David Chaum (Devcon5)

Wonderful to be here with all of you at another Def Con again, and this time I'm delighted to be able to speak. And I thought I would take you back to Berkeley, 1982, when I was a graduate student there, reading Science magazine, and it said that the director of the United States National Security Agency had, it was revealed, written letters to the heads of IEEE and ACM and so on, threatening them with draconian the full force of the federal government if they were to even have a session on cryptography, let alone a whole conference on it. So, what do you think I did? Yeah, organize a conference on cryptography. And I invited everyone.

I did it sort of covertly, and then I was like on stage announce, you know, welcoming everyone as the general chairman of the event, and and the front row was occupied by people who refused to give their organizational affiliation, but all mysteriously happened to live in Laurel, Maryland. And so, when I announced that as an extra benefit of having paid their $100 to attend, they were now members of a new thing, the International Association for Cryptologic Research. And it was a great pleasure to see them all turn green. Because everyone knew that it was over. There was the the UN protects international scientific associations, and so IACR has had uh several conferences every year.

Now it's just three per year, plus a bunch of workshops, and publishes a journal with Springer Verlag. So, it was over. Cryptography was set free. However, we are now at a very different and more fundamental critical juncture. It's less reversible and much more important.

And that is evidenced by the fact that most of the public is starting to become aware of metadata and the dangers of metadata and how it's being used to steal our metal liberty from us. Which includes then all the subsidiary liberties, right? It's It's the big one. Lose that, and it's game over. So, turns out in Berkeley in '82, I was actually working on cryptography for the purpose of finding metadata shredding techniques.

So, the first thing I did was to create like a toy example where some privacy would be needed, so I used voting. Let me just try to explain roughly how this works just to give you a sense that let's say there's a hundred voters, each would take their ballot and multiply encrypt it and publish it a let's say associated with their IP address. Okay? And then they would those all those encrypted ballots would be sent to the first node, which would have the unique ability through its private key to decrypt the outer layer on all those ballots, and then it would shuffle them secretly in a random order that only it knew, and send them on to the next node, and that node would do the same thing, decrypt the layer that only it could decrypt, randomly permute, and send on. And so, the final batch that comes out and is published includes all the cleartext ballots.

But, I say that we have vertical metadata protection because no one knows which IP address is associated with which ballot unless they can compromise all of the nodes. Because each node uses a totally independent uniform randomly chosen permutation. So, this became known as mixing, and there's thousands of papers about it, and it's generally thought of as being used to send messages. See, it it destroys the social graph, the who talks to who and when. That's what we used to call traffic analysis, now it's been rebranded as metadata.

So, then I thought, well, we need some kind of way to pay and stuff like that. And if you I look at regular like the kind of say account-based payments we have like on blockchain today, from wallet ID to wallet ID to wallet ID, what that does is it publishes and proves the whole meta the whole the whole graph, right? It's all it reveals and proves all the metadata. So, that's not a suitable type of of payment technology. Turns out that what I proposed is is actually good enough, not perfect, but it it can be perfected.

So, essentially, it's a simple notion that instead of accounts use digital bearer instruments that are let's say binary denominated. So, 1 cent, 2 cent, 4 cent, 8 cent, 16 and so on. So, each denomination has its own corresponding signature that validates it. And that way as the the these digital tokens migrate through different transactions, they're regrouped and the metadata kind of fades away. Now, we can perfect this by actually washing a whole bunch like all the 1 cent coins through a special mix that then makes every 1 cent coin unlinkable to the the ones that existed previously.

So, we can do that overnight for all coins and then perfect privacy because we have the vertical unlinking by through the mixing of the say IP MAC or your location or whatever to the transactions or the messages that being delivered and then we have the horizontal unlikability or metadata shredding between the say the transactions themselves. So, the the payment system actually was uh built in that you probably heard of it famously in the mid-90s and I airdropped to anyone who wanted a uh who would open up uh a digital shop on the web to receive payments a 100 cyberbucks and you can see the whole the museum of this up on my website chaum.com and bunch of the shops and uh and so forth and then later some banks wanted to issue e-cash so we let Deutsche Bank do it and some other banks in other countries. So, we had the Deutschmarks and Australian dollars, US dollars and so on. And here in Japan, Nomura Research recently was revealed secretly they they licensed eCash from us, but they wouldn't tell us what they were doing with it.

All their employees were allowed to use it and access their bank accounts and buy like Nomura swag and get reimbursed for travel and all this stuff. So, it was actually used here in Japan, but secretly. And so that was uh pretty neat, but now we are in a world where everyone has smartphones and the killer app I can't it's kind of a sick joke, but it's like is is WeChat. Right? So, anyone see the humor in that?

No? Okay, some people get it. Anyways, um have to live there to really appreciate it. So, um this is sort of indisputable because all the major social uh media messaging systems are moving towards payments integrated with messaging. Because that's what consumers want.

They want those two in the same name space and then they want some mini apps or dapps in that same name space, the popular ones. This is this is proven to be what like billions of people actually want. So, what we need to do here is to get rid of this giant you know, metadata capture engine is create a replacement product, something that people can the ordinary users can easily just switch to that does the same stuff for them on their phone as say a WeChat or another messaging platform with payments integrated and and and apps. And but that is based on the metadata shredding technology. So, it'll be just as easy to use and work in a similar way, but the metadata will never even be constituted.

It will be shredded in real time in its distributed form before it's actually able to be abused in any way. However, the the the technologies that I showed you that I developed in the '80s and we implemented in the '90s aren't really suitable for like smartphone to smartphone transactions. There's their performance and other issues. And especially when it's used in a mass scale and because of the kind of speed of transaction latencies that consumers are familiar with in messaging today, something some real breakthroughs are needed. And so, I was uh not really working on this stuff when the Snowden revelations came out.

And and then I was so shocked and I went back to the you know, those crypto conferences that I founded and talked to a lot of people there and I was really quite became quite concerned because it was surprising that they were much more interested in their jobs and in publications and all this and even though they said they were there to protect you know, people from whatever abuse through cryptography, they it no one was really that influenced by the Snowden revelations and I but I was. And so, I took it upon myself to speed up the mixing because it's pretty cumbersome and slow and I found a way to speed it up by over by a factor of a thousand. Took some real work. And no one since I'd published it in 79 as a technical report of CACM 82, no one actually ever found a way to speed it up even though there's thousands of references and bunches of there's a conference every year that's mainly used to be mainly about it. And so the only speed up that occurred was when someone substitute ElGamal for which came out 4 years after I published it for the RSA kind of crypto, but I had defined it in a way that that that RSA was just an example.

So um anyway, so I let me show you some of these breakthroughs now that are enabling us to do exactly what I'm saying. If we pull together as a community, we can create a replacement product for all of these messaging and and combined with payment systems. So we actually I have to say some of you may have have heard of Elixir. That's the our our the high-speed mixing technology which is is actually running in in open alpha and we'll we'll see something about that shortly. And then we have Praxis which is our new very different quantum resistant consensus and payments technology.

And these are sister companies and when you combine them, you get this kind of app that I'm talking about. So let me just now show you a little bit about the the new approach in a in a simplified kind of amusing manner. Suppose Alice wants to send a message to Bob, but she doesn't want anyone to know that the two of them are communicating. Message content can easily be protected by so-called end-to-end encryption, but this does not protect the information about who's talking to whom and when, the metadata, which is increasingly recognized as far more revealing and more challenging to protect. Each member of a team of nodes, in order to protect the metadata, successively shuffles the batch of encrypted messages using its container of secretly arranged tubes and sending the messages without delay through to the next team member.

Even just a single node can keep senders from being linked to recipients. Alice also gets a receipt informing her confidentially that her message was provided to Bob. Team members then destroy their secret pattern of tubes, making way for a new team ready for a new batch. Earlier, each node is chosen independently as a kind of random secret key which input tube to connect to which output tube. Elixxir's breakthrough over the type of messaging I open sourced in the '80s is a way that enables almost all the work to be done well in advance, yielding the only known way to provide real-time metadata protection smartphone to smartphone.

Leap can also pay you me by sending what appears to be an ordinary message, but that actually contains numbers that serve like metal coins and paper money. They're called such denominated unforgable numbers, digital cash, first issued by Digicash in the '90s. Elixxir's breakthrough improvement on the original digital cash now makes it fully distributed and quantum resistant. Elixxir thus is able to uniquely provide a new ultimate metadata level protection of confidentiality in all your online communications and payments, for the first time giving you a protected sphere protecting your digital world today and your digital future. Yeah, thanks.

So, let me just give you an update. Uh we we announced our little effort like a year ago in Singapore. And it's been kind of a wild ride since then. Been traveling all over and, you know, connecting with the community, learning what people want, learning how to understand all the issues and how people think about them. And we built up actually several really great teams, and that's why there are two different entities.

We have a R&D lab in the in the Cayman Islands. And no one else seems to have that, but we can bring people from anywhere in the world there as long as they're not like they don't have lung disease or something. And uh so we we're building up the consensus and the payment work there with our plan to release the white paper this year. And I assure you that when you see it, it will be unlike anything you've seen before, and I hope you'll be as enthusiastic about it as we are. We have received a ton of support from the broader community.

We've got over like 680 node operators that have participated in our open selection process. Now, what do I mean by that? We had a survey first to find out what people wanted from the process. Then we built the process according to what the community wanted. We ran the process, then we had a dispute resolution procedure and everything in it.

And at the end, we got 680 nodes, people that want to support us by running nodes for free in our beta system. And we have divided them into three teams so we can phase them in. So, that's really great, and that well distributed around the planet everything that was one of the criteria. We have also been building out our community. So we've got several thousand.

I don't know what the number is exactly now. People who have signed up. Received our app downloaded our app on Android and iOS which is kind of like a waiting list to be the first users of the messaging system and to be able to participate in other ways. So I would urge you to get it as soon as you can if you have the opportunity to do that which I will give you. And we are also live in the public alpha as I mentioned and when you download the app you can see the dashboard of the alpha network and watch it all all the performance and stuff.

So the main purpose of the alpha is really to perfect the the performance but we've exposed the APIs and the libraries to our our office SDK partners because we don't have the bandwidth to support the whole community yet. So we're kind of rolling that out and trying to get everything really nice and neat before we enlarge the the sphere of people have access to it. So I've been traveling around Asia on this last few weeks announcing mostly at meetups and at conferences not no conferences biggest this one um that our our Cayman lab has come has come out of stealth and that we will have the white paper and that the payment system it houses quantum resistant and it will be able to do the kind of when you mix that when you combine that with the elixir you'll have I hate to have to say it but the world's best money. No question hands down. So that's um kind of what we've been up to over the last year, and here's a sort of preview of the app.

Uh there's a lot of things in it. If you look at those those scroll the menu bar horizontally, you'll see uh you can keep track of all the things we're doing, and uh there's going to be a lot more interesting stuff there. So, I uh let me put up this uh barcode or QR code, if you will, and uh I'd urge you to all uh test Do you think that will that QR code will work? Why don't you see if your QR code reader is capable of reading that code, and then decide if you want to download the the app of your choice, and then you'll have a position in line, and be able to see what we're doing, and watch the dashboard, and uh join our community. So, tha- thank you very very much for uh your attention.

It's a pleasure, and I'll be around uh so, look forward to having a chance to uh to chat with you, and uh thanks again.