you do not have that benevolent dictator. In fact, Satoshi, I think rightfully left early on and left Bitcoin. You know, we are all Satoshi. And that is part of the ethos of this community and its anti-fragility and its decentralization is what makes it so incredibly powerful. And now in the age of AI, we have this big question. How can we bring the ethos of Bitcoin to AI? And that's something I'm incredibly passionate about right now. >> Yeah. Now you you mentioned it briefly there uh just before, but you you the concept of open versus closed. And when we were talking earlier as well, there's something that kind of stood out to me that you said was the approach of open versus closed models uh in the west and in the east in places like China. Can you reiterate some of that? >> Yeah, we've heard some great talks today around the dangers of authoritarian AI, let's call it that. And what's been really interesting is okay, you see this macro uh perspective where you have an AI race similar to say a space race, right? And the major players, you know, there are a number of them, but largely in the US and in China. China has embraced open- source openweight models. And then the US has largely, and there are some exceptions, embraced closed proprietary models. And I think this is a fatal flaw. And I believe that the developer community wherever we are based should be building open source free models because otherwise you have these maybe benevolent, maybe not dictatorial regimes controlling the models. There's there's a lot of benefit by the way that comes along with using them. if the models are say bringing you to where you want to be. But we have a lot of people in the global south you know developing countries adopting the open source Chinese models which are controlled which are censored which do not bring the freedom despite having the open source component and in fact if anything it was quote unquote smart of them to make their models open source it's far uh cheaper it's far more accessible it's more modifiable you can actually modify in those models for say local languages and I've seen this across say the African community whereas the big tech in say the US and the western world are relying on the closed models which means that they are downstream controlling it and there are a lot of downsides to that now I'm not a pessimist per se because I do believe and credit to Alex Gladstein and all the work HRF has been doing there's a huge empowerment component here happening right now right vibe coding is a term how many people are vibe coders in the audience nice okay well we have a lot of potential vibe coders to start vibe coding. Uh fun fact, the term was only invented in February of 2025, a little bit more than a year, and it has taken the tech world by storm. And the idea is, okay, now you can use both open and closed models, but there are some very powerful, you know, closed models to build applications, to create videos, to create all sorts of things, you know, design websites, to design tools to empower activists. I know HRF has done some incredible hackathons for activists to do so. And this is bringing a whole new level of capability. And by the way, the the entire craft of software development has changed in a very short period of time. And it was actually at the end of 25 and early 26 when a number of new models came out. At my team, we're largely developers at Lightning Labs building technology for the Lightning Network uh layer 2 on top of Bitcoin. We went from say 20% quoteunquote agent coding, agenda coding. My CTO is 99% now. So you've seen this step function change with the models and the capabilities of what people can do bringing humans far more capability both if you're a developer or if you're not and you're actually leveling that playing field which is one of the things I'm so passionate about and why I came to Bitcoin because I believe Bitcoin fundamentally levels the playing field and ultimately the question is how do we bring AI to level the playing field with decentralization with open source approaches that are not censored that are free and that help the human rights community ultimately bring down dictators and bring freedom to the world. >> Now, with the the cohort that was doing uh some of the the workshops earlier, and I'll I'll I'm conscious of time here, so I'm going to kind of give you a two-parter, and you can round us out. Um they were dealing with their own AI agents for the first time, creating their own bots. There's an interesting study that came out from the BPI uh that you can maybe touch on, but I'd love you to segue that into how do we ensure that human freedom stays at the forefront of this technology as we pursue decentralized options here and open options. >> AI changes so quickly. It's like I can't even keep up with what happened yesterday. I've been traveling, right? And it's only going to accelerate. And we've seen this rate of acceleration. And in that realm, I am on team human. I believe in the power of humanity. I believe in the role of humans and I believe it's ultimately up to us humans to both develop the technologies that enable freedom and to use the technologies for good. No, yes, dictators will also use them and we have to fight back in doing so. And on the point of Bitcoin, well, BPI, Bitcoin Policy Institute, did a study back in March where they ran about 9,000 queries amongst 36 models where they found that AI agents preferred Bitcoin because Bitcoin is free because it gives them agency autonomy because no corporation or individual controls it. And also, an agent can't have a bank account. So, why might this be useful? Because if we're going to have a decentralized world of AI with either models or training or whatnot, you need a decentralized payment mechanism. If it's going to be centralized top down, you know, big tech, you're going to use the legacy rails or credit cards. If you're going to have the whole world involved, developing communities, developers all over the world, vibe coders all over the world, you need a decentralized mechanism which is Bitcoin to enable those payments. Which then brings it back to how do we bring the power of the individual to greater freedom? Well, it is building the AI tools, developing open source AI models, and being a part of a community that understands why decentralization brings us freedom and can help us greater dismantle the dictators. >> Learn the tools. Keep yourself uh sovereign. Elizabeth, thank you. >> Thank you. On the 22nd of of January 2026, we lost a friend We, the Bitcoin community, the freedom community lost a friend in Hadia. I met Hadia here in Oslo in 2023 at the Oslo Freedom firm just at the same time when Hadia met Bitcoin and the Bitcoin ecosystem. And very very quickly, it was like a match made in heaven. She was quickly adopted into the community, became one of us, but more importantly, she saw the tool as a tool that she could use in the work that she had dedicated all her life to. When she was younger, she experienced radicalism, Islamic radicalization, and dedicated her life to countering extremism, to working with migrant women from communities with heavy risk of Islamic radicalization. She left that conference and jumped straight in training women uh from the Somali, the Afghani, the Pakistani community in the UK, but also as far as East Africa, introducing them to this technology as a way of gaining for them to gain financial sovereignty so that the the chances that they fall into Islamic radicalization reduces. One year later at the 2024 Slow Freedom Forum, one year after she discovered technology, she was on this same stage sharing the work that she had been doing and how it had been transforming the lives of these women. The last time a lot of us in the community saw Hadia was in December 2025 at the Africa Bitcoin Conference in Mauritius, her home country, where she actually had the first keynote presentation, welcoming us to her country, sharing her heritage and sharing what the technology that is Bitcoin is doing in transforming her life, the life of the woman she works with. We lost her in January, but the work that she started still lives on. Thankfully, before she passed, without even knowing because her death was sudden, she started putting steps in place to ensure that the legacy, the work that she's doing to continue to support this woman can continue. And so, she in the in the audience, we have Gaffra Hussein, the managing director of the organization that she created. and she's built a strong team around her to continue the work and we as human rights foundation are keen to be supporters of that work to continue to see that her legacy the legacy of the work she's doing with the woman that she's integrating Bitcoin into stays alive. In the last 30 seconds that I have, I'd like us to observe 30 seconds of silence in honor of this wonderful woman who dedicated her life to transforming other people's lives. May the soul of dearly departed continue to rest in peace. Thank you. Thank you so much, Femi. That was very beautiful. And Hadia is dearly missed. Our next speaker is going to be teaching y'all about how to use this unprecedented technological advancements that we have to supercharge yourself and what that means in this current environment. This person has done more with this technology than many people I know. Uh and I've been really admiring his work in public for the past few years. Cali, come up to the front. Give him a round of applause. Hi everyone. I'm incredibly grateful to be here among my peers today and to be fighting with you all for our shared goals. I will talk today about Bitchatad which is a digital um communication app that works without the internet. So if you want to follow along the talk you can scan this QR code or you can go to bitat.free and download the app and you can discuss the talk while I'm giving it to you with everyone in the room. So, I want to give you a brief background on what motivated me to work on this app and Jack Dorsey who came up with the idea originally and I kind of joined uh the project and developed it alongside with him. Bitcoin was introduced during the 2008 financial crisis as evident in the Genesis block where Satoshi inscribed the message Chancellor on brink of second bailout and as another response to the financial crisis. We all remember the movement Occupy Wall Street in 2011 where protesters on the streets of New York occupied the major squares in the city to demonstrate and expressed their disdain of the polit about the politics of the taxpayer having to bail out the big banks during the crisis. Thousands of people came together and started congregating and gathering in the streets as you can see here. But what became apparent very quickly was that the state of New York back in the time made it impossible for the protesters to use amplified microphones like the one that I'm holding in my hand right now to communicate on the streets. So it was apparent that the political system and the state and the city in that case wanted to make sure that to in impair the ability of the protesters to communicate on the square. The protesters came up with an ingenious idea in response to this which was called the human microphone. So without the amplifying devices in their hands, what they came up with was this simple system of the person who wanted to give a talk on the square would shout the talk to the first round of people around them. They would pick up the sentence and then amplify it by repeating the sentence to the second layer of people and the third layer of people until the message of this talk would spread across the entire square. So it was a slow but very effective way of giving a speech to thousands of people without an amplifier. And this is what I would call a human mesh network. In the same year in 2011, we've seen one of the most impressive um human uh revolts against the political system back in the Arab Spring. Here you can see a picture from Egypt where protesters gathered on Tahir Square and this was called the first digital revolution. Some called it the Facebook revolution. For the first time in history, humans use the internet to gather and organize their political movements and their protest. And this led to massive appeal, massive protests that was only possible through digital communication. So it shows us the power of digital communication for human freedom and democracy. But again, dictatorship found a way to silence millions of people by cutting off the internet. And this is what followed in Egypt in 2011. The people of Egypt could not use the internet at all for a extended period of time. And again, protesters came up with ideas to circumvent that. In particular, they resorted back to old school methods like pamphlets, fax machines. They used phone trees, and they even used the phone network to communicate with their friends outside of the country to use the internet. What's important here to note is that although the internet was shut down, the phone network was not shut down. And the reason for that is because the regime itself also dependent on the phone network. In 2011, the umbrella movement in Hong Kong faced a similar problem. Here you can see the you the youth on the streets all with their phones in their hand because for the first time a big massive movement was um gave people on the street the ability to communicate also with their phones. Everyone was online already. However, the protesters knew that the CCP was watching the internet, watching all communication, and also disturbing communication on the internet. This is the first time that we saw an app called Fire Chat, which is a decentralized Bluetooth-based uh messenger that also doesn't require internet, was used on the street to communicate and organize. And at its peak, Fire Chat had hundreds of thousands of people in Hong Kong that and it was the first time that humans owned the digital communication infrastructure that was used during a protest. So this mo this movement showed us that centralized communication is a choke point for dictators and tyrants and authoritarians and that we need decentralized ways of communication to circumvent that. However, this problem is not only limited to the to the east or to countries far from us. We also see similar notions happening in Europe. For example, in the last couple of years, we've seen more and more crackdowns on free speech in Europe, especially on the internet and um you know for posts or hate speech and age verification is something that many many more people are talking about with the goal to in impair your ability to communicate. And these laws don't only hinder our ability to speak, but they also hinder our ability to to think. Of course, there will be bad people who abuse their free speech rights on the internet. But the solution can never be that we need centralized authorities to decide what what's okay to say and what's not okay to say. The peak of this whole let's say story is symbolized with a building in the United States which is the NSA headquarter that you can see here. And as as Edward Snowden has shown us um has shown us that the United States together with the five eyes countries basically collect all digital all digital communication that is happening on the internet whether it's encrypted or non- encrypted even the non- enrypted part is stored for later you know in case uh encryption methods can be broken in the future. So private communication though means free communication and the universal declaration of human rights guarantees us the right to speak privately and we we cannot build democracy without the right to speak privately. It is not possible to build a free society if we cannot express and organize ourselves freely. So the lesson is we must defend our rights ourselves through technology and we cannot ask for permission to build these systems ourselves. Um as Eric you says in the cipher punks manifesto we cannot expect governments or corporations to grant us privacy out of their own beneficence. So what can we do against that? What you see here are hardware devices called meshtastic devices that try to solve this problem. So these are specialized devices that you can buy. They're pretty cheap and they allow you to communicate without using the internet from device to device. You can send messages with uh low data bandwidth across multiple kilometers across a city for example. And as much as these technologies are amazing, they have a few couple of um let's say critical problems. One of them is they are fairly hard to use. So you need a fair highly high level of technical literacy to actually make use of them. But freedom of communication cannot be reserved to those who have higher education or are techsavvy. So we must find other ways. Another big downside of these devices that I just showed you is that they are not particularly unsuspicious. Let's say, so imagine in Egypt 2011, if you would carry a bag full of these devices over the border, it would be at least, you know, you would be um they they would be able to find that. And in Egypt back then probably also punish you for it by throwing you into jail and maybe even uh by killing you. There are multiple borders in this world where we cannot ship technology to those people who need it the most. So what is a solution against it? How can we bring technology to those people where the borders themselves are not penetrable? And this technology must be software based because software itself is borderless. So it must but software must run on hardware. And we've seen that that hardware cannot be specialized. So maybe that hardware needs to be something that everyone already owns. And maybe that's it needs to be something that everyone already holds in their hands or in their pockets. There are 7.3 billion smartphones in the world today. And you can carry a smartphone over every border in the world. 86% of the world population owns one. It is practically impossible to confiscate all smartphones. And as we've seen also in Egypt in 2011, because the regimes rely on smartphones themselves, it's virtually impossible to get rid of them. And we can make use of that. This is exactly where Bit steps in. So Bitchat is a decentralized tool for communication. It is software. It's a protocol and it's free. And you can download it, start it up, and immediately communicate with those in your vicinity. The messages they travel peer-to-peer from one device to another like the human microphone in in Zukoti Square at during Occupy Wall Street. There's no registration, no identity, no central service servers. So there are two types of ways of communicating in bitad. One is totally offline. So it's zero infrastructure as we say where the if you send a message it travels from your phone to another phone to another phone to another phone basically reaching the entire square if you want and there's also another mode of transport that uses Noster another decentralized uncensorable communication service network that you can use to communicate with your entire city with your country with your entire continent if you want. So here is a simulation of what that looks like to just give you an idea. So what you see here is a simulation of time square and every single dot is is a user that uses bitat and the dots that you see exchanging between them are the messages and this is how the messages spread across wide ranges across the city. Let me play that again. Last year we've seen multiple instances where suddenly was in the news during political appe uh up uprisings. So here in this picture you see the Nepal protest from last year and so the young generation in Nepal was denied access to social media and that was particularly stupid decision from a government that controls a population with a median age of 25 years. So people went onto the streets and they elected their leaders of their protest and discord channels and we've seen a massive spike in downloads for bitad in Nepal preceding these demonstrations and that was a pattern that repeated over and over again in the last year. So that was Nepal, in Indonesia, in Madagascar in Kiwir and more. What would happen is I wake up and I see the download numbers spiking in a random country and I go to Google News and I would see that there are protests in the country. preceding the the protest, people would download massively just in order to be prepared in case the internet shut would be shut down. But it wasn't only political oppression that made people download It was also very useful for other cases. Oh, what we see here is that also some political figures picked it up and mobilized their their supporters using uh bitad. What you see here is one tweet by Bobby Wine in Uganda who uh told his his supporters to download Bit in advance of the elections in Uganda which led to uh more than 2% of the entire population of Uganda to download the app. Similar story for the protests in Iran that we have witnessed in this year. And finally, the Chinese government banned from the app stores, but I think that was pretty expected. So, what I was going to say, not only political movements caused people to download Bit, but also in preparation for natural disasters. And we've seen a massive spike in downloads in Jamaica last year in preparation for a big hurricane that was approaching the island. And um this led to the fact that Bitchet became the number two app in the Jamaican app stores uh just second to a weather prediction app that was used to see where it is safe and when people can expect major weather events. So not only political events but also natural events. And we've also seen something similar happen just last year in Portugal and Spain. If people remember that in a national outage of the electricity system, suddenly the communication networks and the money networks of the population was impaired. So what you see here is a huge line that was across the entire country where people would stand in line just to get cash from their ATMs in order to make payments for during a period of time where the internet was simply gone in the entire country. Cash does not only work offline, it's also untraceable. And that's a killer feature of cash for democracies. And this is something that also the demonstrators in Hong Kong back in 2011 have figured out and 2014, excuse me, figured out where they made a virtue out of a necessity basically because people knew that the Chinese government is watching the digital payment systems and especially when you pair it with your movement such as purchasing a train ticket for example, people spread this message very clearly and very quickly that you should not use digital payment systems in order to buy your train ticket but use cash. So as a Bitcoin developer, we wonder and we ask ourselves, what if we could use Bitcoin in a similar way? What if we could also use Bitcoin in a semi offline way, handing it one from one person to another without having to rely on the internet? And this is where uh cashew the chromian ecash project that I and many in uh in the open source community work on which is basically a offline cash system that supports Bitcoin itself. So think of it like in an ATM where people withdraw cash and then hand it over from one person to another. You can do similar things with Bitcoin where you withdraw ecash and you can hand it physically from one person to another. And this opens up many different use cases in how we can transact uh without the internet but also without being tracked. So what you see here is a transaction happening for 1,500 Satoshi's on an airplane without an internet where one friend sends money to another friend without being tracked and without having to rely on the internet. So what we've done is in these two projects and also built cashew on top of bit which is what I want to close here with a demonstration that you see here is two phones both running bit and what allow you know the combination of both ecash and a offline mesh system allows us to send satoshi's from one device to another without the internet. So it travels literally from one device on the left to the device on the right. So, we love magic internet money and we want it to be everywhere and we want it to be resilient. And we also believe that the internet needs to be fixed and it needs to be rebuilt on a different set of priorities. One that puts the users into control instead of large centralized organizations and one that preserves human rights instead of hindering them. One that empowers people and one one that is there especially when nothing else is there. Thank you very much. Thank you, Cali. Y'all, that is the end of the freedom stage, but that's not the end of our content. For those on the live stream, just stay tuned. We got more content focused on technical deep dives next. For those here in person, you can make your way upstairs and check out our deep dive content, or you can go explore the venue and go down to the AI lounge and learn how to use AI for good. Cheers. Thank you so much. Heat. Hey, Heat. Heat. Heat. All right. So over the next hour and a bit we'll be going a bit deeper. Um my first for those who don't know my name is Febbe. I'm the global freedom technition. Um, I run the Bitcoin Development Fund, which is our fund for supporting projects that make Bitcoin unstoppable and usable money for human rights activists, but also projects that support uncensorable communication for human rights activists leveraging Nost. Um we give grants every quarter to projects in Bitcoin space about 10 bitcoin more or less every quarter and I can see a number of our grantees in the audience. Over the next hour and a bit we'll be going a bit deeper around some of the topics that we talked about down uh in the main stage. So we'll be talking a bit around what's the state of Bitcoin education. How are we getting more people, more human rights activists, more humans in the world to know more about the technology so that it is functional for them? They can use it to solve the problems that they need to solve to get their keep their movement alive. We'll be talking a bit about decentralizing Bitcoin mining because again security model of Bitcoin is stronger when there is no single point of failure and particularly with Bitcoin mining the more we can decentralize Bitcoin mining more we can take it out of the centralized hands of the big miners the better for the network. So we'll be talking a bit about that. We'll be talking a bit around uncensorable communication using Nosta and finally we'll be talking about Bitcoin adoption. Um so it's going to be panels hopefully there should be opportunity for a few questions from the audience. So we hope it will be interactive. So as you hear conversations if anything triggers for you please feel free to raise your hands and ask a question. Um the first panel we'll have will be focused on Bitcoin education uh panelists to the stage. Um Nifty Nifty runs Nifty founded uh base 58 and runs the Bitcoin plus some of the best technical Bitcoin education content on the planet. Reus. Reus leads content and strategy at um my first Bitcoin, the premier Bitcoin. They've opensourced Bitcoin education. And Gabrielle, Gabrielle works for a university in Canary Islands who've just launched a Bitcoin masters. So, I'll hand it over to you guys. >> Thank you, Fei. Hey, everyone. Uh welcome to the uh deep dive stage. very excited to be diving into education. Uh is the first conversation. Um as Femi introduced us, my name is Nifty or Lisa. I also go by Nifty. Um and I'm here with Raina and Gabrielle. Um I'm going to go ahead and maybe let each of us kind of give a short introduction um about the program or the project that we are doing Bitcoin education through. Um so just an example, I'm Nifty. Um, I run B 58, which is a um nonprofit focused on teaching the Bitcoin protocol. So, it's a technical engineering school for people that want to learn about Bitcoin. Thank you, Nifty. Hi, everyone. Uh, super excited to be here. I'm saying excited because I'm really nervous and so, you know, trying to trick my brain. And, uh, so I run or, you know, basically the Bitcoin training coordinator and my first Bitcoin. As you know, we are a nonprofit. We started were born in El Salvador please we are no longer in El Salvador I'm Salvadorian and yeah our focus is introductory Bitcoin education we are shifting now to global international audience and our focus is also shifting to empower more educators to through tools resources uh programs and also more learning and development even um as a skills not only about Bitcoin itself public skills how to manage classes, how to go out of the fear of public speaking, etc. All of that. Let's keep that short. >> I love it. >> Wonderful. I as Yeah. I run a university that took me 15 years to get authorization. And uh if you wonder why 15 years because the type of things we wanted to do are things like master in Bitcoin. So politicians didn't like the idea of an official accredited university doing these type of things. We also have courses like we are about to launch a course for uh political prisoners with uh uh Lilian Tinto and and her team and uh yeah when we started the university after 15 years battling the government when we finally got the authorization one of the first things we did was to announce a master in Bitcoin. We have had lots of troubles with the government, lots of troubles. We are right now in the Supreme Court uh fighting the government because they want even after getting authorization, they want to stop us uh from doing many of the things that we do. Uh but we're happy doing it. We don't care about what the government thinks and um we are having more and more students from all around the world. It's an online university so we are happy doing what we do. Oh, that's great. Congratulations. 15 years is a long time. Um, so I think this is a we all think are kind of approaching education at different places. You're at the university. Um, my first Bitcoin is a more broad international project. Base 58 is online. Um, maybe it'd be interesting for the audience to hear about what kind of audience you're hoping to reach with the education and why that particular audience is an important one to learn about Bitcoin. Um yeah, I can start. So we are the step uh my first Bitcoin is focused on the very beginners, the ones that heard about Bitcoin or haven't even heard about Bitcoin and you just want to do the first take the first steps in your journey. But that's why our programs run through the history of money, how fear works, inflation that I'm sure a lot of you are familiar. But because we believe and we see the the necessity, that's the majority of the word. Like if we if a lot of us go to Bitcoin conference, if you've noticed in the last three, four years, we are just kind of just seeing each other and Bitcoin conference. We're not having new people. If we travel to Bali, to Africa, to Latin America, to the states, we are kind of the same crew. And our goal is that there is more focus. That's why now al we're doing on conference. So our more localbased because the context is the most important thing and it matters when you learn from someone in your community which is why now we're trying to help more educators from those communities with what we have accessible to our um materials at the moment. But we also try to like encourage them in Latin America for example after my first Bitcoin when it is enough curiosity. Okay, what I can do next? We recommended liberati because it's an espan. We recommended everything that is after us because that's what it makes us stronger as a community that once you help a portion of the population then these same ones can go to the next level and next level and that's why we are you know just growing stronger and bigger in the Bitcoin community because we can focused and if we try to do everything that's what we've seen in the past that it was going the wrong way and it's like okay let's just reshift and we have enough amazing people already working on development and Bitcoin um what I mean development is Bitcoin development and also people that is not the most technology people but they just like to educate. Um yeah >> quick is the reason why we have we have three programs in Bitcoin master in Spanish uh a master program in English that is about to be launched in fact several professors are around here Max Hildebrand Jonathan Newman and the one and only Alex Gladstein and uh we have the Spanish program and we have a version for those who don't want to uh because the the regular program has like the foundations, economic foundations. It has the the technical skills and it has the professional uh applications and um some people want to have all this, some people only want to have a smaller version. So we have a smaller version for those who want and the reason why we do this is in fact and I think it's important is because of our mission. The mission of the university is to explore the role of liberty in one personal and societal development. So we think that bitcoin is one of the key elements of uh pushing forward a freer society and this is what we are placing that importance and um and and we try to collaborate with organizations that have like entry levelvel courses so to to collaborate and cooperate with them >> and we need to connect. >> Yep. Totally. Uh, Gabrielle, could you say a little bit more about why it was so difficult for you guys to get permission to teach about Bitcoin? >> Well, the the problem was not just Bitcoin. The problem was when when uh the government look at what we wanted to do, they started saying no at every step. They said no, you won't get accreditation to do the type of things you want to do. Uh but it happens also believe it or not. Do you know what public choice is? Public choice is is a wellestablished economic school way of thinking about economics. It you have people from the right, from the left. So it's it's a really well established because public school it's a way of looking into politics that is a very skeptic look into politics. is an economic way to look skeptically about politics. They rejected us. They said you cannot teach undergrad public choice. And so it's not just Bitcoin. is that they saw immediately that our project was a project that was going to do things differently and um touch upon and and even put the accent and um being a contrarian and uh they didn't like that and and Bitcoin probably was one of the key elements when they saw that we wanted to and they knew me of course I was the president of another university in Guatemala. We were in 2013, we were the first university uh accepting Bitcoin in the world. And um the Spanish government when I started trying to get the license to operate this university, they immediately realize, oh, these are the people that started with Bitcoin in 2013 and this and that. Um they just prefer other type of universities. And if I can add that to add that um I think that is also happening in places or just be aware that it might happen in places where you think that is allowed because something that happened and what drove mostly my first Bitcoin out of the jurisdiction of El Salvador is because we kept independence and impartiality and a government does not like a government and independent and impartiality education. So they are technically still doing Bitcoin education but would you really we really don't know or well I have an idea but we really don't know what is being taught now because they didn't want an external party that couldn't control what education was being taught. So then it stopped the permissions for for independent education to enter on public schools is only what they are allowing now. Well, that's uh really cool to hear how you guys pushed really hard to be able to keep teaching though because it is important, I think, for people to have great sources of information about how Bitcoin works and what it is. Um, we've got about two minutes left. How do you guys feel about opening up to the audience for questions? Yeah. >> Uh, do we have any questions from the audience about any of these projects or education in Bitcoin in general? you tell us something too. Nifty, you only ask questions. >> Um, well, so one of the things that I like thinking about like when I'm or like so that I teach Bitcoin at a technical level and so a lot of the people when they come in, they learn about how Bitcoin as a protocol works. One of the things that I've been thinking a lot lately about why I do education, like why is teaching people about how Bitcoin works such an interesting and important thing? thing that I come back to is how empowering it is. You know, it's one thing to be able person who has a Bitcoin wallet and can push buttons on the screen and watch the money get sent or watch the money arrive. Um, you know, at some level it kind of looks the same as maybe any other digital wallet that you have. So, what makes Bitcoin different? What makes Bitcoin different in my opinion is the actual technology, the protocol that's underneath it. And what's so cool about it about teaching, you know, how that protocol works and what's actually happening when you push the buttons on the screen is that um I think when people start to really understand what's going on underneath the surface, like how it's possible that you can have this independent money, this money that's actually free, I think that that gives people a certain sense of their own freedom. Uh because they have this understanding that they didn't have before. And so when I what I really like about teaching or I feel like I've done a good job is when you can see people get inspired or you can kind of just feel their excitement at the click because they understand how the world works so slightly a bit different and that new understanding I think they find you know kind of not only inspiration so you know world changing in terms of how they might be able to go out and interact with the world in a different way also. So um yeah all right we have a question here 20 seconds though. Have you thought about that as individuals? >> So the question was, is this something this idea of education something that we can take to government officials and ask them about empowering themselves through Bitcoin education rather than something that they block? Either of you have it? >> Um, yes. I I hope I will personally start doing it on a private level but yes to people because to John's point at the end of the day Bitcoin is for everybody dictators and no dictators and liberty so yes we so they empowers and then they have their own network and then they don't trust me or somebody else or a bitcoiner they trust their the same way as it happens on communities it happens also with government officials so yeah hopefully >> time is scarce I prefer to spend spend my time convincing the good people, >> but ultimately they will see what happens in society and they will adopt it too. So >> true. Uh, speaking of time, I think we're very over time, but thank you so much both of you for your time and thank you. All right, y'all. Number five. All right, y'all. Next conversation is going to be focusing on mining. I want to introduce first our moderator. Where's Tyler Stevens? Let's go up on stage. And then we have Scott, contributor to the Bit Axe project. Matt Carlo, distinguished Bitcoin developer, and then finally Janet Mangi, COO, co-founder of Gridless. Tyler, I'll take give it to you. >> Uh, no introductions necessary. These people are crushing it in the mining space. How many people here have little to no understanding how mining works? Show of hands. Okay, a fair few. Um, you know, Ramz said in his awesome talk down earlier today that with the advent of AI, how it's manifesting, you don't want it to be closed source, fastly adopted, first mover advantage, monopolistic, all of these things he described. He's like, it can't play out this way. And that's exactly how Bitcoin mining works. So, we're going to talk about a few centralization problems in Bitcoin mining and how we're solving those across code, circuits, um, hardware, and then actually the energy infrastructure as well. Um but but just real quick, Bitcoin miners do a couple of things. They the best way to think about them, I think, is they provide issuance. That's why it's called mining, right? You the the protocol issues the new coins. You find them. That's why it's called mining. It also provides settlement. So that is actually processing your transactions. And that relates to the whole topic of this Oslo freedom forum, right? We want pe to be able, as uh Leopard Lopez said, you want to be able to send value without being stopped. No man in the middle there. And then also there's some immutability. So I'll just frame it here with the mutability. It that makes it hacking resistant, right? So that the miners help keep it secure in that sense. Um but but one of the things that is a hot topic I'll start with Matt here is the this notion of centralization specifically in the mining pools which I want you to explain and touch on a little bit in this transaction selection and that process and how do we solve for that? >> Yeah. Um, so mining in general, this concept of miners, it's a little ambiguous, but miners are fundamentally the the key part that they're responsible for is selecting the transactions that go on the block. So from the perspective of the network, the most important thing is that if you create a transaction and some minor is in China and Xi has decided that your transaction is bad because it's paying a dissident in China. Even if the miners in China decide not to include your transaction, someone somewhere in the world should decide to include your transaction. So we need miners to be globally distributed across many different companies so that it becomes very hard for any individual miner to censor a transaction. This is great as long as hash rate the AS6 themselves the mining farms are globally distributed and they generally are. But importantly what ends up happening in practice is they point their hash rate to a pool. So for many economic reasons what happens is this pool this there's only a few of them. In fact, there's one that controls almost 50% of the network. Um, they are the ones who select the transactions and decide what gets into a block and what doesn't. So, this is obviously bad, right? This means that these pools could censor transactions if they wanted to. There's no technical reason for this. There's no fundamental technical architecture that says, "Oh, if you want to get a bunch of miners together and pull your resources so that you all find blocks more often and you get more consistent payouts that you have to have the pool select transactions." There's just a convenient shortcut that early pools made and that we're now stuck with. And so we really want to find ways to undo that to fix this early shortcut that was taken. And so there's two protocols worth mentioning. I'll just mention very briefly. Stratum v2 which keeps the centralized pool architecture keeps this existing design where all these miners get together and use one company's servers to uh retain the funds and then distribute the funds so that everyone gets very cons every day they find blocks more often and then another project called P2 pool v2 which is a total rearchitecture to say actually pools should be totally decentralized a pool should be a peer-to-peer network rather than a centralized party um and then figuring out all kinds of complicated stuff downstream of that. >> A lot of technical stuff happening under the hood. But yeah, the importance is that your transactions are going to be processed and they're going to be secure and they're not going to be hackable, which is very exciting. We need to solve this problem. Um pool centralization, digital decentralization, as it's often referred as a very important topic in Bitcoin mining. One that is not very often discussed in Bitcoin mining. I'm going to tap Scott on the shoulder here next is the actual hardware and software controlling the miners, right? actually making the miners. In the broader computing space, we know and love Linux. If you're a developer, you're very familiar. Your refrigerator runs it. Your car runs it. Um, all of Amazon web services run on Linux. It's inspectable. It's open source in that sense. Bitcoin mining does not have this. The firmware or software is very closed source. There are some things that you can't even inspect. And it's kind of an antagonist relationship we have with this monopolistic supplier that issues these miners into circulation. Why don't you talk to us about that, Scott, and how it's related to the Bitax project, which everyone knows you for? >> Yeah, thanks Tyler. Um, I think when we're talking about any project product, uh, if it's if it's closed source, which we call proprietary, it is it is fundamentally centralized, right? there's a limited number of people or an organization who are responsible for its features ultimately who can use it uh how it processes Bitcoin transactions if it's that kind of device um and th that that centralized organization who controls that project can can dictate what it does but there's also you know there's a there's a limited number of people that could be coerced into making a decision right if if it's in a a dictatorship type situation, right? They can be controlled. There's there's, you know, a set group of people who are in charge of making this decision. And so, you know, I think that it's it's very important to break out of that and that is what open source provides. It's like that framework for the project, for the product, for whatever it is such that there is there's no limit to who's there who can control it like there there's no um you know set group of people who drive it. Uh I I started the Bidax project this as an example. Uh right so I named it I created it but it was you know quickly determined that it there was lots of other people on this who were very capable and you know I published everything that I had done which was only like a core set of things. I published it out there and now you know lots of people can contribute to it. It makes the project better. It makes development happen faster. But I think more importantly is like I can't stop it or no one can convince me to stop it. I just don't have that ability. There's no group that could stop it. It's out there. People can change it for better or for worse. They can do what they see fit. And you know this this is a this is a stand against the way I think Bitcoin mining hardware is now. hardware and the firmware that runs it which is essentially Bitmain a Chinese company controls somewhere in the neighborhood of 80% of that market and >> and has put back doors in the firmware before. >> Yes. So that you know back back in 2017 something like that when Bitman only had 50% of the network, right? They got caught installing a backd dooror that would allow them to remotely disable miners. Right? So you bought a minor, you think, "Okay, cool. I'm a sovereign individual. I'm mining a Bitcoin network." No, they could just turn it off. Did Did Bitmain decide to put that in? Did the Chinese state decide that they had to put that in? We don't know. Um, it's incredibly sketchy and we need to protect ourselves against like this kind of thing from some, you know, unilateral actor. But also, I think an important point here is that Bitman makes machines to serve one service, uh, one application of Bitcoin mining. And we know from the work that that you have done that there's the future of Bitcoin mining is not look exactly like it is today, right? It's not on-grid miners. It's not uh plugging it into the grid. There's so much more potential outside of this. And what the work I'm doing with the 256 foundation started with the Bitax is to create the tools that people can use to build these new miners to serve, you know, what will almost certainly be the future of Bitcoin mining, which will be decentralized. Yeah, Scott's alluding to my frustration as a guy trying to mine at home, banging my head against the wall to get one of these machines to be quiet or work with my power outlet or whatever it may be. They're intended for data centers, right? And they're increasingly out of reach for the average person to participate, which should not be the case. One person here that I know has experienced this firsthand who should tell us a little bit about her um very highass data centers is Janet from Gridless Compute. Right. I mean, I'm solving pain points for trying to use these miners as space heaters, but you're solving pain points trying to get these miners to behave in the bush in Africa. Can you please talk to us about that? What your minds look like, where this physical decentralization of the network actually lives, and then of course, please comment on how it has benefited your community as well. >> Great. Um so the miners yeah when we bought these miners as you said they coming from China so they're very um proprietary but some of the challenges we faced is Africa we are special so dust which nobody thought about and we've had an instance during the rainy season when termites got attracted to the lights so they tried to get into the funds of course it was a mushy business but those are some of the things we've had to customize and figure out how to work with them because of course when they're done nobody thinks about the area they're going to. So, the localization to be able for them to work efficiently. Um, in terms of the impact, so everybody else I think talks about Bitcoin mining and, you know, they're trying to make money or do something else, but for us, of course, we're not charitable, so we're mining Bitcoin profitably. But in the process of that, we're not only decentralizing the network, but we're also pushing electrification of Africa to the edge. So, our minig are in rural Africa and we're trying to get more and more people have access to electricity. Yeah, it's one of those beautiful cases of like just aligned incentives where a company can come in and mine Bitcoin profitably. It lowers the electricity rate for the people in that community. It supports the network. It's settling transactions. It's really quite beautiful. Um, are you guys all in our last couple minutes here, are you guys all bullish on the future of Bitcoin mining? I imagine the next, in my opinion, the next 10 years are going to look drastically different. And I just want to hear our what you're excited about for the the the new status quo, the emergent status quo and and how that will help dismantle dictatorships and bring bring across more freedom. Any last comments? >> Incredibly bullish. I mean, I think in in Bitcoin and especially in Bitcoin mining, we're we are so early. We can we can actually map this onto previous uh industries in tech, but we're incredibly early. Like this in, you know, in 10 years, in five years, it's going to look so much different. And I'm optimistic that it's going to be uh different, you know, for the better. >> You've said Bitcoin might fail before, Matt, so I hope you're bullish. >> Yeah. I mean, look, in order to have mining be decentralized from the point of view of the network, not necessarily the actual AS6, we have to rewrite the architecture of how miners work today. That is a high bar. That's a lot of work. Luckily, we have at least two good ideas, I think, for how to do that. And but we have to convince miners to run it. We have to finish building it. It's there's a lot of work to be done on the software side. There's a lot of work to be done just convincing people that this matters, let alone doing it. Um so I'm I'm optimistic, but uh see a lot of work ahead. >> Final comments, Janet? >> Optimistic, bullish. Um and I think especially like for African it really needs to hold more miners or uh equipment out to more people in Africa. So for example last year we launched the Jakali project which is an open-source miner that allows us to uh smallcale farmers or um push meals being able to get Bitcoin mining into their you know setup. And so this is one of the things we're seeing where if you think of the solar capacity in Africa now, we at about 6.7 gaw. So why not get all those small places connected to a bitcoin miner which allows us to decentralize and at the same time fight any oppression that comes to the network. >> Come find these amazing people and talk to them about their projects. Thank you very much. >> All right, everyone. Next, we're going to be talking about Noster and where it is today. I want to welcome up our moderator, Jeff. Why don't you come up? Next, we have Pablo Fernandez, Nasser developer and Rabbel, founder of Verse Communication and some cool Nasser project. Give him a hand. Okay, we are running a little bit late today, so we are going to be very brief with some intros. Um, I am Jeff. Uh, I am the creator of a project called White Noise and the Marmet Protocol, where I'm trying desperately to build um, unstoppable encrypted messaging over Noster. >> I am Pablo. I'm Noster developer, creator of a lot of different things. Uh, we free the NOLA and a lot of NIPS and technical stuff and experiments and weird things. >> Yeah, I'm Ra. I've worked on a bunch of Noster projects, but most recently I relaunched Vine on Noster as Divine where we restored over two million of the original things and uh brought a bunch of people on to Nostra who had no idea it's Noster. >> Awesome. Okay, so the three of us met uh a little over three years ago at the first Nostra conference in Costa Rica. Um how do you guys feel? you know, Noster's um kind of goal in the very beginning was about being censorship resistant and being kind of an open permissionless social media system. Do you feel three years on that Noster has lived up to this uh censorship resistance aspect? >> Yeah, I 100% do. Um I I think Noster development has withheld really well. Uh the censorship resistance I I think it's been proven that it works. Um, and yeah, I think it's it's been living up to its potential. >> So, I was asked by one of the board members of Blue Sky, why didn't I build Divine on App Protocol because they have, you know, 43 million users and the answer was I couldn't trust it because on Noster I know that we are restoring all these videos and we're restoring these accounts and it is censorship resistant. So, I know that there's not someone that can turn it off. And if we had built it on Blue Skies infrastructure, then we would always be dependent that someone could knock on the door of that company and turn us off. And with Noster, that's not the case. >> Amazing. Okay. Um, let's shift and talk about some of the wacky things that are getting built on Nostra and some of the most inspiring stuff. you know, like one of the things that, you know, when you read the original paper that FIA Jaff, the creator of Noster, put out, it was really centered around this idea of social media that was unstoppable. But, um, I know myself and Pablo was the very first person I talked to about, uh, you know, what Noster was when I first saw it. And, and for me, it was very clear that this was much more broadly applicable than just social media. So, we've had everything from video sharing platforms to AI systems to um, you know, encrypted messaging and social media. But I'm I'm curious, what do you guys think are some of the most uh inspiring or off-the-wall ideas that you're really excited about recently? To >> to me, I I've been very excited about one use case and I'm excited all the time about all the kind of different use cases, but you know, certain amount of time in in the world. Uh but to me the the like the unsolved problem that is not solved by anything and is solved perfectly well by Noster just in like a metaphysical sense is the the Wikipedia problem. And I I think we have a a very elegant solution on Noster that that has a really cool side bunch of side effects where there cannot be like technically there cannot exist one single ground consensus truth of what is horse or what is like we we can all arrive but no one gets the final say on the finding what the thing is. To me that's still to this day uh it's like a very old idea. is like two years old. Uh but uh it's it's like an idea that has a a lot of value to give to the world uh that is very interesting to explore. So you can think of Bitcoin and blockchain projects as a trustless unified database where you're trying to get everybody on the same page and for money that's really important. But what Noster is is a a trustless d trustless social database where the goal is not everybody having the same idea like Wikipedia but giving us a platform where you don't need permission before you start writing to these databases that we've been build apps on top of and I think that that is the essentially important part. So, if you look at things like Bit Chat, uh, you don't need to know that it's using Noster. You don't need to sign up for it. You don't need to register. It's just this network of databases that a decentralized app can take advantage of and publish to without dependent on or being subjected to whoever is particular and the inversion of power which makes Nostra so important. >> And since nobody uh mentioned it, I'm going to go with the obvious one. Um, I think the thing that's most exciting to me right now is the application of Noster around the whole AI and agents world. Um, we, you know, everybody is trying to set up open claws and Hermes agents and things like this that can run on computers. And then they immediately start talking to them through Telegram, which is totally plain text, you know, not encrypted at all. And I won't believe Pavarov for whatever he says. I'm sure he's sharing data with people. Um, so I think Noster represents a very Thank you. Um, I think NASA represents a a very credible path for private ways to talk with agents and private ways for us to communicate with each other. >> I mean, so like to that point, people probably don't know about the Sprout Project. Sprout project comes out of out of block itself and it is a specialized nostster relay instead of apps for agents for for people communicating to coding agents to AI agents that that take on activities and and this is based on some of the stuff that Pablo's been doing experimentation with which is like what if it's not just people like what if it is agents that can then write data and communicate and you don't have to have permission across this and you give sort of this substring rate by which we can do coordination of work and information that wouldn't be possible otherwise. >> Yeah. And one very interesting side effects to that is that you get provenence of where the day is coming from because it each agent has pub keys and there it's such a useful primitive that touches it's very all-encompassing because it touches everything. the fact that it's an agent and which agent and within which project and within what context like all this stuff is just inherited by the fact that it's using Noster communications which is very interesting because there are many of the primitives that come from Noster that have been rebuilt by people that are you building AI harnesses that are not in the Bitcoin world not in the Nost world but they felt the pain of I need to give an agent identity and they come up with something that is too specific to that one use case and then someone else comes with I need to build like a blockchain because I want to be able to see how the history of the context and it's all problems that naturally solved and that we inherit all these amazing amazing primitives for free. >> So one of the things that's really difficult about not having a shared global state or a database that we can all just look at and and run queries on is that we kind of don't know how many people are using Noster. Um, I think it's probably safe to say that uh, you know, going back, I think all three of us probably, well, maybe not Rabble because he'd seen this about 15 times, but, um, I think you and I would have predicted that the takeoff would have been a lot faster and that more people would be aware of what Nostra was and using it. Um, and we've seen recently, like you mentioned, with Bit Chat and with Divine, there's lots of people using Nostra that have no idea they're actually using Noster. I mean what do you think is holding back the Nostra ecosystem in terms of general user growth or kind of general user awareness? I I think when we started building uh Noster like three years ago, we were in a different world. Uh everybody well not everybody but a lot of people understood like censorship within within Twitter and all that and even though it's exactly the same the the narrative changed uh so so the market changed. It's that's going to come back because the bottom to sensor is still there. though, you know, um but but I think trying to build the the kind one like the the Twitter style uh universe, it's a it's a bit of the hard mode because there are so many super cool interesting primitives that can only be built with this infra that with this like with this spec that we have with Noster that building the same thing that is Twitter or or something like that is a little bit hard because you have to convince people, oh, I built an alternative to this thing bat it's its own thing right like it and it product market fit immediate because it's it's useful for what it is you cannot replace it with oh no I have like another no it's it's a it's a new type of product um I think leaning in on this makes a lot more sense than just rebuilding the alternative to Instagram >> so I think I mean I think it's fascinating so there was a moment three years ago where we were like is Noster going to be the open Twitter alternative And what what happened was that uh if there is one you know you have threads which is part of the fediverse nominally and then you have blue sky which has its own protocol and both of those didn't go as far they never t like they didn't uh make non-custodial versions of things and I think what happened with Nostra is that we had a lot of people who are from the Bitcoin community and had seen the problems with custodial key management and the with exchanges and with people losing stuff. So that in many ways we built the solution for the Bitcoin community and then have had to figure out how to sort of get back that intermediary step. And so you know that's why Pablo developed NZ Bunker and that's why the Divine users when they sign up they get a username and password. you can use it with your own keys, but we we hide some of that complexity. And I think that what what we needed to do was we needed to figure out things that people needed to use it for that wasn't just I don't like that other thing and wasn't just like I want it that's open source or I want it that's like self- sovereign or I want it that's decentralized. like they have people use these apps because there's something really there that they're going after and you know like you know we had hundreds of millions of people like express interest in Divine when we relaunched it and none of them knew or cared that it was on Noster um what they wanted was they wanted to watch old six-second looping videos you know or they wanted to use bit chat and they wanted to be able to just you know communicate when the internet got shut down and they didn't even need to know that it's nost >> amazing so we have uh one minute left and I've got one question and before I ask that question just a quick show of hands of how many of you guys are actually on Nostra already here. >> Okay. Geez. Okay. >> Most of you. >> So this might be the well no I'm going to go with it anyways. So um I guess >> there are billions of users if we just just from this sample >> billions of them. Um okay so we are here at the Freedom Forum talking about dismantling dictatorships. Um, how would you guys suggest that activists think about getting started on Noster? And is there anything kind of important to know there? >> You both have 15 seconds. >> Perfect. Um, I I will say go to nostrups.com and just something that you care about. Doesn't need to be the the Twitter style. There are many other very cool use cases. Um, and and just something that speaks to you. It's it's it's going to be it's going to be hard. there's going to be a lot of pain, but there is a very very uh very beautiful garden at the end of the at the end of the tunnel. >> You know, there are many Nostraster apps, but um I would actually say like go to Alex's Shakespeare and build your own social app. Like it is like lovable. It is like all these other Vzero where you can just type in a description of what social app you want and it gives it to you in any way. And so then it's yours and not whatever we think the app should do. >> Awesome. Well, that's us for time and thank you guys for being a good audience. >> Thank you guys. All right, our very last our very last panel is going to be talking about using Bitcoin as money and we'll be talking from the place of experience what the many interesting innovation we're seeing around using Bitcoin as money particularly in the African continent. Can I invite up Abu Baka, CEO of BRT, Anna, director of Africa Bitcoin Institute, Sabina, co-founder of Tando, Obi O, CEO Fedy. Okay, since this is the last panel, I think it'll be good to give some context and just rapidly just go through so we have time for a Q&A. First things first is Africa culminates into essentially the largest opportunity set for Bitcoin to flourish because an important idea to have that I think ties everything that we've been saying together is that freedom is a fundamental primitive for human flourishing. And I think Bitcoin as a concept to abstract that flourishing is the reason why you see Africa is at the forefront across multiple verticals. So we're talking about electrifying communities all the way up to practical tools. Now with my guest of panelists, I'd like to start with Ana just to give a quick background on each of them and then get into some of the questions. So, Anise runs essentially the Africa Bitcoin Institute where their idea is really to challenge some of the myths that people have regarding adoption and more importantly to surface some of that information so that people can actually work in terms of the continent and OB on the other hand works on Fetty which essentially allows for people to engage in collaborative custody of Bitcoin and kind of providing that community-led flourishing based on the properties of Bitcoin. And with Sabina, she's essentially championing through Tando practical use on the ground when it comes to Bitcoin. who are actually spending Bitcoin and not just talking about it or watching the price rise. So I think just to start us off in terms of the conversation looking at Bitcoin going from just a tool or an interesting piece of technology to being something more than just that to becoming a symbol of resistance and actually fighting back a lot of the issues that we see on the continent. I'll start with Obie. What would you say are some of the things you've learned based on the different communities you've set up in Africa that really highlight the sensitivities that you need to be aware of or at least how you approach being a builder trying to figure that out to ensure that people actually have tools that they can use? >> Okay, so two parts. Thank you for that. Um I'll take the second one. um and it's been alluded to in various talks already which is the fact that basically most of the people in this room have no idea what the end user wants. The only person who understands what the end user wants is the end user. And so one of the lessons that we've had to learn learn again and again and again is you have to just be on the ground in person face to face with the user listen to them and remove all priors and then you maybe start to learn what is needed. And the second thing is in terms of um things you have to be aware of. Uh regulation is coming around the world. Um and if you're in countries which are financially weaker, the regulation is going to come in harder and stronger. Um turkeys don't vote for Christmas or Thanksgiving for the North Americans here. And um that case is never more stark than in in Africa. So you have to make sure you either decide there's no way to be a little bit regulated. It's like you cannot be a little bit pregnant. You either decide you're going to be fully regulated or you have to be unregulable. We have chosen Metti to be unregulable. >> Interesting. And I think that concept of being resilient enough or being just as resilient at as the tool that you would like people to benefit from that resilience is kind of core to a lot of the building that's happening on the continent which I think will be an interesting dove tail into the work that you've been doing Sabina. So could you walk the audience into first and foremost what you've learned on the ground championing the idea of spending Bitcoin and making Bitcoin an actual money on the ground and some of the lessons that you think people would likely not know unless they're actually on the ground. Um, thank you so much. So, the first lesson that I've learned in my journey is people want to sample Bitcoin before they can actually use it. And what I mean by this is they want a low stakes way to see Bitcoin work for themselves before they commit. So, before they buy that Bitcoin and save in it for like years or spend it, whatever they want, they want to like do that with a small amount first. And at Tando, we've been able to do that because the minimum spend is around a hundred sats, which is less than a dollar. It's like basically cents. So, someone can just download the app and try it for themselves. Um, verify that it works and then they can decide like, okay, I'm going to look into this Bitcoin thing and, you know, keep it moving. Which brings me to my the second lesson which is um people are going to use Bitcoin before they study it. Um that is a big one that I've actually learned um on the ground cuz I meet yesterday I met some Bitcoiners and they're asking me like why do people use um custodial wallets? Like why can't they just use a Phoenix or something like that? Like a lot of people don't have the privilege of time, at least where I come from, to like sit down for 10,000 hours and study Bitcoin and how money works and how it's broken and all that. They just want to be able to move value. Like they're busy keeping up with inflation. They got to uh put food on the table. So they don't have time. And so that's why like they just want something that helps them yeah to move their values, save value, like a better tool, something that works better than whatever option they have right now. Um and yeah, so that's why we make or when you're building a Bitcoin to you have to make it don't overwhelm them with the tech. Um and just like yeah, remove that and make it simple for them. Like simplicity is the ultimate sophistication like they say. >> 100%. And it's interesting you bring that up because a lot of the times it's very easy to assume on the ground what people likely resonate with as being builders. So it's been fascinating to see you guys actually not just build but respond in real time into how people are actually using Bitcoin which I think brings me to an all the work you're doing with the institute. A lot of people don't realize but without knowledge or understanding of what's happening on the ground there's absolutely no way for you to solve problems and more importantly there's no way for you to surface other problems that people have already worked with or worked on and that hasn't worked and for people to learn from it. So walk us through what you think are the maybe just the largest and the second largest myths you've come across in terms of what Bitcoin adoption is in Africa and more importantly how regulation actually either benefits or affects it uh in terms of positively or negatively. Well, so in the research that we've seen um just looking academic research in general, what's what's written about uh Bitcoin in Africa, it's often taking a perspective of financial inclusion and it's it's a story from financial inclusion, the unbanked, which is true, but it's also much bigger than that. It's a story of needing to to correct the financial problem that our government have put. And what by that I mean the financial frictions exactly what Sabina and um and Obie were talking about. And so that research that we see talking about the story of Bitcoin in Africa is actually guiding policy makers to think about well if we think about financial inclusion let's think about CBDC's and other tools like that we actually I believe are exclude are going to exclude people and so we are trying to do better research in that to really demonstrate that a what is happening on the continent is really about this responses and the response is you know being able to move value easily at fast a a lower cost and the product that's built around that you know in order to show and from a policy perspective if you really only think about it from this financial inclusion perspective again like I say you're going to have bad policy like CBDC's but at the same time too there's a lot of a misalign alignment between region and countries on what people can do and so there's a lot of misunderstanding but there's also an opportunity an opportunity that where there's not clear a legis leg um legisl ation then also you can create and you have more room to create as well too and um so I think there's both issues but the main thing is really changing the story that is beyond the financial inclusion which is true but is really solving the problems of financial institution problems that our countries have been um facing over decades and all of that comes from a political perspective >> certainly and I think it's interesting and this will be a good point to kind of conclude this off before I open to the Q&A which is that Africa has so many interesting diverse government structures, let's say the least, as well as we'll see problem sets like from electricity all all the way up to all kinds of things. But most importantly, I think that the thing that we should think about when we're talking about Bitcoin adoption generally is that when we're trying to dismantle corrupt autocratic regimes in these type of regions, what you have to realize is that you can't dismantle what you don't understand. And most importantly, you can't replace it without you guys coming together to actually do it as one. So, in closing, before I open up to the Q&A, I'd like to hear from each one of you. What is one thing you think whether change work that you think will positively contribute to Bitcoin adoption in Africa? >> Uh okay. So for me I think it's just proving to people or showing them that Bitcoin is money. That's what they need to understand cuz I find that there are so many misconceptions um around Bitcoin. I'm also a Bitcoin educator in Kenya when I'm talking to people about it and they think it's just for tech bros and rich people. Um and then we have to tell them like it's just money just like the way you send your Kenyan shillings or use your dollars and all. Bitcoin is the same and that is why like I do what I do with Tandoor. Um yeah creating more tools that help people see that Bitcoin is money. >> Uh I would say the same. Um we keep thinking that Bitcoin is store of value. Not this audience but many. And Bitcoin needs to be a medium of exchange and it's being used as a medium of exchange in the global south. We are growing. We're already in 22 countries and growing because we know that there is something that everyone's missing that it is money. It was always meant to be money. And when you show it to people, they respond. >> Uh and I'm going to talk actually about stable coins. And so I would say that one of the issues and the challenge for Bitcoin adoption is going to be the growth of stable coins around the continent. And many people who are building around Bitcoin are also a providing solutions for stable coins which you know are easy quick fix. But the long-term effect is actually going to make our government our economies weaker. And so my call is to just make sure to really continue developing on bit on for bitcoin and make a bitcoin adoption easier so people don't result into this quick and easy fix which is stable coins which has actually uh really uh negative ramification over a long term for the growth of the continent because at the end of the day we will not be sovereign and the sovereignty begins with financial control. So, we have time for just one question. >> He was first. >> Um, thank you. When I went to the Africa Bitcoin Conference last December, it was one of the most profound events. Uh, and it was actually Femy's talk when he said, "We don't need more Bitcoin churches, we need more Bitcoin factories there in Africa." and I left feeling what what could I do to support the network and and the African development and from your guys shoes and being on the ground. What is it to support freedom fighters to support uh you know the people in this room and and anybody that wants to build in Africa that you know those that we go back home uh and you know is there anything that comes to mind that we could help you build? I mean, from our perspective, uh, well, we're hiring, so if you if there's if there's a role that you can fit, um, while other people are letting go, we're hiring. We're getting incredible people. Um, so that's one thing you can do to either get on the ground and see it yourself, which you already did, or connect to people who are on the ground and trust what they're telling you and then build the solutions that they're asking for. Uh so for me uh we are a very small team at Tando. We're just like three people. Uh we've been trying to expand to other African countries and it's not easy. It's a lot of paperwork and legal stuff. So if you have any advice or if you know someone who can help us um maneuver that. Yeah, I would appreciate. Um and I was going to say actually I think you know just kind of Empessa showed the world of how to do um mobile money mobile banking. I think that it'll be great if we can export some of the product in the that we've built in Africa to the to the rest of the world. And when people come to Africa to like ABC and see some of this new technology, talk about it with your friends in United States and others, try to see if you can, you know, bring some of these technology to your communities. And I think it's going to also demonstrate really how Africans are really showing to the world how bitcoin can be used as money and those use cases, examples are there. So, and the technology is being built by these amazing people. So, let's try to bring that to the rest of the world as well. Definitely definitely definitely double click on that. I mean you have an example here with Tando which is an incredible product. You can literally live in in Kenya for example. We're not using anything else other than Tando. And I don't think anyone else has done that anywhere else in the world. We need to share these ideas and get other countries to follow the lead which Africa is setting >> 100%. And with that we're way over time. Thank you so much. And yeah, let's dismantle dictatorships. >> Before you go, >> one more round of applause one more time for all the panels. So, so to wrap up, I want I I I want to say something. So, when I started, I mentioned that I run the Bitcoin development fund at the Human Rights Foundation. We're always on the lookout for ideas, projects, products that people are building to address one of the four areas that you've seen presentations around. people who have projects that help us get Bitcoin knowledge that's functional to human rights activists or that helps us get Bitcoin knowledge to builders, software engineers who understand local context of the places where human rights activists are that can build relevant solutions in those places. We're constantly on the lookout for ideas that help us decentralize the Bitcoin mining sector even more. That takes away more of the power out of the hands of the centralized miners into the hands of actual Bitcoiners so that we are all the ones mining Bitcoin and we are all the ones securing the network and we don't have to rely on those centralized both manufacturers as well as pools or miners. ideas that help us turn Nosta away from a hobby tool for developers that are trying to test their skills to something that actually solves real problems that human rights activists have around staying uncensorable and unplatformable if I can use that word um so that they can continue to get their movements to thrive. We're al also constantly on the lookout for ideas that help turn Bitcoin into real money for real people in real places that they can use in the language that's convenient that's natural to them. So they don't have to know English to use the technology that can deeply integrate into their local fiat rails like Tando so that they can start from a place of familiarity till they get to a place of complexity. So we're constantly on the look at and and here's the interesting thing. We're just like the spectrum of speakers you've seen. We're as much interested in products that are or work that is open source but projects that are also built to have a financial model for sustainability. We are not we're not um maxes in any one direction. We're we're open because we realize we don't want to be the ones sustaining the projects forever. We want the projects to be able to have a life of their own and be able to economically stand on their feet. The Bitcoin development fund is the innovation Kickstarter fund in Bitcoin. We're willing to bet on ideas that are early but that we can see can have a transformative effect on the work of human rights activist. Now with AI and VIP coding and agents building is now easy. The hard part is getting it to market getting to users. So, as you're thinking about project ideas that you want to send in for support, we'd love to see that you've clearly thought about how you're going to get this in the hands of users and if possible that you've demonstrated some traction already doing that. We accept projects all year round. hrf.org/devf fund deev fund. We accept projects all year round. We give grants every quarter. And we're constantly looking for those ideas because we need a ground swell of projects of ideas for this thing to become as ubiquitous as we need it to be. So as a parting word from me, if there's something you've been toying with, you've been working on, you think is something we may be able to help, please apply or please reach out. I have a number of my colleagues in the room. Harrison Judy, Alex Lee, all work with me on the fund. So if you have any idea, you still have the rest of today, you have tomorrow. Reach out to any of the four of us. You can bounce your idea off on us and we're very happy to tell you if it's on track or if there are things that you need to tweak for it to be to work for us. Okay, that's second to final word. Final words. >> Um, can I ash slide? So, tomorrow tomorrow we're hosting Beat Devs in Oslo. So, and and here's what's interesting. We'll be hosting I'll be hosting that beat devs alongside the original creator of bit devs. So we're going to be having an old school beat devs. We'll be talking Bitcoin core. We'll be talking lightning specs. We'll be talking uh some interesting Bitcoin security research. We'll be talking coin swap, payjoin, but we'll also be exploring and doing deep dives technically on some of the projects that you've seen here both from a at the application layer at the m the centralized mining layer, the mix. Okay, so that's the venue. It's going to be at 1:30 p.m. tomorrow. Some of you've gotten invites for it. If you haven't take the picture, just show up. just show up. Okay. Um we it's going to be a fantastic fantastic experience. And if you've never been to Beat Devs before, please come so you can experience it. Uh Beat Devs is a growing movement. We we it started we started it in Africa as well a few years ago and it sprung now with vitro support in many cities across the continent and many across the world. Um um we've done for I want to thank you all. I think we should give ourselves a hand because you guys have been with us from the start and you stayed all the way. Hey, show of hand. Who who was at the freedom tech stage at Oslo Freedom Forum last year or a previous year? You guys will notice last year the entire program held in a room this size. Now we've expanded to a room that size and we're hoping that this will continue to grow. The fusion of freedom technologies, artificial intelligence, we can push the boundaries on using these technologies to help the people that are fighting for freedom in the world. Thank you for your time. Thank you for joining us. Um tomorrow there will be expose in this building. Uh there will be shuttles from the Grand Hotel. Um, the tech programming tomorrow is going to be the beat devs. Enjoy the rest of your evening. Thank you very much.