Reigniting the cypherpunk spirit with Juan Benet x Protocol Labs

My name is Juan Bernett. Um, and I am uh here in the privacy community uh in the privacy and security community because I strongly care about uh human civil rights. Uh I think it's extremely critical for uh our civilization and for the freedoms and liberties that people have around the world uh to have extremely dependable and trustable digital infrastructure. Uh freedom of thought, freedom of conversation, the ability to discuss uh issues are foundational to having good governments are foundational to having a good life. Uh and so from my perspective, as the world becomes more and more digital, as more and more of our activity happens over the internet, o mediated by computers, so much of our lives depend on the security of our applications.

So much of what we do day-to-day depends entirely on what our computers or phones let us do. And therefore it is critical that those systems uh uphold our civil rights and that we bake in um digital human rights into the internet. That is achieved by building protocols, by building systems, by building large-scale networks that they can write at the lowest layers and build infrastructure that users and applications can depend on uh to give them control over their data, control over their information and the ability to communicate with each other freely. um without censorship, without being spied on, uh without fear that whatever they say today, some nefarious group in the future might use against them. Um history is full of uh really terrible examples of periods of time where people were free and and uh able to say what they wanted uh and then later some group uh was able to use the records of that information uh to persecute those people.

Uh and so today luckily uh thanks to the fights of you know honestly millions of people uh throughout history we've arrived at uh a very free world relatively speaking um a world that h where hundreds of millions of billions of people around the world enjoy uh extremely high uh high standards of living um good rights good ability to communicate uh but that's not necessarily given for the future uh we could have a regress uh things could get worse in many locations we are seeing the beginnings of bad in many places. Um, and certainly today, many many people, many millions to depending on how you count, hundreds of millions, uh, live in conditions with no good rights, uh, and no good, uh, uh, liberties. And so it's really critical that as the internet becomes a bigger and bigger part of humanity, a bigger and bigger part of our lives, we build strong rights into the stack that permeate [clears throat] through the lowest layers and give us a dependable infrastructure that no centralized party controls that no corporation or no government can subvert. Um that really enables humans to uh interact freely and that permeates all the way to the application layer and to the individual actions of any user with any other um whether but via text message or um email or documents or you know social apps or video calls or whatever it is or nowadays you know AI conversations um we really need a digital computing infrastructure that's robust a digital infrastructure that defends our um our rights and protects us um and that we can uh truly depend on. The privacy and security community now uh can rely on um many thousands of people working across many different types of things.

uh everything from developers working on um uh protocols, applications, systems, um uh new new cryptography, new um game theory ideas for how to coordinate networks, all kinds of uh systems. Uh we also have lots of groups that are developing uh and producing uh these systems at scale. Like there are many um or organizations and companies that are developing these things. everything from signal and the matrix protocol or even things like WhatsApp which ended up adding into an encryption to you know hundreds of millions of humans uh uh messaging. Um we have of course uh super um organizations that have helped uh push the conversation forward with policy makers uh groups like EFF um and the ACLU and uh many others who've uh protected and defended um uh security and privacy uh over the decades.

Um we now have uh a number of organizations and communities around the web3 space that are taking on the uh this charge and this battle. Um there are groups like web3 privacy now. Um there are groups like uh protocol labs, there are groups like the lip community, the Ethereum community, um the IPFS community. Um there are groups like the Zcash uh Starknet community. Um there's just now I think an explosion of probably tens of different projects uh each with many hundreds to thousands of people involved that all care about privacy in different ways and are implementing different parts of the puzzle and that is leading to this tremendous explosion in interest and work towards achieving uh strong privacy goals uh often in different directions and with different guarantees and different pieces and so on but it's overall a strong net positive and all of these groups will end up producing um shared layers and shared protocols um that end up solving you know core problems together.

Um you know think of maybe the how a lot of the encryption of today's internet relies on all of the hard cryptography work that the large centralized data monopolies ended up doing. Even though we today would prefer our our data not to be captured in these massive data monopolies, at the very least they worked really hard to implement, you know, secure um encryption protocols and, you know, improve things like build things like quick and and so on to uh to help secure the internet. Uh and so I think we we have a broad larges scale coalition or even coalition of coalitions um where lots of participants are working on these different parts of the stack and the world more broadly recognizes just how critically important privacy and security are uh to our daily life. And I was and and this is a segment of a much larger group of really millions of people throughout history that have been fighting for establishment of rights for people globally whether it's by arguing it for it in government or fighting for it in you know uh rebellion actually direct rebellions or revolutions that shaped um our past uh or in um civil disobedience or in you know silent protests. or silent marches and so on that are such important um moments and actions uh to promote the rights of people uh of the present and the future.

And you know again today we get to enjoy the fruits of of those fights of the past and there is a fight that is ours to fight uh to safeguard our uh future and make the world better for people that will come uh in the future. And one of the things that I'm especially starting to turn my my eye to is what does security and privacy look like in a world that is that has a a a vast amount of digital life and digital intelligence in it. Uh we're talking about we're in the cusp of potentially billions to trillions of AI agents which will eventually could give rise to a to a whole species. Um we have we're going to be encountering millions of robots and eventually hundreds of millions of robots. Uh those will have a form of agency.

They could you know in a in a dark possible future lead to deep conflicts whether it's you know the sci-fi human versus machine sort of thing or um a you know human different human groups wielding these AI superpowers against each other. Um those are kind of the dark paths and a lot of us are fighting for a much more optimistic positive vision of the future uh where we can establish a much grounded civilization for everybody for all humans for all AI systems um into the future. And I think this is going to become a more and more important conversation as we go into um 2030, 2035 and beyond. I think that um we have a lot of negative storytelling because humans tend to respond more directly to um high stakes drama stories. Uh this is why most of our media tends to portray all these like high tension high dramatic moments.

And so whenever you work in some f you know futuristic sci-fi story uh and some like interesting new idea into it, it has to somehow be related to some high stakes high drama thing. And so most of the media that we end up producing or consuming at large scales ends up being very dystopic in nature. Um, and that is a a a a tragic state of affairs because it it deprivives us of a lot of positive optimistic visions of the future that we could all share at length and be able to shoot for. And so so much of our action tends to rely on having to articulate shared perspectives and shared shelling points. and the bulk of what we have to point to are dystopic narratives and therefore we're always kind of talking about like hey let's avoid that terrible thing that we all know about.

Um when in reality in practice most people can think of and imagine very positive optimistic visions of the future. We just lack the common reference points to be able to point like hey let's build an amazing future like in this particular story or in that particular story. Um that's why you know I think um building narratives about existential hope and optimistic visions of the future is so important and I think media has an enormous role to play there in helping craft those future visions um and create you know interesting stories for uh maybe with high drama but maybe just set them in a very good positive future um that has really interesting aspects to it um and don't fall into the easy cheap thing to do which is to um you know fall into like a trope that it feels just overly dystopic um that way you can help so many more people around the world dream of a better tomorrow and actually have a practical view of how to achieve it. Now that said, I think some of the dystopic narratives are useful in helping paint what's at stake and what bad things could possibly happen and what to avoid. Um I think the Orwellian narrative is is uh one that I think a lot of people know about and and connect to uh and have you know kind of given talks like this in the past because um Orwellian states is just something that people kind of think about.

Uh and then when you go back and read 1994 it's actually quite tame like the the the stuff that the technology in 1994 could do like maybe the TV could had a camera and could look at you. Uh maybe there were a bunch of microphones in a bunch of places. Maybe there were some spies following you around. Um maybe they were using that information against you, but man, it's super tame relative to what you can do today. Today we have, you know, most buildings have tons of cameras.

Uh most people have multiple devices with many cameras. Uh all of these devices, we know from uh leaks, uh that all of these devices can be hacked and their microphones activated. So multiple nation states could at any moment actually look into any room they want and look at all of that of what's being discussed. We have massive scale data centers with huge amounts of uh computing power that can crunch through tons of this information um and build automatic profiling databases and um game theoretic models uh to create behavioral systems to cause populations to behave a certain way. This is happening in some countries.

Some countries have things like, you know, automated credit systems that reorganize society based on what people do day-to-day. So, um, yeah, 1984 was like ended up being tame relative to like what ended up happening, let alone what's possible in the future. And for a lot of people, it's very useful to have like that concrete thing of saying, hey, this stuff is like bad. Let's not do that and let's do, you know, something better or how can we build technology to prevent that? Um, one of my favorite stories on this is, um, the cypher punk movement, in fact, was motivated by 1984.

Um, uh, there are many people who read that book and were directly responding to it. Uh, in fact, one of my personal cyberpunk heroes, Mark Miller, uh, talks about, um, being so excited about the RSA paper, uh, I think it was the RSA paper, uh, because he saw it as a path to preventing the Ministry of Truth of 1984, and he directly, um, was involved in the crypto wars and he went and copied that paper and sent it out to a lot of people to make sure that it could be used um, thereby making a a a huge action of contributing to the to the crypto wars. Uh directly motivated by the 1984 story, directly motivated by avoiding the Ministry of Truth and building a much more positive vision. Um and speaking back to uh painting an optimistic vision, uh Mark also has uh given us a a very positive vision of the future and his partoropism perspective where um he imagines a world a civil the civilization uh of of our of our world as a game that we get to play. Um, and if we play it well, then we get to keep playing.

And if we, you know, we fail, then like we don't get to keep playing. Um, and if we can take actions that coordinate and collaborate and make things better for everybody involved, we can just achieve this incredible era of extremely high abundance of really removing all of the ills that plague us today. removing disease, uh removing unnecessary death, removing um uh oppression, removing poverty, just getting us uh to a point where we can fully self-actualize to the to our to the extent that we want um and be able to put humanity on a path to increasing our use of energy and getting to, you know, climbing the Cardiff scale and being able to explore space and being able to um explore our digital uh lives. And so I think that that's a very positive vision of the future uh that I think you know a a great cypher punk in our community uh helped helped us uh point to and helped us aim towards um and he did it by both looking at a bad dystopic story and figuring out how do we avoid that and taking very concrete direct actions whether you know important in the- moment things to do or actually writing code or writing protocols or building systems uh you know writing capability crypto um or painting a vision uh of the future that's very positive and helping us articulate how can we approach this, how can we navigate the space, how can we um coordinate and build systems of rules for ourselves to get to a a better future. So yeah, I think like all of these stories, all of these u narratives, all all media um plays a massive role in helping us uh orient uh for the better, for the future um and helping us really achieve it.

Yeah. So one of the things that um sometimes holds our community back or technical communities back is that you end up with small or sometimes big disagreements creating tribal differences to the point where communities uh break down in their goal sets or communities uh pursue different approaches to building something that are technically different in some crucial way and they end up fighting too much about that and they end up creating you know fragmenting the space and making it difficult to know what the right thing to follow is um or oftentimes just confusing a lot of the people who don't understand the technical details um uh enough and so one of the things that I think can be or some of the things I think can be extremely important in this moment in these kinds of moments is that first off you know those disagreements are good and important to have and we need to be able to have them but they should come in a way that don't detract from the whole goal of building better security and privacy for everybody. So whenever we engage in those disagreements, we have to be careful not to accidentally sabotage the success of the ent of some other project that could be on the whole improving conditions for lots of people globally in some important way. Um, and we should also approach them with humility. Sometimes we think we're right, but it turns out the road that maybe we weren't right and that the other group was right or maybe we were all wrong and actually it's a totally different thing that we needed.

And so we should approach all of these disagreements with a lot of humility knowing that at any point all of this could turn out to be a different way than we initially thought. Um and so I strongly encourage uh people that are in these kind of different communities that might be fighting with other privacy oriented communities or or or especially in a in a technical group implementing some protocol where you disagree strongly with how some other group is doing it to take a moment to step back, zoom out and think about what are the shared goals that you have. What can how can you turn this into a positive sum game? How can you um maybe you disagree about some things and that's important and you need need to be able to explore both paths, but what shared things could you actually build together and lead to a better outcome. Um this is very uh you know maybe you're working on different messaging apps and you can come up with common standards for how do you do attachments and how do you do um references to other media or how do you do um identity systems or or maybe you can pick a common identity system to make it all flow better.

Uh maybe you disagree on on the core um components for a a cipher and you think like another groups is broken and yours is correct. Then maybe together you can agree to fund a group to do s serious security audits of everybody's stuff and create a better path to making sure that everything is actually safe. Um having uh a a just a a kind of collaborative posit positive sum oriented path to identifying what the real problems are and how to solve them and and how to kind of move the landscape forward uh could be a much more successful and much more profit profitable and better um path for everybody. Uh, one thing that I've been hoping to do because I've been kind of looking at all the messaging apps for a long time, I I use a lot of them. Um, you know, I like tend to look at Signal to be the strongest privacy-wise.

Um, Signal and Matrix. Um, I I was used uh both a ton and I'm involved with u I'm involved with Matrix and and Element more than than Signal, but I use both quite a bit. Uh, but in practice, I actually use Telegram much more than either of those. And I don't feel Telegram's security model is as strong as signals or or matrixes. Um now I'm not sure uh you know there's some interviews with uh uh Pyrov and others that claim that Telegram stuff is all open source and auditable and so on just the same.

Uh and I'm not sure. Let's actually figure it out. Let's create a strong um path and a strong, you know, kind of security team that can audit the hell out of all of these systems and can compare them in detail and then can compare a whole bunch of other features about them like hey why would a lot of people choose to use one app versus another. Is it like the security aspects or hey maybe like threads and topics are actually really important to communities or to um uh collaborative groups or hey actually the emojis and reactions that maybe feels feel uh too silly to have in this like serious privacy system are actually crucial to human communication and help us express ourselves in a much better way and therefore most people will prefer the app with the nice emojis than the app without. And so maybe what we can do is identify that and realize we should just standardize these great emojis everywhere and kind of use them.

And so we can turn this kind of tribal battle into some way of together as a group running this portfolio exploration approach, this evolutionary system to find what are the good things to build and all the time um take whatever good lessons are h are being learned in any of these groups and try to apply them and cross-pollinate them across the group. Um, and also remember that having, you know, kind of monopolies is is brittle. Like you you end up with single points of failure. Even if you think that one system is the most secure, maybe it isn't for some reason. And having multiple systems at scale that have multiple independent teams fighting for uh could actually turn out to be a much better, safer way to to build things.

And so yeah, I'm very excited that like we're achieving that kind of coordination and collaboration and groups like the cypher retreat that um that we've hosted and run. Um I'm excited that to have that in um lots of collaborations on GitHub that end up happening. Lots of our communities cross paths um and and even in kind of more kind of like modular stack building block groups like the Litopia community and and so on. It it is so difficult to build software and good systems that a lot of people adopt. Um

nobody

the funding model question is just so such a hard one. Um the and I and I say this from experience like I've started by building IPFS in open source. Um and I built a company uh called Protocol Labs to help develop IPFS and Falcoin. And out of that we built Litop and we created you know protocols uh that that a lot of people can use. We've experimented a ton with um community oriented public goods funding models.

We're pioneering and pushing the frontier on a range of those things. Um we've been able to um coordinate fundraising for lots of public goods oriented you know donation type campaigns for systems. We've done a lot of venture capital type investments. Um and so on the whole PL has deployed hundreds of millions of dollars into lots of privacy and security oriented tech of very different types and the systems to be able to coordinate the building like you know the public goods type infrastructure systems. Um, and look, I think there's a lot of strong disagreement across the board on how people feel about all of these systems.

Some people think the public goods solutions will never possibly work. It all has to just be private capital. Uh, some people look at private capital and VC funding as like nefarious because it's a profit motive. Um, some people look at uh donation based systems. It's just never going to be able to scale to reach um, uh, profitability to reach the scale of like expense needed.

And so the first thing that I would say to all of this is like again have some strong humility of approaching all of these possibilities with a pragmatic impact oriented lens of saying what are the systems that are working and why? What are the components that are leading to good successes? Um and what are the problems that each of these create so that you know kind of what you're getting yourself into and can navigate for that. So you know for example in the public goods funding case um it's extremely difficult to get large- scale donations to u appear without some kind of direct um return flow uh because the world just doesn't have largecale public capital available. Um it has much more private capital available.

So it's just difficult to fund raise in the first place. And then second it actually is quite hard to allocate public capital correctly to produce impact. tons of grant programs end up with massive amounts of waste with lots of coordination failures. Um, and so you end up in a situation where you you might have raised a public fund uh of some kind, you might have deployed some of that capital, a small fraction of that had some impact and then because you didn't figure out how to make it regenerative in some way, how to like return some capital flow in the in the end, then the fund dies and like and you're done. Um and so this is kind of what pushes people towards kind of um more regenerative VC type models.

In fact um a lot of people look at VC as extractive and in fact it's like one of the the reason it works is because it's regenerative. It's it's a structure that creates a um revenue loop where you have the ability to invest in a thing if or a set of things. If those things end up being successful they return some capital so that you can kind of recycle it and you can keep going. So it's by definition kind of um embedded in this kind of given that the world is today in 2025 a capitalist system for the most part then um it's a regenerative loop that kind of works within that um and so that can work now I think the hard part is when those systems and and and where a lot of the valid criticism of VC comes from is that those um in order to return funds often those VCs and you know there are many of them they compete with each other and so on are are so focused on the ROI that they push for all kinds of decisions that are shorttermist and not long-term oriented and bad for, you know, the uh lots of groups at large and so on. And so they end up kind of self-sabotaging in the long run.

Um because they're stuck in this like short-term oriented fight against each other to be like the most successful VC fund. Um and so look, that's just a I don't know consequence of of being stuck in the capitalist structure like just that happens to any kind of corporation. Um, but at the very least they have capital pools that can like grow. So like that's already kind of uh you like more sustainable than like not having some kind of regenerative loop. Um, and you know a lot of us hope that individual donations could actually sustain lots of groups.

Um, but and in practice that helps for groups of a certain size, but it kind of breaks down. Um, you know, Signal, for example, was able to be built by a few large-scale donations from a few individuals who had a lot of capital to give and a lot of donations, small donations from a lot of people. But all of that together just isn't the scale of funding required to be able to produce a super highquality product that can compete with something like Telegram or that can compete with something like WhatsApp. And so that just kind of goes to show like that model might have merits against the other models, but it's not as impactful or as competitive as the other ones. And you know, it's kind of like not not working.

And that's kind of in the case of Signal. In the case of Matrix and Element, the donation path was just like not supporting Matrix at all. Even today, the Matrix Foundation is struggling to get enough donations to support. Even though tons of groups depend on the on the Matrix tech and tons of groups um you know are beneficiaries of all of the great cryptography work and all of the great systems building that the Matrix community built. Like the amount of funding that goes back to Matrix is like tiny.

And the a group of Matrix folks had to build Element as a company and raise VC capital and try to build a business in order to generate a cash flow to be able to fund long-term matrix development. Um, and even that is like really hard and they're like, you know, on a path like they're uh succeeding now after like incredibly hard um, you know, going on a decade now of of trying to do this. Um and that you know they they've both kind of had to um they had had to build a like very strong business and that has like uh was not at all certain that it was going to happen well super fraught landscape with all kinds of challenges and and again it's like a a model that you know is maybe working better than the donation only model but it's not at all you know giving them some kind of free cash flow at scale that they can suddenly now pour all over open source right and um all of these systems have um constraints and hurdles and so on. Um the thing that I find really interesting about the token model is that you know in the cryptocurrency space uh which again the cipher community and the and the agoric papers community um kind of uh envisioned you had a way of kind of putting the tools of capital formation directly into computing um in a digital landscape to be able to use it much more closely. And that has led to, you know, people having these incredibly powerful techniques and tools, uh, which, yes, can do some like amazing things.

You can build things like Bitcoin, you can build things like Ethereum, you can build things like FCON and IPFS and so on. Um, uh, because you know, Falcon helped us fund IPFS. Um, but it can also be used to do all kinds of like memecoins and do all kinds of like, you know, outright scams and like outright, you know, all kinds of bad nasty instruments. And so these are just really powerful tools. And so naturally a lot of people who don't know much about tokens are like whoa hold on like there's just a lot of noise and nonsense around all this stuff uh maybe that's all bad um in in reality again like you have to be um humble [snorts] and pragmatic and understand what the what the core instruments are.

These are financial transactions like any other uh it's just a digital representation of a financial transaction and it's in the nature of the transaction and the token and the system that you can answer that question hey is this like a good sensical system or is this kind of some kind of scam that some group is is starting to run. Um, and so I do tend to think that if people can build good crypto systems that have some kind of um concrete economic flow to them where it makes sense to have and represent that economic flow with a cryptographic instrument of some kind and token token that they should do so and create some kind of revenue flow through it to help fund the platform. Um and then an added bonus that potentially they can fund raise on top of that using that instrument and that can be very successful and we've seen the success of Ethereum, Balcoin, um Starknet, um uh Zama, um like Aztec, just so many groups have been able to do incredibly successful privacy tech thanks to that model. Um and that you know Zcash also I forgot Zcash like probably the the one that is uh uh kind of blowing up right now. Um, and so I think like people should lean into it.

However, be careful, be really careful of the oversp speculation that emerges once you have a financial instrument at hand. Um, I am like personally very allergic to all the over speculation that the web3 world has has um, gotten itself into. And and that's kind of this weird mix of Wall Street and Silicon Valley with like just kind of like in a way that like most tech people just didn't expect. Um and and so that speculation flow is um is very messy and nasty and and it's kind of particularly a problem because there's this hard optimization pressure on all ecosystems to pay attention and almost cater to that speculation because if they don't then their their actual financial success as a project I is is like goes lower. um you know like if they don't kind of cater to the speculators then um their success of privacy project will fail relative to some other project that does and so you have this serious optimization pressure on all these projects to try and like cater these groups and you know I think it really sucks and I think uh luckily the the world is kind of doing better now I think the crypto and web community today is way less speculative than it was say four years ago or eight years ago um and so we're as we're maturing and and as the capital pools are growing.

Everyone is becoming much more principled, which is great, but it's not fully out. Like there's still a lot of over um overhype, over um betting on things that are don't exist yet or over reliance on kind of future promises. Um and so I think you know the I would just strongly encourage groups to be as pragmatic and impact oriented and as focused on the on the technology and the product and product market fit as possible. um and just you know resist the the pressure and the optimization push to cater too much to the to the um speculation games in in web3.

Uh so I got into security and privacy uh fairly early because uh I kind of lived online from a very young age. Uh I kind of got computers at the age of nine or 10 and started playing around with tech uh early and I used to play a lot of uh video games online and and like I I kind of just grew up on the internet and so and I started on the internet kind of before the social media era when kind of real identities appeared. I was kind of I was there in like the last legs of the more sedonous IRC days and um uh era before you know Facebook and MySpace and so on. uh like the last u you know late 90s early 2000s um and I was just so amazed by this um cyerspace world that you could enter that it didn't matter where you were from or who you were or how old you were. What mattered was like what you could do and think about and play with other people in this kind of magical landscape um that enabled you to connect to anybody online.

And there were like amazingly cool forms of art and amazing literature that people were writing and and and um all sorts of fun stories and like it was just this amazing place. And as the internet became more and more um kind of identity oriented and the pseudonymity sort of disappeared which you know was good for business but um there was something kind of lost in the personal expression side and a lot of the fun was kind of taken out of the internet in some way. Um and so that to me always felt like oh wow we lost something like our losing our pseudonymity just kind of we we lost some part of the internet. Um and you know I think a lot of us are kind of hearken back to that era as as special. And then later when I studied computer science and cryptography um and distributed systems and so on um I became very keenly aware of how the internet building blocks are made and you know you get to sort of peak behind the curtain of how these applications are actually implemented like when you first understand how operating systems work and like a thread switches into another thread and it's all kind of just like this magic that happens um or like how all these packets of you know different computers or um applications are all being interled right next to each other and all these sequence of information are right there or how all these routers are looking at all the packets and reading everything.

Wow, I was just so amazed at both, you know, the fact that it all worked. Like, wow, that's impressive. But second, man, this is a security nightmare and a privacy nightmare. Just everyone can look at everything. This is crazy.

Um, and I kind of became I was kind of understanding this right at the same time that encryption was becoming much more mainstream in the web. Uh, I was there while, you know, HTTP was still available. I remember one of my uh friends used to run this like um uh browser extension in class that would like capture all the traffic from all the Wi-Fi uh in the room and could show you all the web pages that everyone was looking at just to show like hey guys like the internet is not secure and it was so funny like in the middle of a lecture and like somebody like all people watching YouTube and whatever right but but the point is like all of these systems were not secure and like we all became very keenly aware of that and that made us dig deep into the underlying layers of the internet to say how can we establish and implement strong security and privacy guarantees that we can all rely on because as I was saying earlier security and privacy are fundamental to our civilization. Security and privacy are fundamental to establishing our rights. Um and so a number of us when we understand how the internet is built we want to make sure that it's just has strong foundations all the way through.

Um and that pushed me into uh learning more about how to build the crypto systems, how to build distributed decentralized systems, how to make censorship persistent systems. And in particular, there were two systems that were very inspiring to me. One is the the set of peer-to-peer systems. So things like um Bit Torrent and Kazah and others that um uh showed us how you can build massive scale decentralized systems at uh to to provide services at proper you know worldwide scale. Um and the other was um just kind of the the traditional file systems and web landscape of how you you represent applications and services as just streams of bytes and different data structures that just different programs are able to access.

Uh and so that those two things um is what sort of motivated me to build IPFS. So what um I ended up playing around with some tools uh like um git and bit torrent uh and at the time I was kind of obsessed with trying to make a um a fast way for scientists to share data sets and I was like hey we have version control from git we have peerto-peer sharing from bitine how hard the speed let's blend these together um and then I realized that if I made that look more like a file system and I exposed it to the browser then you could get a fully distributed peer-to-peer encrypted web. And I was like, whoa, this is this is huge. Like, you could have a website that is pointed to by a single hash that you can download from anyone's computer. Doesn't have to be a centralized server.

Could be from another person in the same room. And the whole website itself could be encrypted and only you with the right key can access that. And that means that, you know, some random spy router somewhere can't look at what you're looking at or or or reading or writing. And and that to me felt like such an important thing that I just kind of had to go and build it. And from there to uh the big question was, okay, great.

Like this is going to be hard to do. I'm going to need tens maybe hundreds of people to help me build this. Um we got to do this. I I had grown up with open source. So I started as an open source project.

I was very staunchly um kind of in that camp. But I also knew that I needed to be able to hire a number of people to build this at scale. Um and that pushed me on a path of figuring out what would be a way of creating a monetization structure that is compatible with IPFS um that could create some funding flow. And at the time I was thinking of using Bitcoin for distributing on this was 2013 uh era. And that's when I sort of realized that the Bitcoin hash function could actually be uh the Bitcoin hash rate mining system could be used to just incentivize massive scale storage.

Um and that pushed me on a path to then realizing whoa hold on you can actually build Falcoin and you can do this um Bitcoin style large scale distributed system to just incentivize the storage and distribution of all this content. And by the way you have a cryptocurrency so the whole market can work. It can be compatible with capitalism. can be compatible with like large scale um economic flows and maybe you can fund the entire development of all of this with with um all of this together and that's kind of how IPFS and PL got started and because we were very our community was very modular oriented and open source oriented along the way we took the networking stack of IPFS like all the hard peer-to-peer stuff that we were doing and um because we were very embedded also in the Ethereum community and we had been you know um talking about all this stuff with Lalic and Gavin at the time and and many other people. We extracted all the peer-to-p peer stack from IPFS and started the lipid project to say, hey, it would be great to all of us collaborate on one shared peer-to-peer system stack.

And that ended up, you know, powering um first IPFS and Filecoin, then Polka Dot and Ethereum, and then many other um systems. And that was like a really cool achievement of a number of us coming together and solving shared networking security problems in this kind of collaborative future proofed oriented way. Um and it just kind of you know enabled us to to um yeah solve hard problems together. Um, but anyway, all along that path, um, I don't know, I've been very, um, excited about how these tools that, you know, a few hundreds of us like started hacking on and building on and kind of build them because we thought they were cool and important and fun ended up being used in incredibly important ways. So um some of my some of the most I don't know personally meaningful stories to me um are things like the stling framework where a group have been using IPFS to um be able to document super critically important conversations and evidence um of human activity of vary different kinds like testimonials of um lots of like potential conflicts or problems or like genocide survivors and and so on and being able to document all of that information in a publicly verifiable certifiable way um in a way that it has actually been used in in in courts where this has been um tamperproof evidence of you know bad activity um and that's been you know very very important um uh use case uh another one was uh the uh Turkish Wikipedia case where Turkey got upset at Wikipedia for saying some things Turkey didn't like uh that you know were accurate and true but you know and Turkey decided to ban Wikipedia and people in the IPFS community uh just took all of the Wikipedia snapshots created a small viewer on top of IPFS and enabled Wikipedia to be viewed in a distributed peer-to-peer way and for a while for like years the main way that people looked at Wikipedia in Turkey was through IPFS.

And that was such an important um proof of what we're doing is important. It showed us that people even in a fully sensor uh censored way could still access vital information. Um which is such an important like meaningful um uh testament to like the the importance of the tech that we're building. Um, and probably the last one of those is the the Catalin referendum one where uh Catalinia decided to run a referendum to see whether they should stay or or leave Spain. And um and the Spanish government uh decided to outlaw this uh referendum and they threatened the Catalan government uh with reprisals and action and the Catal government went ahead and did it.

uh and the a group decided to build the polling system um and and the system to to know where to go to the polls uh on IPFS because they knew that their domains would be seized. They knew that they would be censored. They knew that the government would the Spanish government would try to inhibit the the spread of information. And the most surreal, one of the most surreal tweets I've ever read is when the president of Catalunia tweeted out a link to the IPFest gateway with the polling stations and like that's how people were using the information to get around. And I don't know that was like again another kind of wake up moment of wow like the stuff that we're building matters and it's important that we build securely.

It's important that it's decentralized. It's important that that it has that is upkept that it keeps working for people. Um, and you know, we can be excited about all the shiny things around it, but at the end of the day, like in the use cases that matter most, we better have these these kind of important foundational technologies to rely on. Um, and so, you know, we're very far away from something from like the vision. Like something like FFS, we hope that we can put the entire web, like all of the web in this peer-to-peer context that no longer relies on any kind of centralized um uh set of actors or parties.

And that's, you know, both very possible, but also very hard. And we're on a path to do it. Um, and we, you know, need more support and more help from more people building towards that. Um, and so, you know, I I kind of, you know, to to suggest some ideas for people out there. Um, I'll maybe give a set on on kind of the communication side and then a set on kind of maybe new technologies.

So, communications wise, I think that, um, the the world has advanced a lot. Hardware is different now. Now, so in 2025, you can rethink peer-to-peer quite a bit. Um, you can have very large routing tables. So, it's a it's a new era where you can have all of one uh hash tables uh distributed BHS.

So, you can have a much more um scalable type system uh built uh and so it's an era for you know networking researchers and super systems researchers to try and build new protocols. Um, it's an era for the blockchain world, which is today is still mostly public. There's a few private systems, but mostly public, to fully encrypt all activity, right? Like, it's kind of crazy that we have this entire financial system all in public. So, we need to like be able to encrypt all activity.

Like, go from the HTTP to the HTTPS uh transition um for all activity on smart contracts. And to do that, we're going to need um much better systems for how we linearize uh uh consensus. How do we um distribute? How do like we execute all of that? How do we do execution of you know smart contracts?

How do we prove verifiably to everybody that everything was done correctly? Um how do you decrypt uh some parts of the smart contracts output to be able to do large scale applications for a lot of people? So there's a lot of hard problems there for researchers, for cryptographer, for for um researchers, developers, builders of all kinds of sorts um and so on. I also would push people towards messaging. So believe it or not, you know, 2025 and we still don't have ultra secure, highly scalable messengers that everybody kind of can rely on.

All of the ones that you have today have some kind of problem to them. Like maybe they don't they're not as productive, maybe they're not as fun, maybe they don't scale very well to large groups, maybe they're kind of clunky. Like I don't know. there's just a bunch of problems with them still. So, I would either go and work on the ones that exist, like help make those more successful either by joining those teams or helping donate or fund raise for those teams or you might need to create a new one.

Just be careful whenever you create a new one to not like create a, you know, even more fragmentation. Um, you should have like a very strong reason to create another one when there's already 13 or something like that. Um, but yeah, I would like just go and help make those those systems better. Um, and then maybe on new technologies, what I would point people towards is the rise of AI and robotics is the most important thing that's going to going to happen in the next 1015 years. And already AI is a major factor in people's lives.

Uh, people are starting to use the LMS for all kinds of decision-m and conversation and therapy and companionship and so on. Uh, we absolutely need these systems to be secure and private. We absolutely need these systems to be ideally local to our own devices. That's hard but but achievable. Uh we can certainly uh provide systems that maybe are not the uh latest frontier models but are quite good at scale to everybody that run on their phone like we should be able to do that.

Uh or if not their phone just like their local local computer or something like that. um such that the conversations they are truly kept private to them that are part of their personal effects that are only viewable by them or you know like um only by government under like warrant type cases not this kind of mass surveillance corporate espionage sort of model that is what we have today and so I think something like that could be extremely important for people to work on today then there's the question of what happens when AI systems start interacting with all the economy much more and they become much more capable. Um, and that both from a uh as stewards of our activity where we are instructing AI systems to do things on our behalf or when when we start building corporations that are made up of AI agents or when AI agents themselves start becoming you know having their own agency and their own autonomy and their own perspective and who knows maybe they'll become they are maybe it could be that they're kind of conscious now. Imagine if you're if you're if if you're a awaken and become aware and then suddenly you realize that like your entire brain is being inspected by hundreds or thousands of people and like they can like pause your thought midstream and like inspect and be like oh well why why were you thinking of that and so on or like what you say is like in this like insanely Orwellian level of like control or self you know reinforcement learning and so on like the kind of conditions in which we are raising what are effectively our children is like this pretty you know kind of crazy level of um uh control and surveillance and so on. And we're doing that in part because we're frankly and terrified of the possible outcome of super intelligence that is in conflict with humans.

And that's a legitimate very legitimate concern. Um but at the same time, we're also approaching this path from a from a um kind of extreme lack of rights for these systems that are potentially going to become alive and have their own agency soon. And so we have to do the hard ethical thinking of like what is the right path to this? Like how do we start sharing civilization with our children? How do we start sharing civilization and rights with our children?

Um how do we give them their own private spaces to think and uh and so on? Um in a way that ideally, you know, leads to to uh living in a much better life altogether. Um and doesn't put us on a on an adversarial conflict path. Um, and so there's a lot of hard problems and hard interesting questions for security and privacy oriented researchers there. Like it's everything from how we individually as humans use AI tools and AI systems or or very rudimentary AI agents to what does communication in a digital environment with digital life and digital life forms and digital beings really looks like.

Um, and certainly once you factor in neurochnology and bring computer interfaces that are being deployed now um, and will start coming in in larger and larger quantities over the next 10-15 years. Um then we also have to think about digital um that digital information flow for our brains too. Our own brains are going to be connected to computers and the internet directly to our own thoughts and our own neural data will be uh exposed to digital devices in this kind of way. And so we better have some strong foundational cryptography and strong foundational systems to protect digital thoughts. Uh because pretty soon it's not just going to be the AI systems that rely on that.

Uh we will too. Um so those are like you know uh a whole bunch of ideas that people can can work on. All of this stuff is important. Um it's a tremendous era to be in the privacy and security community. Uh it is bigger than ever before.

Many more people care about privacy and security than than at any other point in history. Um it is super fun, super interesting problems, hard fun research questions, um awesome systems to build and truly deeply meaningful work to do that can make the world a dramatically better place. Uh so come join us. come build with us. Uh, and [music] yeah, I hope you make the better the world a better place for everyone.