Pushing Blockchain Limits with Hardware Accelerated Provable Computation - TOKEN2049 Dubai 2024

[Applause] all right good morning everybody I uh promised myself I would not be hung over for this talk but uh didn't exactly go that way it was very nice to see some old friends uh I was hanging out with the Polkka do guys last night I saw Dr Gavin wood Bjorn Ed hes Obie my old friends seen lots of old friends here it's very nice to see you all um so today we're going to talk about uh uh what we've been cooking up at Telos and uh hopefully it might resonate with some of you um interestingly enough I was just in Hong Kong at the uh web 3 Festival I gave a talk on recursive proofs um in particular something called snark Tor which actually Alberto our ZK lead our lead researcher is here developed and shortly after my talk vitalic came on stage and sort of somehow outlined basically our entire strategy um this is just a brief uh view into the history of Telos because a lot of people don't know or might not know much about Telos um it was actually launched in 2018 um in uh 2021 it came onto my radar if you guys remember uh the uh it was called the GU return data corruption via data copy exploit um in 2021 this thing almost brought down ethereum and actually it was discovered while fuzz testing the Telos evm so in some weird way uh Telos is very well connected to the history of ethereum of course once that exploit was discovered um rapid action was taken um the coordination was incredible and uh ethereum was saved um and more recently we launched nnar Tor and I'll talk a little bit about that um in some of our strategies going forward just a little bit of some of our memes we've got this really fast evm it's been around for five six years now 300 million plus blocks so I think as a R&D effort it's been an interesting um project um okay so getting into the more substantive part of this discussion what's sort of preventing Mass adoption um my good buddy Alberto often uses the example of uh renting a vehicle using a blockchain system in a crypto payment um well you wouldn't do that if it meant your license would stay on chain forever uh and so with ZK and in particular our view on recursive proofs you can actually just selectively disclose some proofs um and preserve your privacy and we think that essentially this abstracted to pretty much most real world use cases is one of the big hindrances uh preventing adoption and recently we've had lots and lots of discussion about ZK snarks um I guess probably a lot of people have a really good sense as far as uh some of the things ZK snarks can do again the emphasis from our standpoint is really about user data protection um but uh scalability as well is something we think very deeply about so some of the issues with uh DK snarks at the moment um obviously there are efficiency questions um when we look at the landscape across ethereum we see lots and lots of uh shared sequencers centralized provs a weird Patchwork of Frankenstein um Solutions I mean these are very very very hard problems and there's really talented hardworking dedicated people um working on them but uh these are very hard problems um okay so this kind of leads into um where we're going with all of this um and I appreciate you guys uh sticking with me through this um so in February Justin Drake uh tweeted picture of this uh chip and it's the uh first narc proving Asic um and then uh and then like I said said uh last week the other week I was actually in Hong Kong I spoke at the web 3 Festival um I didn't sort of describe everything that we're doing just a little piece of it and vitalic did his uh wonderful keynote and somehow laid it all out um we couldn't believe it but uh basically ZK snarks aggregation Hardware software co-design and I think that's sort of one of the really interesting and important um themes that's emerging right now now that uh everybody in ethereum is starting to pay attention to uh and then our friends Axel um wonderful people hardware company um also focused on snark proving A6 okay so coming back to Telos um if you guys are interested in checking out the white paper this is snark tour um this is our uh recursive proof aggregator we believe that recursive proofs will play a very very important role in the future to come um particularly if we're talking about Hardware software codesign and especially if we're talking about Hardware acceleration okay so this is basically one of the things I'm very uh fortunate and happy to announce today uh Telos will be launching its Hardware accelerated zkm L2 um this thing will be uh featuring snark Tor and we hope to provide a data protection um hub for ethereum and Beyond um and I'll talk a little bit more about what exactly we're doing and more importantly uh some of our new partners um okay so yeah this stuff is very very very hard it's very hard to build highly performant ZK systems um there are extremely talented people very hardworking people dedicated to this stuff um there are across ethereum uh and other networks um there's tremendous talent for example when you look at something like Alo uh when you look at moned when you look at some of these other teams um they've raised tremendous amounts of money from very smart people they're attacking very hard problems and I think um this is my personal opinion but I think the future of ethereum um is going to necessarily depend on Hardware software codesign the extent to which uh we can effectively and efficiently prove snarks with as6 in particular and then the extent to which we can make all of that function very well across the ecosystem um and so this is sort of our view on it at least part of it at a high level um in particular um ZK rollup efficiently verified on ethereum um okay so you know a lot of you know me or I guess some of you have been my friends for a very long time I've been in ethereum since the beginning I was kind of originally the consensus dude I guess I spent some time at polygon as well and uh I remember the early days when we started off as a proof of work Network and then along the way we sort of developed this narrative that actually as6 are really bad and destroying the planet and sort sort of like we have to move away from that um but I think think that there's been a lot of maturity um there's tremendous innovation in so far as um clean Renewable Power all sorts of different uh projects out there in the ecosystem which are incentivizing away the problem that used to be perceived in the past as far as as6 not being something that we could uh value um and so coming back to as6 I think is part of the future of ethereum um and so Hardware software codesign is going to be very critical if we're going to attack the uh ZK problem in an efficient way we have to think about that from the ground up how do we um design the two in conjunction with each other such that when we accelerate it's not just some functionalities that sort of perform better but the whole system um achieves significant significant Improvement so I'm also very uh very happy and privileged and honored to uh introduce some of our new partners today Telos is going to be embarking on this journey with a coalition of very good friends um starting with our friends at ponos ponos technology um maybe not super well known in let's say the broader um ethereum mainstream or retail world but when it comes to the bleeding edge as far as researchers are concerned um these are some of the best in the business very hardworking people and they've been focused on ZK research for many years now um and I'm very very happy and honored to announce them as our partners um not too long ago we also announced Jerome uh many of you know Jerome Jerome is the founder of a thean France the founder of SEC and also the founder of kth uh he's come on on board as our evm adviser um our good friends at atka this is uh one of the leading bleeding edge incubators in Paris they incubated groups like Moro and mango for example very good friends of ours uh also advising us and then digital mob um some of you might remember these guys as basically consensus Romania um Milad and dragos and some of these guys they built many many um key projects in ethereum over over the last 10 years um so this is basically what our uh basically what the new Telos looks like um and together we're going to be focused on Hardware software codesign and tackling these problems together um just a bit more about pono's technology um so the CEO of ponos is a guy named Slobodan uh slow very very good friend of mine he lives in Lugano uh there's a wonderful University there um actually Lugano is very interesting for a number of reasons um I was just there uh last month we had some conversations with the plan B guys and uh I have to say um the work they're doing Lugano is spectacular there is a density of tremendous Talent there lots and lots of researchers um really great people working on really tough problems um and so ponos has been uh actually these guys were um if some of you might remember part of the original polygon ZK uh team and they've worked with a number of other uh groups across ethereum and other networks as well for quite some time um intensely uh focused on basically uh the intersection of Hardware software code design nzk uh just a little quote from slow um give you a second to take a look at it and so we believe that uh so we're betting everything on ZK um and we're taking a hardware software code design approach we think that um and uh and we were just talking backstage and and it was like sort of not much not much and then suddenly everything all at once and that's how we think this is going to happen in other words when the infrastructure reaches a point uh where the efficiency and the scalability is there it's going to be very very difficult to use anything other than a fully provable system okay where every single computational step storage operation every single action that happens is fully provable um this is I think going to be um necessary and so uh Slobodan is a very elegant speaker and I'm sure he'll be participating in a lot more conferences and I think you guys will hear a lot more from him soon uh these are just some thoughts from Alberto who's actually here in the front row um and Alberto's uh is a wonderful person he's done a great job snark tour is something he's been working on for some time um in conjunction with uh a couple iog researchers as well really wonderful people and Alberto has helped me understand and all of us understand the extent to which um user privacy is fundamental okay and it's not just fundamental for some of the reasons we've talked about in the past um it's fundamental because it unlocks the vast majority of all the possible use cases out there which we currently can't serve um and and so I would even sort of expand a a little bit more on the uh car rental example and I would come back to uh identity um and so you can establish your can canonical identity on chain and you can have Persona constructs that describe various attributes about your identity and those Persona constructs can collect attestations and you can imagine over time this forms the notion of reputation and now that reputation can be persistent portable and transitive meaning you can take it with you you can log into any application anywhere in the world with your identity you don't need um some intermediary like Google or Apple to do that for you um or Facebook um however this is sort of an idealistic Vision that we cannot realize without this technology because you're not ever going to put the full extent of your identity on chain um if you can't selectively disclose okay if you can't give granular authentication in other words if everything is just there for for everybody to see we can keep talking about these use cases and we can keep institutionalsales or rather data protection um user data protection we're just not going to realize those use cases um and so I know lots of people talk about dpen as an example I think it's a great emerging um sector lot lot of great real world asset stuff happening and so when you think about something like that it's not primary issuance where Banks make money it's on secondary trading okay and we can construct things like permissioned actors on permissionless systems um or we can construct fully provable infrastructure that gives you full control and sovereignty over your identity against which you can have various Persona constructs that could look like u a business person for example an Institutional Persona a certain Department with an institution which can have uh the appropriate attestations which then give you um access to whatever it is that you're doing in other words you can have granular authentication native into the system um and so you can see if we realize ZK at scale and we believe that looks like Hardware software codesign leveraging things like recursive proofs then you can Implement things like self- Sovereign identity a tremendous scale and unlock the full potential of what this technology can do let's say in the case of the institutional um Jesse Schulman is a CTO of Telos wonderful guy lives in uh spends a lot of his time in California but he's all over the world um he's been with Telos for quite some time um and uh yeah the uh it's just it's just kind of funny to me because you know ethereum has been such an important part of my life and many of my friends uh a lot of whom are here uh and it and and until those came onto my radar actually because of that exploit in 2021 which Sentinel by the way to their credit wonderful uh Auditors um discovered and uh sort of in some way saved ethereum at that time um so just a little bit about our ZK Journey um we started doing research particularly on recur of proofs uh some time ago um Alberto was thinking very intensely with that alongside Danielle one of our other ZK research leads and uh and Jesse they spent some time together in San Diego we had a workshop um we published a snar Tor white paper in January and uh over the last four five six months we've been collaborating intensely with our partners at ponos technology uh we actually had a intense Workshop um uh in March it was wonderful we got spend some time together in Milan uh today we're making this announcement as a sort of First Step introducing everything um and where are we going I guess or where are we going to try to go so um we're also a title sponsor at sec in July um some of you might remember me as the guy that sponsored SEC with my face which was kind of like a funny thing Jerome's one of my best friends and nobody would give him any money during the co stuff so I ended up sponsoring it and we just donate money to charity this year I've given that sponsorship to Telos and so we'll be there at sec we're going to be demoing a proof of concept um basically snark tour accelerated and maybe playing a game or something like that um we're going to be releasing a new new white paper uh shortly um we're going to offer our existing Telos uh Community the opportunity uh to participate um with their existing tokens so we will not um we won't be doing an Ico we've never done an Ico we won't be doing anything like that or any kind of token sale um it's just going to be something that's available to our existing Community uh and then and then going into the early part of next year hopefully the L later part of this year but we'll see um we're going to start releasing our first iteration um of a hardware software code design um zkv ml2 um thank you all very much for your time and uh here's just a few links if you guys want to follow on Twitter or connect very happy to meet with you all um so super cool and thank you very much [Applause]