1 00:00:13,900 --> 00:00:17,500 This is the second part of a two-part series on the state of 2 00:00:17,500 --> 00:00:21,000 the ZK ecosystem. And we're speaking with 3 00:00:21,000 --> 00:00:25,300 annerose, the founder of ZK podcast, they kzk Summit ZK, 4 00:00:25,300 --> 00:00:29,200 validator, ZK, hack, and probably a bunch of other zks, 5 00:00:29,400 --> 00:00:32,500 as well as As Kobe, Kobe gherkin, who is the head of 6 00:00:32,500 --> 00:00:36,200 research at geometry and works with Anna and seek a validator 7 00:00:36,200 --> 00:00:40,300 and seek a hack and is also cryptography, advised it Sealab. 8 00:00:40,600 --> 00:00:44,600 If you've missed the first part of this double episode, go back 9 00:00:44,600 --> 00:00:48,100 to last week's and maybe start with that one because it is 10 00:00:48,100 --> 00:00:54,000 segue quite nicely from there without much further Ado. 11 00:00:54,000 --> 00:00:57,600 Let me tell you about our sponsors this week. 12 00:00:57,900 --> 00:00:59,700 Our sponsor this week as Tallyho. 13 00:01:00,000 --> 00:01:03,000 Kelly how is an open-source wallet, redefining the wallet as 14 00:01:03,000 --> 00:01:07,000 a public good with Teddy, how he can safely connect to defy and 15 00:01:07,000 --> 00:01:10,200 web 3 plus a lot more. You can view a 10 ft is in the 16 00:01:10,200 --> 00:01:13,700 wallet across a theorem polygon optimism and a Bertram they have 17 00:01:13,700 --> 00:01:15,900 Ledger support. You can swap between assets and 18 00:01:15,900 --> 00:01:17,500 view. All of your account balances 19 00:01:17,500 --> 00:01:22,200 across in their portfolio tab. Currently, they're running a 20 00:01:22,208 --> 00:01:25,100 layer to Adventure that rewards users for exploring the 21 00:01:25,100 --> 00:01:27,600 arbitrary ecosystem. With hey Leo, you can get a 22 00:01:27,607 --> 00:01:31,200 space dog and ft and be entered into the giveaway for another 23 00:01:31,200 --> 00:01:34,700 nft head over to Terry dot cash and check it out. 24 00:01:35,600 --> 00:01:38,900 And with that, let's head over to the second part of the 25 00:01:38,900 --> 00:01:44,300 interview with Kobe and Anna. So I think it might be a perfect 26 00:01:44,300 --> 00:01:46,400 time to start talking applications because I think 27 00:01:46,400 --> 00:01:50,000 that was where you wanted. You wanted to sort of hear some 28 00:01:50,000 --> 00:01:52,900 of these like novel, cool ideas, right? 29 00:01:52,900 --> 00:01:55,000 Absolutely, that's what I'm here for. 30 00:01:57,300 --> 00:02:00,100 All right. So yeah, we did before this 31 00:02:00,200 --> 00:02:02,100 interview. We compiled a bit of a list and 32 00:02:02,100 --> 00:02:05,200 a lot of these are coming, by the way, from these interviews 33 00:02:05,200 --> 00:02:07,900 that we've done recently and like trying to pull together 34 00:02:07,900 --> 00:02:10,500 like what you know what Dan? For example, this interview we 35 00:02:10,500 --> 00:02:14,200 did with Dan Bonet. It was episode 56 like four 36 00:02:14,200 --> 00:02:16,200 weeks ago. The wrong number. 37 00:02:16,200 --> 00:02:19,100 Yeah, it's a good number but yeah, it's a real mess up. 38 00:02:21,200 --> 00:02:23,800 We in that he went through a bunch of examples so we can 39 00:02:23,800 --> 00:02:26,300 definitely share those. I also know in the interview I 40 00:02:26,300 --> 00:02:29,500 did with Matt Greene, I learned about some new ideas and 41 00:02:29,500 --> 00:02:33,300 generally, I mean, look, if you look at what 0x park releases in 42 00:02:33,300 --> 00:02:36,300 their videos, when they do showcases, that's like an 43 00:02:36,300 --> 00:02:41,000 amazing space to find out about like new ideas and not at the 44 00:02:41,000 --> 00:02:45,000 truth is not all. Of them are feasible at this 45 00:02:45,000 --> 00:02:45,900 moment. I think. 46 00:02:45,900 --> 00:02:49,700 But still like, set the goal, like put the, you know, flag in 47 00:02:49,700 --> 00:02:51,700 the sand, in the future, and then, let's build those tools. 48 00:02:51,700 --> 00:02:53,300 So we can actually make these things possible. 49 00:02:54,400 --> 00:02:58,000 I'm curious, if you've been keeping up to date on this Kobe, 50 00:02:58,000 --> 00:03:00,800 have you been paying attention to like Constitution doubt to 51 00:03:02,000 --> 00:03:06,800 just a bit? So like I like the concept of 52 00:03:08,600 --> 00:03:13,800 this Anonymous multisig. So That's kind of the what the 53 00:03:14,000 --> 00:03:21,000 new Clio team is doing so they're using privacy technology 54 00:03:21,000 --> 00:03:25,400 to like, one of the things that they're doing is to hide who 55 00:03:25,400 --> 00:03:28,400 exactly participate in multisig. But Inquisition doubt, they 56 00:03:28,400 --> 00:03:33,000 using it to hide the total amount that you would be. 57 00:03:33,200 --> 00:03:34,800 Yeah, exactly. The using this to hide the 58 00:03:34,808 --> 00:03:37,600 amount that it would be because we don't know what happened with 59 00:03:37,600 --> 00:03:41,500 the first constitution though. We're just because the amount of 60 00:03:41,500 --> 00:03:44,900 public. It was just outbid and so yeah, 61 00:03:44,900 --> 00:03:49,900 I think they're using this in a very cool way to how to hide 62 00:03:49,900 --> 00:03:52,600 this, but yeah, that's the extent. 63 00:03:52,600 --> 00:03:56,700 They know it. So yeah, this was the private 64 00:03:56,700 --> 00:04:00,200 Dows. One of the reasons for this was 65 00:04:00,600 --> 00:04:02,600 like and this is what we talked about in that episode with Dan 66 00:04:02,600 --> 00:04:06,100 was, you know, when you're doing bedding, when you're actually in 67 00:04:06,100 --> 00:04:10,100 an auction and you can see what other people have as their Max 68 00:04:10,100 --> 00:04:12,400 than it's pretty easy to win the auction and I think that's kind 69 00:04:12,400 --> 00:04:14,400 of what happened. So yeah, this is a cool. 70 00:04:14,600 --> 00:04:16,300 This is a really cool. Use case for this. 71 00:04:17,100 --> 00:04:19,899 We do have penumbra which is like a privacy-preserving deck 72 00:04:19,899 --> 00:04:21,700 so that's like and there's been a lot. 73 00:04:21,700 --> 00:04:24,300 Like Manta also had proposed something like this. 74 00:04:24,300 --> 00:04:26,600 There's a lot. Different like ways to think 75 00:04:26,600 --> 00:04:31,200 about what in add X should be private and what can be private 76 00:04:31,200 --> 00:04:32,800 and what, maybe can't be private. 77 00:04:33,400 --> 00:04:36,300 And so those that's just been an interesting conversation space 78 00:04:36,300 --> 00:04:40,300 and like development space to Define that I think penumbra's 79 00:04:40,300 --> 00:04:44,500 Solutions really cool. And I just in terms of status 80 00:04:44,500 --> 00:04:48,100 though, like, you know, they are still test Nets and like there's 81 00:04:48,100 --> 00:04:50,600 a lot to build before you can actually use these things. 82 00:04:52,100 --> 00:04:54,300 Maybe we can mention like some cool experiments that we've seen 83 00:04:54,300 --> 00:04:58,400 from Xerox, Parc, right. So I think one thing that was 84 00:04:58,400 --> 00:05:01,700 really cool. Was this hay and on experiment. 85 00:05:02,400 --> 00:05:07,200 So maybe you've seen that on Twitter, but they they've been 86 00:05:07,200 --> 00:05:14,200 taking these groups that maybe have some relation among the 87 00:05:14,200 --> 00:05:15,900 people. So for example, all the people 88 00:05:15,900 --> 00:05:19,300 that participated in the Dow and got hacked, you know? 89 00:05:19,300 --> 00:05:26,400 So and And now, all of these people can send a message as a 90 00:05:26,407 --> 00:05:31,600 participant or as a member in that group without revealing who 91 00:05:31,600 --> 00:05:34,600 you are. So this is something that 92 00:05:34,600 --> 00:05:37,600 they've published which is which is really nice. 93 00:05:37,600 --> 00:05:43,300 Like you can you can do that and some other cool things that I 94 00:05:43,300 --> 00:05:47,100 think technically it's very advanced. 95 00:05:48,600 --> 00:05:52,900 That's called ethos actually. I don't remember what? 96 00:05:53,000 --> 00:05:57,100 Okay, it's degree of separation. That's the the acronym. 97 00:05:57,800 --> 00:06:03,300 So it allows you to prove. Oh yeah, who you know? 98 00:06:03,500 --> 00:06:05,800 And like who that person knows? And so on. 99 00:06:05,800 --> 00:06:10,200 So how far in terms of the chain of people are you from vitalik? 100 00:06:10,200 --> 00:06:13,000 That was the demo of this, they did. 101 00:06:13,700 --> 00:06:18,600 So it uses actually like, recursive proofs and Extremely 102 00:06:18,600 --> 00:06:23,200 complex Machinery, inside snarks, to make that happen, and 103 00:06:23,400 --> 00:06:25,300 just to prove this degree of separation. 104 00:06:25,300 --> 00:06:30,800 So that's something unique. And another cool experiment that 105 00:06:31,300 --> 00:06:35,200 they did was cabal, which allowed you to create these 106 00:06:37,800 --> 00:06:40,500 discs or channels or telegram channels. 107 00:06:40,500 --> 00:06:48,100 That have pseudonyms, that are derived from again a group Of 108 00:06:48,100 --> 00:06:52,700 people that, you know beforehand. 109 00:06:52,800 --> 00:06:57,500 But when they join the channel the nobody knows who they were 110 00:06:57,600 --> 00:07:00,200 from that group. So those are some cool 111 00:07:00,200 --> 00:07:07,800 experiments that we've seen. And I think we could also 112 00:07:07,800 --> 00:07:10,100 mention some of them experiments, right? 113 00:07:10,500 --> 00:07:12,500 Yeah. So Dan told us about this 114 00:07:12,800 --> 00:07:16,900 project where they are trying to almost do like image provenance 115 00:07:17,500 --> 00:07:22,100 even after it's been cropped or changed, so trying to prove 116 00:07:22,100 --> 00:07:24,500 that. Like and I think the example are 117 00:07:24,500 --> 00:07:26,500 the use case here is like newspapers that will crop an 118 00:07:26,500 --> 00:07:30,500 image but you want to know that it actually is from an original 119 00:07:31,300 --> 00:07:34,400 Physically or a digital file or Digital Image. 120 00:07:34,600 --> 00:07:38,100 And so there's a zkp to like, prove that it's actually from 121 00:07:38,100 --> 00:07:42,200 that original one and I know on our show we talked about maybe 122 00:07:42,200 --> 00:07:46,400 even extending this to music or to film and like being able to 123 00:07:46,400 --> 00:07:50,300 prove like if you edit film that it's connected to an earlier 124 00:07:50,300 --> 00:07:54,000 version of that video footage, which is pretty exciting. 125 00:07:54,000 --> 00:07:56,700 This is this is like one of these non-locking ones. 126 00:07:57,300 --> 00:08:00,000 There is another non blockchain. One to mention this is mentioned 127 00:08:00,000 --> 00:08:01,600 on the episode. I did with Matt Greene where 128 00:08:01,900 --> 00:08:04,600 apparently, and I don't know any real details on this, but 129 00:08:04,600 --> 00:08:06,800 apparently, Google is looking to use EK PS2. 130 00:08:07,300 --> 00:08:11,300 Like comply with GDP are rules, but still use private data. 131 00:08:11,300 --> 00:08:15,300 So it's almost like to go around some of the privacy rules like 132 00:08:15,300 --> 00:08:17,600 using it to like oh we can make it private. 133 00:08:17,600 --> 00:08:19,300 Was you keep these? But like actually doing things 134 00:08:19,300 --> 00:08:23,600 that normally they wouldn't they shouldn't be doing right? 135 00:08:23,600 --> 00:08:26,000 I don't know I actually don't know enough about this to cash 136 00:08:26,000 --> 00:08:32,200 that much shade, but yeah, I want to mention also like a 137 00:08:32,208 --> 00:08:35,600 project that is live, which uses it and that's his Mo. 138 00:08:36,000 --> 00:08:38,799 So we actually had them present, as you can hack, and actually 139 00:08:38,799 --> 00:08:42,700 create these attestation badges. It's pretty cool. 140 00:08:42,700 --> 00:08:45,100 Like it's, I don't know. Kobe. 141 00:08:45,100 --> 00:08:47,400 Do you want to, you want to go through how it works? 142 00:08:47,800 --> 00:08:50,100 I can help. Yeah, sure. 143 00:08:50,300 --> 00:08:55,200 Actually, like, it's interesting because when they started out, 144 00:08:55,200 --> 00:08:59,200 like I was talking to the seasonal Founders and like, 145 00:08:59,700 --> 00:09:03,000 actually I think this is is when I was looking at the Stark mad. 146 00:09:03,000 --> 00:09:08,400 So I was doing a project based on blind signatures that allow 147 00:09:08,400 --> 00:09:13,800 you to get an attestation from like some issuer. 148 00:09:13,900 --> 00:09:16,900 But after you get this at the station, the issuer cannot track 149 00:09:16,900 --> 00:09:19,500 you anymore because you're blind it in the process. 150 00:09:19,500 --> 00:09:23,500 So it was, it was a construction that I learned from privacy 151 00:09:23,500 --> 00:09:27,900 paths which is a project from cloudflare from years ago, and 152 00:09:27,900 --> 00:09:30,900 but I think nobody really played it in. 153 00:09:31,100 --> 00:09:33,900 This context. So that that was fun. 154 00:09:34,500 --> 00:09:38,000 And this is my people also created something of that sort 155 00:09:38,000 --> 00:09:42,900 when they started out and that's what allowed them to create 156 00:09:43,200 --> 00:09:47,200 these badges, which are privacy-preserving. 157 00:09:47,900 --> 00:09:52,000 So you have some issues that maybe you trust or put some 158 00:09:52,000 --> 00:09:56,100 trust in but nobody can track. You not even them after you do 159 00:09:56,100 --> 00:10:02,200 that, but they have sense developed much more And they're 160 00:10:02,200 --> 00:10:07,300 using SDK and membership proofs. Like the ones that are others, 161 00:10:07,300 --> 00:10:11,000 like the cash and the projects on lithium Foundation are using 162 00:10:11,700 --> 00:10:17,100 and basically everyone that does privacy-preserving transfers and 163 00:10:17,100 --> 00:10:23,700 it's there now doing other kinds of at the stations and that have 164 00:10:23,700 --> 00:10:27,700 even less trusting them but still preserve. 165 00:10:27,700 --> 00:10:32,000 This unlink ability properties. They seem to emerge like people 166 00:10:32,000 --> 00:10:36,800 kind of treat them like Poe apps but they have a month Christine. 167 00:10:38,300 --> 00:10:40,400 Yeah. Like something private and 168 00:10:40,400 --> 00:10:42,200 detached underneath which is cool. 169 00:10:43,600 --> 00:10:46,000 Yeah no this is most one of the projects we've been meaning to 170 00:10:46,000 --> 00:10:48,400 have one for a while. So yeah. 171 00:10:49,100 --> 00:10:50,500 Nice. Yeah. 172 00:10:50,500 --> 00:10:52,500 Other like so there's some there's some projects that have 173 00:10:52,500 --> 00:10:56,000 been around for a while like semaphore or CLR fund. 174 00:10:56,100 --> 00:10:58,600 Jill are funds like allowing you to do get coin? 175 00:11:00,500 --> 00:11:03,100 Type donations but privately as far as I remember. 176 00:11:05,400 --> 00:11:09,200 Yeah and then I'm trying to think like there's two areas 177 00:11:09,900 --> 00:11:12,800 like well actually before I do that Kobe is there any other use 178 00:11:12,800 --> 00:11:15,500 cases that come to mind? I think we've covered our list. 179 00:11:15,500 --> 00:11:19,600 So yeah, I think the thing like maybe there's Worth to mention 180 00:11:19,600 --> 00:11:22,400 that there is a big group in this room Foundation. 181 00:11:22,400 --> 00:11:26,600 That's called like PSE like privacy is going Explorations. 182 00:11:27,000 --> 00:11:30,000 They're doing a lot of projects that push the boundaries in. 183 00:11:30,200 --> 00:11:33,300 MZ K. So semaphore was in there, which 184 00:11:33,300 --> 00:11:36,900 allows for anonymous messaging were signaling more correctly. 185 00:11:38,000 --> 00:11:44,300 Sailor fund is based on Maisie, which is kind of a project that 186 00:11:44,300 --> 00:11:48,300 doesn't fit in the Box. Exactly of privacy or 187 00:11:48,300 --> 00:11:53,300 scalability because it is snarks to prevent collusion between 188 00:11:53,300 --> 00:11:58,500 people. And so it creates a more secure 189 00:11:58,500 --> 00:12:02,200 environment for voting. So that's that's really 190 00:12:02,200 --> 00:12:05,800 compelling. And there's, there's a bunch of 191 00:12:05,800 --> 00:12:07,400 other projects that are being work there. 192 00:12:07,400 --> 00:12:11,700 So, for example, there is a zk0 pro, which combines private 193 00:12:11,700 --> 00:12:17,800 transactions with an optimistic roll-up kind of construction and 194 00:12:17,800 --> 00:12:22,200 generally the doing like cool things. 195 00:12:24,700 --> 00:12:26,200 But I know what was the question? 196 00:12:26,200 --> 00:12:28,600 You know, I need to ask you for any other use cases. 197 00:12:28,600 --> 00:12:30,700 I think what you're saying is like, if you want to find the 198 00:12:30,700 --> 00:12:35,000 good, use cases head to those places I think like, yeah. 199 00:12:35,000 --> 00:12:37,200 I think, keep watch it. Yeah. 200 00:12:37,200 --> 00:12:39,800 Actually, as you're saying this, I was just thinking about, like 201 00:12:39,800 --> 00:12:44,700 so like any project that does EK like any network, they're going 202 00:12:44,700 --> 00:12:48,300 to have like a ZK, education arm and they're going to want, you 203 00:12:48,300 --> 00:12:50,700 know, like they're creating a lot of content and ideas for 204 00:12:50,700 --> 00:12:53,000 their network. But there are like these three 205 00:12:53,200 --> 00:12:54,800 Three groups. I think we count. 206 00:12:54,800 --> 00:12:57,100 ZK hack in that. It's like zika, hack, Xerox, 207 00:12:57,100 --> 00:13:01,700 Parc, and p, and p SE. Because like in that case, I 208 00:13:01,700 --> 00:13:04,000 guess PSE still invariance theorem and actually their ex 209 00:13:04,000 --> 00:13:06,300 Parks variant theorem. But like generally, there's 210 00:13:06,300 --> 00:13:11,100 little bit more of a, like a neutrality for where, and how 211 00:13:11,100 --> 00:13:15,200 you could deploy these things. So they're just like ideas and 212 00:13:15,200 --> 00:13:18,100 libraries like things, you can probably rate recreate in other 213 00:13:18,100 --> 00:13:21,700 places and so yeah, it's definitely like my source for 214 00:13:21,800 --> 00:13:25,400 for use cases. So, there are two big there, two 215 00:13:25,400 --> 00:13:27,900 big topics, we haven't covered. And that's ZK. 216 00:13:27,900 --> 00:13:34,400 Bridges and Z km L, which is like the new very exciting 217 00:13:34,400 --> 00:13:37,200 field. People are jumping into. 218 00:13:37,600 --> 00:13:40,700 Yeah, so maybe we can talk audits. 219 00:13:40,700 --> 00:13:44,200 It's also something that I really want to talk about for a 220 00:13:44,208 --> 00:13:48,000 bit cool. But ZK Bridges sounds good. 221 00:13:48,000 --> 00:13:51,900 So there's there's the 16th Bridge. 222 00:13:52,500 --> 00:13:56,300 It also came out of CR-X Park in a way. 223 00:13:56,300 --> 00:14:01,200 It's a little bit. It's a piece of infrastructure 224 00:14:01,400 --> 00:14:08,700 that most people didn't really expect, right? 225 00:14:09,800 --> 00:14:15,100 Can you talk about how it works? Yeah, happy too. 226 00:14:15,200 --> 00:14:21,600 So usually when you want to get the bridge like if you want the 227 00:14:21,600 --> 00:14:26,300 Holy Grail in some sense or at least something very, very good. 228 00:14:26,600 --> 00:14:29,600 You want something, which is a light client Bridge so that you 229 00:14:29,600 --> 00:14:33,700 can verify the light Clan protocol of another chain inside 230 00:14:33,700 --> 00:14:37,500 your chain and that allows you to get really high guarantees. 231 00:14:37,500 --> 00:14:40,100 Because you don't trust, let's say like a multi, see great job 232 00:14:40,200 --> 00:14:44,900 of small Committee of Sign errors or in an optimistic 233 00:14:44,900 --> 00:14:47,000 Bridge, which is like very secure. 234 00:14:47,000 --> 00:14:52,500 But there is this synchrony assumption or at least it's a 235 00:14:52,500 --> 00:14:56,100 game theoretically secure and not like Zac the matically 236 00:14:56,100 --> 00:14:57,100 secure, right? Yeah. 237 00:14:57,100 --> 00:15:01,200 And like, there is like Network dependency on like latency and 238 00:15:01,200 --> 00:15:03,200 stuff of that sort. So, yeah, exactly. 239 00:15:03,200 --> 00:15:06,400 It's like different kind of security properties. 240 00:15:07,100 --> 00:15:10,700 But alive client Bridge allows you to overcome a lot of these 241 00:15:10,700 --> 00:15:15,000 difficulties specially in. Is the tab and deterministic 242 00:15:15,000 --> 00:15:18,700 finality where you can be certain that after a few blocks, 243 00:15:18,700 --> 00:15:23,400 things will not change and which is true, for example, for 244 00:15:23,400 --> 00:15:27,200 ethereal 2.0, but not through for theorem one. 245 00:15:27,200 --> 00:15:34,500 So, and ZK Bridges, make it possible for chains that have 246 00:15:35,000 --> 00:15:40,900 let's say, incompatible light clients and communicate with 247 00:15:40,900 --> 00:15:44,100 each other. So, When you want to verify the 248 00:15:44,100 --> 00:15:48,300 theorem 2.0 consensus. Then you have to verify a lot of 249 00:15:48,300 --> 00:15:52,200 be less signatures and you have to do it on a very specific 250 00:15:52,200 --> 00:15:56,500 elliptic curve, that that's pretty hard and it's pretty 251 00:15:56,500 --> 00:15:59,100 compute-intensive in the context of a blockchain. 252 00:15:59,900 --> 00:16:03,500 So with the zika bridge, you just found it in the toll 253 00:16:03,500 --> 00:16:07,700 protocol inside the snark and then you just verify the proof, 254 00:16:07,700 --> 00:16:10,700 which is like like we mentioned before, a lot of L ones already 255 00:16:10,700 --> 00:16:15,700 support Primitives for proof. If ocation so you can do that 256 00:16:15,700 --> 00:16:19,500 but you cannot do the light client directly so zika Bridges. 257 00:16:19,500 --> 00:16:22,800 Give you that. So for example, if it's just a 258 00:16:22,800 --> 00:16:27,500 way of kind of getting the stage of the of the other chain, yeah, 259 00:16:27,600 --> 00:16:31,200 crust is Leon to the chain that you're coming from, right? 260 00:16:31,700 --> 00:16:34,500 Yeah, exactly. So you now that you know you can 261 00:16:34,500 --> 00:16:37,300 verify the left hand you know know what is the latest state. 262 00:16:37,400 --> 00:16:41,400 So from there, you can start making queries and check 263 00:16:41,400 --> 00:16:44,400 balances and checks. Check transfer messages or 264 00:16:44,400 --> 00:16:48,500 things like that. So that connects them in a very 265 00:16:48,500 --> 00:16:51,500 nice way. Yeah. 266 00:16:51,500 --> 00:16:55,200 And like if you look I did an episode while back on like 267 00:16:55,500 --> 00:16:57,800 different kinds of bridges and different security for different 268 00:16:57,800 --> 00:17:01,300 kinds of bridges. And these cryptographic Bridges, 269 00:17:01,300 --> 00:17:07,000 tend to do best on the security front and that's where like IBC 270 00:17:07,000 --> 00:17:10,800 is like that. But I funny because I thought, 271 00:17:10,800 --> 00:17:13,400 for a long time, people were just like you can't do that on a 272 00:17:13,400 --> 00:17:15,300 theory. Mm, you can can't. 273 00:17:15,400 --> 00:17:17,599 That was the way. Yeah, everyone's like you just, 274 00:17:17,599 --> 00:17:20,300 it was a paradigm shift, right? I mean, to you, just putting 275 00:17:20,300 --> 00:17:22,300 everything. In the ZK S. 276 00:17:22,300 --> 00:17:26,700 I mean, it's it's kind of it's a trick but it's a trick that 277 00:17:26,700 --> 00:17:27,900 works. Yeah. 278 00:17:27,900 --> 00:17:30,500 And it wasn't practical like it two years ago, right? 279 00:17:30,500 --> 00:17:34,900 Like now things have become much faster and much more advanced 280 00:17:34,900 --> 00:17:37,800 and people have become better at doing this. 281 00:17:38,100 --> 00:17:40,700 And now it's practical. It wasn't practical to years 282 00:17:40,700 --> 00:17:41,700 ago. So it's called. 283 00:17:41,700 --> 00:17:43,600 Yeah. And there are so there's the 284 00:17:43,600 --> 00:17:46,600 succinct Labs. There we actually did for like 4 285 00:17:46,600 --> 00:17:49,600 Z, KV, we do these events sometimes focused on specific 286 00:17:49,600 --> 00:17:52,200 networks so we did one Nah, never mous and actually talked 287 00:17:52,200 --> 00:17:55,100 about ZK Bridges. So there was like, we were able 288 00:17:55,100 --> 00:17:57,400 to find a few other teams doing something similar. 289 00:17:57,400 --> 00:18:00,300 There is like a ZK, IBC. There is a team called seek a 290 00:18:00,300 --> 00:18:03,700 bridge. So yeah, that's maybe that's 291 00:18:03,700 --> 00:18:05,100 also over on the YouTube channel. 292 00:18:05,100 --> 00:18:08,500 So if you do have a linked I can find it, you have New Foundation 293 00:18:08,500 --> 00:18:10,300 that also did this with mean, any Theory. 294 00:18:10,300 --> 00:18:14,800 Mm, yeah that is also has one now as UK bridge and there's a 295 00:18:14,800 --> 00:18:19,900 team coming out of Berkeley as well called I think specular. 296 00:18:20,000 --> 00:18:22,900 So Yeah, that's quite a few teams now. 297 00:18:23,000 --> 00:18:29,300 And in a way this this is something that we had no. 298 00:18:29,300 --> 00:18:32,000 So basically the first deployment of the succinct 299 00:18:32,000 --> 00:18:38,200 Bridge was between the gnosis test net and and and girly the 300 00:18:38,200 --> 00:18:42,800 theorem test that. So in a way we've kind of we've 301 00:18:42,800 --> 00:18:54,300 been tracking this very closely, Yeah, and but the thing is, and 302 00:18:54,300 --> 00:18:58,500 this kind of leads me to my next topic, auditing. 303 00:18:58,500 --> 00:19:02,300 This is so difficult because basically, with a bridge, and I 304 00:19:02,308 --> 00:19:05,200 mean, with the number of Bridge facts, we've seen this year, 305 00:19:05,800 --> 00:19:07,900 let's hope that's. So, we've seen the last of them 306 00:19:07,900 --> 00:19:10,600 for this year, at least, but whatever. 307 00:19:20,500 --> 00:19:24,100 And the number of funds that are typically in bridges and kind of 308 00:19:24,100 --> 00:19:27,300 security is obviously of the utmost concern and I mean, 309 00:19:27,300 --> 00:19:30,400 obviously it sucks with vulnerabilities and more or less 310 00:19:31,400 --> 00:19:37,300 anything. But basically if it's if it's if 311 00:19:37,300 --> 00:19:40,200 it's a game or something obviously it's not you know, the 312 00:19:40,200 --> 00:19:43,600 consequences and that large but yeah. 313 00:19:43,600 --> 00:19:48,100 How do you audit these EK s? How do you make sure that the ZK 314 00:19:48,100 --> 00:19:51,200 circuit because I mean, the byte code itself. 315 00:19:51,300 --> 00:19:55,300 It's not exactly human readable. So yeah, how do you go about 316 00:19:55,300 --> 00:19:57,800 this? So first of all, can I just give 317 00:19:57,800 --> 00:20:05,500 a depressing thought about the security of, like, ZK moments 318 00:20:05,700 --> 00:20:09,000 and bridges in general? This is yes. 319 00:20:09,100 --> 00:20:11,400 But I will counteract your thought afterwards. 320 00:20:11,400 --> 00:20:13,100 But no, go for it. Go for it. 321 00:20:14,400 --> 00:20:17,700 Okay, let's try. So the depressing thought that a 322 00:20:17,700 --> 00:20:21,800 lot of the heck's that We've seen have happened on, let's say 323 00:20:21,800 --> 00:20:25,800 the integration layers. So you have this complex 324 00:20:25,800 --> 00:20:30,600 Solutions, like the circuits and all that, and let's say the 325 00:20:30,600 --> 00:20:35,300 optimistic protocols and that part, but a lot of the hacks 326 00:20:35,300 --> 00:20:39,000 happen on the integration part, let's say on the smart contracts 327 00:20:39,300 --> 00:20:45,100 and so on where you find a bug that happens after you verify 328 00:20:45,100 --> 00:20:50,600 the complex part. And I think that's a lot. 329 00:20:51,500 --> 00:20:54,400 That's the part that where hackers put a lot of effort in 330 00:20:54,800 --> 00:20:59,200 and kind of something that is concerning me, is that when they 331 00:20:59,200 --> 00:21:01,600 start looking at the more complex Parts, maybe maybe 332 00:21:01,600 --> 00:21:04,700 they'll find more holes. So we'll see. 333 00:21:05,700 --> 00:21:07,600 Oh, you're saying like maybe. Yeah. 334 00:21:07,600 --> 00:21:10,000 Okay. So, like the Hat, doctors have 335 00:21:10,000 --> 00:21:13,000 any didn't even need to go. That deep the hacks happened. 336 00:21:13,300 --> 00:21:16,800 Exactly. I mean, I'm totally with you. 337 00:21:16,900 --> 00:21:20,600 I mean Busy the possibility, space for making for introducing 338 00:21:20,600 --> 00:21:25,200 vulnerabilities income tax systems is so large and even if 339 00:21:25,200 --> 00:21:30,900 you look at, if you look like at like the recent Finance bridge 340 00:21:30,900 --> 00:21:34,500 hack and so on, it was actually pretty sophisticated already. 341 00:21:34,500 --> 00:21:39,600 So basically the yeah. So how we actually thought about 342 00:21:39,600 --> 00:21:41,800 this? So obviously we thought about 343 00:21:41,800 --> 00:21:44,400 this for a long time because I mean, I mean we're German 344 00:21:44,400 --> 00:21:46,700 Engineers, right? So we like thinking about 345 00:21:46,700 --> 00:21:49,700 security Education's. Yeah, and audits. 346 00:21:49,700 --> 00:21:53,900 So this is why audits? Yeah, but so far no major hacks. 347 00:21:53,900 --> 00:21:56,900 So, I mean, so far, you know, for a project that's been in 348 00:21:56,900 --> 00:22:01,400 this space for so long. No sir said very few, like 349 00:22:01,400 --> 00:22:04,100 vulnerabilities. And yeah, that's good. 350 00:22:04,900 --> 00:22:06,800 Yeah. I mean, then turn off been any 351 00:22:06,800 --> 00:22:09,100 day. I mean, when you prevent audits 352 00:22:09,100 --> 00:22:11,900 of ZK systems, we do. I mean, just so, you know what, 353 00:22:11,900 --> 00:22:15,700 ZK hack when we introduce these puzzles, they're often based on 354 00:22:15,700 --> 00:22:19,400 historic. Here's so actually like if you 355 00:22:19,400 --> 00:22:22,700 look at the puzzle page on the zika hack site, it's basically 356 00:22:22,700 --> 00:22:25,500 meant to show you. Here's how like a bug can 357 00:22:25,500 --> 00:22:27,300 happen. Here's a vulnerability did 358 00:22:27,300 --> 00:22:31,000 happen, but I think so. I mean the Z cash bug for 359 00:22:31,000 --> 00:22:36,600 example could have been horrific the fact which they don't 360 00:22:36,600 --> 00:22:40,800 believe it has been. I did here, I said because it 361 00:22:40,800 --> 00:22:44,500 would have basically, you would have gotten more in the field, 362 00:22:44,500 --> 00:22:46,200 it pours, right? So basically, you would always 363 00:22:46,200 --> 00:22:49,300 own after its own Their particular system, there was a 364 00:22:49,308 --> 00:22:52,200 way to kind of prove that it didn't happen because of when 365 00:22:52,200 --> 00:22:55,900 they did the upgrade and the number of tokens likes at some 366 00:22:55,900 --> 00:22:58,400 point like you had to move. Yeah. 367 00:22:59,100 --> 00:23:02,900 Yeah not what everyone has but like just a total and so like 368 00:23:02,900 --> 00:23:06,200 had there been this exploit was just like mint Infinity tokens, 369 00:23:06,200 --> 00:23:10,000 you would have seen it but that's like a specific case 370 00:23:10,100 --> 00:23:12,900 where like like there's only like five people in the world 371 00:23:12,900 --> 00:23:15,600 who understood what was happening anyways and like they 372 00:23:15,600 --> 00:23:19,700 all worked there. Like that's I think. 373 00:23:19,700 --> 00:23:21,500 Why? That didn't get exploited. 374 00:23:23,200 --> 00:23:28,100 Yeah, yeah, but yeah, but that could be how we were thinking 375 00:23:28,100 --> 00:23:30,900 about this. This kind of security is kind of 376 00:23:31,000 --> 00:23:33,100 the its kind of strength in numbers. 377 00:23:33,300 --> 00:23:35,700 Basically, it's like you need to add redundancy. 378 00:23:35,700 --> 00:23:38,400 So as you put like a sieve in a sieve and a see if and then 379 00:23:38,400 --> 00:23:41,500 basically you have something that doesn't is not very seve 380 00:23:41,500 --> 00:23:44,400 like anymore. And so basically our idea and 381 00:23:44,400 --> 00:23:48,900 basically how we are going to Because you don't want to test 382 00:23:48,900 --> 00:23:51,500 anything in production, right? So, basically, how we are going 383 00:23:51,500 --> 00:23:58,500 to kind of move this to production without endangering 384 00:24:00,300 --> 00:24:05,300 client funds or user funds is by actually creating a mighty sick 385 00:24:05,300 --> 00:24:07,900 of bridges. So basically you have you have 386 00:24:07,900 --> 00:24:10,900 the old Omni Bridge which is basically multisig base bridge 387 00:24:12,200 --> 00:24:15,700 and you put it put it in the mighty sick with the new ZK like 388 00:24:15,700 --> 00:24:19,400 language and basically if Agree on a payout, payout happens. 389 00:24:19,400 --> 00:24:21,200 If they don't, agree, pale doesn't happen. 390 00:24:21,900 --> 00:24:24,800 And then basically, you have to you have to in order to 391 00:24:24,800 --> 00:24:28,900 compromise that you have to find a vulnerability in the ZK, 392 00:24:28,900 --> 00:24:32,000 lifetime bridge, but at the same time you actually have to 393 00:24:32,000 --> 00:24:37,400 compromise a majority of the sinus for the mighty Sigrid. 394 00:24:37,600 --> 00:24:41,400 So basically that that would make it like at least you know, 395 00:24:42,200 --> 00:24:45,700 an order of magnitude more difficult to kind of break it. 396 00:24:45,800 --> 00:24:48,900 And we're also working with our Others seek a light line, Bridge 397 00:24:48,900 --> 00:24:51,200 teams. So, basically that in the 398 00:24:51,208 --> 00:24:53,800 future, we will have two or three independent 399 00:24:53,800 --> 00:24:56,200 implementations of ZK like time Bridges. 400 00:24:56,400 --> 00:24:58,700 And basically, there will be an Andrew. 401 00:24:58,900 --> 00:25:03,600 So basically, if these a great payout happens, if these don't 402 00:25:03,600 --> 00:25:06,800 agree, nothing happens. And basically in that way you 403 00:25:06,800 --> 00:25:12,500 can reduce your exposure significantly but no, Masti 404 00:25:12,600 --> 00:25:16,200 because not only you don't have to find just one vulnerability 405 00:25:16,200 --> 00:25:18,700 in one Bridge. You actually have to find 406 00:25:18,700 --> 00:25:24,000 vulnerabilities in three different implementations of ZK 407 00:25:24,300 --> 00:25:27,000 client Bridges. Obviously, that could happen. 408 00:25:27,000 --> 00:25:30,700 If there's like, if you have like a ZK, like like one of 409 00:25:30,708 --> 00:25:34,800 those bugs where there's a, you know, bug in a paper for 410 00:25:34,800 --> 00:25:39,200 instance, like what happened with Z cash at the time, but it 411 00:25:39,200 --> 00:25:44,400 becomes a lot less likely that everything kind of has the same 412 00:25:45,000 --> 00:25:47,500 breaking point. So, No II. 413 00:25:47,500 --> 00:25:49,000 Think I think it's very responsible. 414 00:25:49,000 --> 00:25:51,800 I think it's a good approach like to add Security in depth. 415 00:25:51,900 --> 00:25:55,200 I'm a very big fan of security in depth in general. 416 00:25:55,400 --> 00:26:00,300 So, for example, you know, sometimes even people don't like 417 00:26:00,300 --> 00:26:03,600 things like sgx, right? But they seem like sgx. 418 00:26:03,800 --> 00:26:09,100 Yeah, Fox, it's really difficult to trust but if you ask you, 419 00:26:09,100 --> 00:26:11,000 other dozen other. Yeah, exactly. 420 00:26:11,000 --> 00:26:14,100 Like if you had it as another component of security in depth, 421 00:26:15,100 --> 00:26:18,200 It is something that might add security. 422 00:26:18,500 --> 00:26:22,000 So here if you do like at the multisig bridge and you know, 423 00:26:22,200 --> 00:26:24,100 maybe maybe it's easy to bribe, all of them. 424 00:26:24,100 --> 00:26:26,100 You know, there are some rich people that can drive ball of 425 00:26:26,108 --> 00:26:30,500 designers but it adds some measure of security. 426 00:26:30,700 --> 00:26:34,600 So it is nice and I think that's a very responsible way to 427 00:26:34,600 --> 00:26:37,000 approach it. So but yeah like if we're 428 00:26:37,000 --> 00:26:41,400 talking about like auditing the circuits themselves, I think 429 00:26:41,400 --> 00:26:46,900 it's still pretty hard task. I remember like I was doing some 430 00:26:46,900 --> 00:26:50,800 what it's like, when I when I was encouraged, we also like 431 00:26:50,800 --> 00:26:53,800 audited sapling like the zika shop. 432 00:26:53,800 --> 00:26:59,800 Great for it came out and it seems to me that still the most 433 00:27:00,700 --> 00:27:04,200 successful or the most, elaborate, all that's happened. 434 00:27:04,200 --> 00:27:11,700 When people do manual reviews and they see specification in 435 00:27:11,700 --> 00:27:14,700 front of them and they vet the specification first. 436 00:27:14,800 --> 00:27:18,800 First and then they see that the code matches the specification 437 00:27:19,400 --> 00:27:23,300 and so on and I think the Z cash team has been spectacular in 438 00:27:23,300 --> 00:27:31,400 writing good specifications. And that was something that in 439 00:27:31,400 --> 00:27:36,600 my opinion makes them system. Like I'm have a high confidence 440 00:27:36,600 --> 00:27:43,500 in their system because of that and but Today, we still do a lot 441 00:27:43,500 --> 00:27:45,400 of money a lot. It's there are some companies 442 00:27:45,400 --> 00:27:49,100 that try to improve on that. Like there has been work where 443 00:27:49,100 --> 00:27:51,900 the Kestrel people that now are part of failure. 444 00:27:51,900 --> 00:27:55,400 I think that they've done some other things occasionally. 445 00:27:55,400 --> 00:27:58,400 Yeah, exactly. And very dice is also doing 446 00:27:59,100 --> 00:28:04,300 developing a bunch of tooling around that but still not yet 447 00:28:04,300 --> 00:28:08,800 widely used. But one of the things that I 448 00:28:08,800 --> 00:28:14,600 think, contribute More practically to security. 449 00:28:16,400 --> 00:28:22,800 And still don't happen as much as I would like is the creation 450 00:28:22,800 --> 00:28:29,600 of common components that you can reuse, and that get audited 451 00:28:30,000 --> 00:28:34,400 and then get deployed and have a bunch of steak in them, in some 452 00:28:34,400 --> 00:28:38,100 sense, because they're deployed in a high-stakes places. 453 00:28:39,300 --> 00:28:41,800 And that increases the confidence in the components 454 00:28:41,800 --> 00:28:44,300 themselves. And then if you use this as a 455 00:28:44,300 --> 00:28:48,200 major part in your Application can just quickly back like that. 456 00:28:48,200 --> 00:28:51,300 Exactly. And they think we've done some 457 00:28:51,300 --> 00:28:54,000 of this when we were developing semaphore. 458 00:28:54,000 --> 00:28:58,200 So, we audited this component and some other people were using 459 00:28:58,200 --> 00:29:03,900 them. And I would like to see that 460 00:29:03,900 --> 00:29:09,700 more now in the new Realms of the cave, where we use long and 461 00:29:09,800 --> 00:29:13,100 other kinds of systems. It's harder there because you 462 00:29:13,100 --> 00:29:16,400 also have a lot of sex. Build. 463 00:29:16,400 --> 00:29:24,300 You can customize it a lot, but if we find the community plonk 464 00:29:24,500 --> 00:29:27,800 construction, that we can line on, maybe we can start 465 00:29:27,800 --> 00:29:31,800 developing these components that people could reuse. 466 00:29:31,900 --> 00:29:34,200 And that, that makes a big difference. 467 00:29:34,200 --> 00:29:37,900 Yeah, I like that. Yeah, I think for that also 468 00:29:38,400 --> 00:29:47,800 having some code of some sort of, you know, benchmarking to a 469 00:29:47,800 --> 00:29:52,800 which tells you how much certain components have been used in 470 00:29:52,800 --> 00:29:56,200 what amount riding on them as been. 471 00:29:56,500 --> 00:29:59,300 I think that would be super. Super useful. 472 00:29:59,700 --> 00:30:00,900 That would be nice. Yeah. 473 00:30:02,200 --> 00:30:04,300 Yeah there was a last point, right? 474 00:30:04,900 --> 00:30:09,200 Yes. But it's the KML and I don't 475 00:30:09,200 --> 00:30:13,900 know anything really about that. What what is he KML? 476 00:30:14,200 --> 00:30:15,900 Oh, so that's an interesting thought. 477 00:30:16,100 --> 00:30:17,900 I think it's because it has become. 478 00:30:19,000 --> 00:30:22,100 From 0 to 100 popular in some sense. 479 00:30:22,400 --> 00:30:27,200 It like the last month, maybe maybe a month or two. 480 00:30:27,200 --> 00:30:30,500 So yeah. But it's a topic that become 481 00:30:30,500 --> 00:30:34,900 very popular. So Z KML in general is proving 482 00:30:34,900 --> 00:30:39,800 that you evaluated model that you got the results from 483 00:30:39,800 --> 00:30:44,700 validating machine learning model or a neural network and 484 00:30:44,700 --> 00:30:47,500 proving that you did it correctly with the right model 485 00:30:48,300 --> 00:30:51,300 with With signed inputs or so on. 486 00:30:52,200 --> 00:30:58,900 And in that way, you can run this in different environments. 487 00:30:58,900 --> 00:31:03,500 Not necessarily the environment that consumes the result but 488 00:31:03,500 --> 00:31:07,300 enjoy like be confident that you've got the correct result. 489 00:31:07,300 --> 00:31:14,300 So, one example would be for like face liveness detection, 490 00:31:14,300 --> 00:31:22,100 where you would see someone has Like you can do this for kyc or 491 00:31:22,400 --> 00:31:27,900 things of that sort and but you have other use cases. 492 00:31:27,900 --> 00:31:32,000 You can think about, let's say, G PT 3 today or chat GPT where 493 00:31:32,400 --> 00:31:36,600 you want to know that you're getting the results from the 494 00:31:36,600 --> 00:31:42,400 right model that was at least claim to have been trained on 495 00:31:42,400 --> 00:31:48,900 some specific on some set of inputs and you will then be sure 496 00:31:48,900 --> 00:31:51,800 that it you get the result from that model. 497 00:31:51,800 --> 00:31:55,200 And not something that someone else invented that didn't 498 00:31:55,200 --> 00:31:58,600 include all the history of some region in the world. 499 00:31:58,600 --> 00:32:04,200 Let's say so we recently had Jensen on are you familiar with 500 00:32:04,200 --> 00:32:10,000 them Jensen? Jensen, GE n Sy n. 501 00:32:10,000 --> 00:32:15,400 It was episode 471 basically it's about deep learning compute 502 00:32:15,400 --> 00:32:19,700 and kind of distributed basically making that Muted. 503 00:32:19,700 --> 00:32:22,600 And basically there you also have this checkpointing problem. 504 00:32:22,700 --> 00:32:26,600 They basically, you have two people have to prove that they 505 00:32:26,600 --> 00:32:31,700 actually did, run your the data through some sort of model and 506 00:32:31,700 --> 00:32:37,100 basically, how involved at was. And yeah, we also talked a bit 507 00:32:37,100 --> 00:32:40,000 about how difficult that actually is to prove because 508 00:32:40,000 --> 00:32:43,400 it's not deterministic, so it's yeah. 509 00:32:44,000 --> 00:32:46,800 One other angle on this. So like I have not done a deep 510 00:32:46,800 --> 00:32:50,100 dive so I only like, really all I've understood though is 511 00:32:50,100 --> 00:32:51,700 there's like two ways of thinking about it. 512 00:32:51,700 --> 00:32:55,300 One is the way that Kobe just described if like attesting to 513 00:32:55,300 --> 00:32:59,600 the model but I have heard it all so discreet. 514 00:32:59,600 --> 00:33:03,600 Like I've just I've heard the idea of doing kind of like the 515 00:33:03,600 --> 00:33:07,100 model making building the model on encrypted data or like and so 516 00:33:07,100 --> 00:33:09,800 it's not so much ZK. I think people kind of don't 517 00:33:09,800 --> 00:33:13,000 think that's homomorphic encryption exactly it's probably 518 00:33:13,000 --> 00:33:18,700 going to be like fhe or 80 somehow but yeah just like the 519 00:33:18,700 --> 00:33:22,700 topic of ZK. Also, it's often like super 520 00:33:22,700 --> 00:33:26,300 mixed up with all the other Advanced cryptography like MPC 521 00:33:26,300 --> 00:33:28,800 and FHA which is great. I think they shot. 522 00:33:29,700 --> 00:33:31,400 Yeah. Yeah I mean sometimes you're 523 00:33:31,400 --> 00:33:33,700 actually mixing them and sometimes just like in people's 524 00:33:33,700 --> 00:33:37,200 minds they're like very complicated cryptography over 525 00:33:37,200 --> 00:33:42,500 there difficult. That's there does seem to be 526 00:33:42,500 --> 00:33:47,700 some really cool overlap between the ML and I worlds and ZK now, 527 00:33:48,000 --> 00:33:50,600 which is cool. I'm excited for the new year to 528 00:33:50,600 --> 00:33:53,700 like meet some of these people, building this stuff. 529 00:33:54,200 --> 00:33:57,900 Yeah, some really good people working on this on the KML so 530 00:33:57,900 --> 00:33:59,500 I'm excited to see where it goes. 531 00:34:01,300 --> 00:34:06,100 Quit. So it seems like while like for 532 00:34:06,100 --> 00:34:10,699 the last three years or so. Most of the progress has kind of 533 00:34:10,699 --> 00:34:16,300 hinged on the succinctness property of CK proofs. 534 00:34:16,600 --> 00:34:22,800 It seems like it now, Ventures more into like the Privacy 535 00:34:22,800 --> 00:34:26,400 preserving Ram that obviously can also offer, would you agree? 536 00:34:26,400 --> 00:34:30,000 Oh, and I see Anna kind of weak. Like the Hat. 537 00:34:30,400 --> 00:34:32,500 There's been though. Those two, I think the 538 00:34:32,500 --> 00:34:36,400 blockchain the wanting to scale. The blockchain Accelerated. 539 00:34:36,600 --> 00:34:40,100 This is synchronous for sure. And like the idea of using it as 540 00:34:40,500 --> 00:34:44,900 validity instead of privacy. I think those ideas were really 541 00:34:45,300 --> 00:34:49,800 awesome and they were like super easily kind of understood and 542 00:34:49,800 --> 00:34:54,100 then used, but I do think like, as what I've what has happened 543 00:34:54,100 --> 00:34:58,100 is like DK connecting to blockchain accelerated, ZK 544 00:34:58,100 --> 00:35:02,900 development, so much that now the Libraries are quite good and 545 00:35:02,900 --> 00:35:05,200 they're getting better. They start to be able to be used 546 00:35:05,200 --> 00:35:07,300 even outside of blockchain and often. 547 00:35:07,300 --> 00:35:11,800 In those cases, it is privacy. That's the main part so. 548 00:35:11,800 --> 00:35:14,700 Yeah, I think of examples. Well I did I gave this one 549 00:35:14,700 --> 00:35:17,700 example that I don't know that much about like the video thing 550 00:35:17,800 --> 00:35:20,600 Google. Well the video one but like 551 00:35:20,600 --> 00:35:23,200 Google and gdpr. I mean there was years ago 552 00:35:23,200 --> 00:35:26,500 cloudflare used it in a non blockchain context. 553 00:35:26,500 --> 00:35:29,200 I mean we've talked to people who did like password 554 00:35:29,200 --> 00:35:33,900 protection, Barely touching the blockchain, but I think what 555 00:35:33,900 --> 00:35:38,000 I've understood is all the major like tech companies. 556 00:35:38,000 --> 00:35:43,100 Web two companies have teams building ZK stuff outside of the 557 00:35:43,100 --> 00:35:46,400 blockchain context. I think Apple probably is doing 558 00:35:46,400 --> 00:35:50,100 something but obviously, like, I don't know because they don't 559 00:35:50,100 --> 00:35:54,900 tell you stuff but true private very private. 560 00:35:55,700 --> 00:35:58,800 Yeah, I don't know, I think I think we're going to see a lot 561 00:35:58,800 --> 00:36:03,300 of non blockchains EKU. Cases that use privacy in the, 562 00:36:03,700 --> 00:36:07,200 in the near future. I think like, in some sense that 563 00:36:07,700 --> 00:36:13,200 these non blockchain use cases, we see a lot of, let's say, very 564 00:36:13,200 --> 00:36:20,100 specific instructions for very optimized protocol for concrete, 565 00:36:20,100 --> 00:36:24,100 use case, like you said in cloudflare and and what others 566 00:36:24,100 --> 00:36:26,900 do. And maybe sometimes people don't 567 00:36:26,900 --> 00:36:30,700 bundle them with ZK because we're used to snarks, which are 568 00:36:31,100 --> 00:36:36,200 General purpose but definitely a lot of advanced than super 569 00:36:36,200 --> 00:36:42,200 interesting cryptography happen all around and but yeah. 570 00:36:42,300 --> 00:36:46,600 But actually about privacy. I think if we if we look at it 571 00:36:47,000 --> 00:36:51,800 we actually see would see that the first deployments of ZK in 572 00:36:51,800 --> 00:36:53,900 blockchain was about privacy. Right. 573 00:36:53,900 --> 00:36:57,900 Like scalability came later in the let is truth. 574 00:36:58,600 --> 00:37:01,300 Yeah. Oh I just remember to use We 575 00:37:01,300 --> 00:37:04,000 forgot to mention by the way, and one that's hyper relevant. 576 00:37:04,000 --> 00:37:07,100 Today, it is on chain mostly because proof of reserves and 577 00:37:07,100 --> 00:37:10,900 zk0 yeses and it's actually, I would call it a. 578 00:37:11,100 --> 00:37:13,300 It's a, it's both it's attestation. 579 00:37:13,300 --> 00:37:16,500 But often the reason you're using ZK is also for privacy. 580 00:37:17,000 --> 00:37:20,100 So like so that you're not like just overtly showing everybody's 581 00:37:20,100 --> 00:37:23,900 bank account crypto accounts. You actually can retain some 582 00:37:23,900 --> 00:37:25,500 privacy. Yeah, that's kind of a good 583 00:37:25,500 --> 00:37:28,200 hybrid example. It is a good use case. 584 00:37:28,600 --> 00:37:30,900 No. So what are you guys? 585 00:37:31,000 --> 00:37:37,900 He's excited about for 2023. I well I'm going to be KML is 586 00:37:37,900 --> 00:37:41,100 the place I want to explore. It's already been declared on 587 00:37:41,100 --> 00:37:42,700 the last show. We did of the year. 588 00:37:42,700 --> 00:37:47,200 The Troon is like I predict there will be more Z km L 589 00:37:47,200 --> 00:37:50,600 guests. So I think we're going to we're 590 00:37:50,600 --> 00:37:52,800 going to start bringing on some more folks like that. 591 00:37:53,500 --> 00:37:55,300 That's true. I think that's one very 592 00:37:55,300 --> 00:38:00,200 interesting topic. I'm like you want to go more 593 00:38:00,300 --> 00:38:05,200 like yeah, go for it. So I'm turning the excited to 594 00:38:05,200 --> 00:38:12,300 see the development or the let's see which like the tooling 595 00:38:12,300 --> 00:38:17,900 around ZK and just Advanced cryptography mature, like one 596 00:38:17,900 --> 00:38:21,100 analogy that they had. In my mind is that, you know, 597 00:38:21,100 --> 00:38:24,200 when you start out with electronics, you just connect 598 00:38:24,200 --> 00:38:28,900 like an LED with. Battery and then you have like 599 00:38:28,900 --> 00:38:32,100 you turn on, it's very nice. That's kind of where we were 600 00:38:32,100 --> 00:38:34,600 with the tooling like a year ago and ZK. 601 00:38:35,100 --> 00:38:40,400 So now we're at the part of like maybe something that is more 602 00:38:40,400 --> 00:38:44,400 akin to an Arduino, which is still quite crude and then you 603 00:38:44,400 --> 00:38:49,900 put it on on this board where you put the wires, but it's 604 00:38:49,900 --> 00:38:52,800 still quite messy. So you have this component that 605 00:38:52,800 --> 00:38:56,800 you can connect, but I think we're getting to the stage where 606 00:38:57,200 --> 00:39:04,400 It's become something that feels very real and you can design 607 00:39:04,400 --> 00:39:08,000 extremely complex things that interact with each other. 608 00:39:08,000 --> 00:39:11,200 And you have efficient components that you can connect 609 00:39:11,200 --> 00:39:13,600 like transistors. Like, this is kind of an analogy 610 00:39:13,600 --> 00:39:16,400 that I have in my mind. So I'm excited to see this kind 611 00:39:16,400 --> 00:39:22,800 of tolling, develop and people. Mix mixing and matching these 612 00:39:22,800 --> 00:39:28,200 kind of things. But generally also, So, seeing 613 00:39:28,700 --> 00:39:32,200 other kinds of cryptography, intermixing with those. 614 00:39:32,200 --> 00:39:35,900 So we've seen some of that also, from them bonus Group, by the 615 00:39:35,900 --> 00:39:39,000 way, like this collaborative, snarks approach, where you 616 00:39:39,000 --> 00:39:44,300 combine MPC with snarks, like use proven ZK, the correct 617 00:39:44,300 --> 00:39:47,000 execution of a multi-party computation. 618 00:39:49,100 --> 00:39:54,100 So I would love to see more of that happening and I think it 619 00:39:54,100 --> 00:39:56,300 can unlock bunch of interesting use cases. 620 00:39:57,800 --> 00:40:01,600 Yeah, I think it's the use cases like this past year. 621 00:40:01,600 --> 00:40:04,700 Like, I mean, a lot of what we mentioned today are actually 622 00:40:04,700 --> 00:40:08,100 newer to us to, like they've come up in the last year, not 623 00:40:08,100 --> 00:40:11,700 all of them, but a number of those, I remember years ago, 624 00:40:11,700 --> 00:40:15,600 having a really, like, kind of having just such a few number of 625 00:40:15,600 --> 00:40:19,000 use cases, like trying to pick like, oh, yeah, you can do 626 00:40:19,000 --> 00:40:23,600 payments kind of you can scale but yeah, this time around it's 627 00:40:23,600 --> 00:40:26,600 like there's a real, There's real list and I think it's going 628 00:40:26,600 --> 00:40:29,900 to keep growing. Yeah, I think that I think. 629 00:40:30,100 --> 00:40:33,900 Yeah, exactly. And I think more more let's say 630 00:40:34,600 --> 00:40:37,800 Engineers that do not specialize in your Toga fee would be able 631 00:40:37,800 --> 00:40:43,100 to access these Technologies, which is not quite true today. 632 00:40:43,300 --> 00:40:49,300 So I hope that will happen and that will spark more ideas and 633 00:40:49,300 --> 00:40:55,700 more use cases 2023. What about fears for 20:23? 634 00:40:55,800 --> 00:40:58,400 What is something that we absolutely must get, right? 635 00:41:00,100 --> 00:41:05,000 Cool. Funny my fear for 2021 had come 636 00:41:05,000 --> 00:41:09,900 true actually which was like a large-scale ZK project gets tied 637 00:41:09,900 --> 00:41:15,800 to International crime and it did I think I mean we didn't 638 00:41:15,800 --> 00:41:18,400 talk much about tornado but I think it was a reckoning for the 639 00:41:18,400 --> 00:41:23,000 ZK space and it's pushed teams towards compliance and some 640 00:41:23,000 --> 00:41:28,800 away. I think fears are that. 641 00:41:29,100 --> 00:41:32,100 That people use EK for nefarious purposes that are worse than 642 00:41:32,200 --> 00:41:33,700 Than tornado. I don't know. 643 00:41:34,300 --> 00:41:39,000 That's what I would not want. Yeah, I actually would, let's 644 00:41:39,000 --> 00:41:43,700 say that what we must get right? Is how to how to get people to 645 00:41:43,700 --> 00:41:48,500 use privacy in a way that doesn't hurt them because I 646 00:41:48,500 --> 00:41:54,200 think a lot of people today use privacy and like like the used 647 00:41:54,200 --> 00:41:56,300 tools in general and maybe they don't think about the 648 00:41:56,300 --> 00:42:02,100 implications. So yeah by technological ways in 649 00:42:02,300 --> 00:42:04,400 It's kind of this could be addressed, right? 650 00:42:04,400 --> 00:42:08,100 So for instance, for tornado. There was the idea that you can 651 00:42:09,100 --> 00:42:13,500 have curated lists of addresses. You don't want to be mixed with 652 00:42:13,500 --> 00:42:16,300 and then you can prove that you weren't mixed with these offices 653 00:42:16,300 --> 00:42:18,200 and so on. Then we can just be like an 654 00:42:18,200 --> 00:42:21,800 adblocker where basically you subscribe to some service that 655 00:42:21,800 --> 00:42:24,500 kind of tells you who who not to play with. 656 00:42:24,900 --> 00:42:30,300 And so I think there's actually technological solutions for lots 657 00:42:30,300 --> 00:42:32,100 of things. But what what? 658 00:42:32,200 --> 00:42:34,800 I'm worried about sorry I'm hijacking my own question. 659 00:42:36,200 --> 00:42:43,400 What I'm worried about is that privacy as a concept somehow 660 00:42:43,400 --> 00:42:47,300 gets discredited that it kind of goes down the drain because 661 00:42:48,600 --> 00:42:56,800 people tired to nefarious purposes or basically where it's 662 00:42:56,800 --> 00:43:01,100 been used in the various ways where I would kind of pause it. 663 00:43:02,400 --> 00:43:04,600 And it's a value, right? Basically, it's kind of like 664 00:43:04,600 --> 00:43:09,800 something that that kind of you have you subscribe to our you 665 00:43:09,800 --> 00:43:16,000 don't. But I do believe that privacy 666 00:43:16,000 --> 00:43:21,100 should be a human, right? And I think kind of the erosions 667 00:43:21,100 --> 00:43:24,500 of that. How do you feel about that? 668 00:43:24,500 --> 00:43:27,400 Well, I just I was about to say I think actually tornado in a 669 00:43:27,408 --> 00:43:31,300 way and like a lot of things this year, if anything, right? 670 00:43:31,300 --> 00:43:34,400 This at this moment I actually think there's a renewed interest 671 00:43:34,400 --> 00:43:37,200 in having privacy. I actually think the last few 672 00:43:37,200 --> 00:43:39,400 years there's been more of like a less a fair. 673 00:43:39,700 --> 00:43:44,500 Oh well I actually I think I feel more of like a realization 674 00:43:44,500 --> 00:43:47,100 that it's not just like a fun nice to have but more of a 675 00:43:47,100 --> 00:43:52,400 pillar, if you really want the world, you want, And I think, I 676 00:43:52,400 --> 00:43:57,500 don't know. I felt that at least the tone of 677 00:43:57,600 --> 00:43:59,500 discourse has actually shifted in that way. 678 00:44:00,300 --> 00:44:02,600 So I'm hopeful, I'm hopeful that in the new year. 679 00:44:02,600 --> 00:44:05,700 It actually continues to be relevant, if not more relevant. 680 00:44:05,800 --> 00:44:08,000 But we may go through another phase, where we don't care. 681 00:44:08,100 --> 00:44:10,000 That would, that's totally possible. 682 00:44:10,900 --> 00:44:14,400 I mean, I think that is also, like levels of privacy that have 683 00:44:14,400 --> 00:44:20,400 not been properly explored yet like I want privacy from Um, 684 00:44:20,400 --> 00:44:23,200 let's say, I don't want this random person to see what I'm 685 00:44:23,200 --> 00:44:27,500 doing, for example, in my bank account today, but I'm okay 686 00:44:27,500 --> 00:44:31,500 with, I don't know the bank seeing this or some other 687 00:44:31,500 --> 00:44:35,600 entities saying it. So this is the level of privacy 688 00:44:35,600 --> 00:44:39,800 that like like it's not all or nothing like that doesn't 689 00:44:40,800 --> 00:44:42,400 control. Yeah. 690 00:44:42,400 --> 00:44:45,900 Permissions. Or who you reveal it to or Jack 691 00:44:45,900 --> 00:44:48,000 you want privacy from the public? 692 00:44:48,100 --> 00:44:50,800 Definitely. I think this is definitely That 693 00:44:51,100 --> 00:44:54,600 I would like to say but there are levels to this. 694 00:44:54,900 --> 00:44:58,100 I also I do think that the builders, you know, they've made 695 00:44:58,100 --> 00:45:01,800 decisions but I think another decision that's been made is to 696 00:45:01,800 --> 00:45:04,100 use exactly what you described Federica. 697 00:45:04,100 --> 00:45:09,600 These like new architectures or like techniques to actually 698 00:45:09,600 --> 00:45:12,900 avoid the bad behavior but still have these systems for good 699 00:45:12,900 --> 00:45:15,900 actors. I think they're think. 700 00:45:15,900 --> 00:45:18,300 I think a lot of teams are thinking more about that there. 701 00:45:18,300 --> 00:45:21,100 Had been this attitude of, like, oh, well, we decentralized and 702 00:45:21,100 --> 00:45:24,500 we're not in trouble, or we add a viewing key, and then we're 703 00:45:24,500 --> 00:45:28,900 not in trouble, but actually, like the effectiveness of these 704 00:45:29,000 --> 00:45:32,200 are questionable. I think like there are some new 705 00:45:32,200 --> 00:45:35,200 techniques that haven't been explored haven't been used and 706 00:45:35,200 --> 00:45:37,200 maybe they'll be more. And I think people thinking 707 00:45:37,200 --> 00:45:40,700 about those techniques is really important because I think we 708 00:45:40,700 --> 00:45:43,400 want, like you said Kobe. I think we want a place where we 709 00:45:43,400 --> 00:45:46,500 can act pretty I've ately, but we also don't want just Bad 710 00:45:46,500 --> 00:45:50,900 actors to come in and read like use those those tools in ways 711 00:45:50,900 --> 00:45:53,900 that just like hurt Society. Yeah. 712 00:45:54,600 --> 00:45:57,700 I think this, these are really good last words, Anna. 713 00:45:57,700 --> 00:46:02,700 Yeah, you happy with these as last words, for the year, your 714 00:46:02,700 --> 00:46:05,000 last night. Well, my last words for the 715 00:46:05,008 --> 00:46:11,100 year. Let's say good, thank you both 716 00:46:11,100 --> 00:46:13,700 for coming on. Yeah, I'll add a lot. 717 00:46:14,500 --> 00:46:24,500 I am so confused. So my I think I think I would 718 00:46:24,500 --> 00:46:28,000 love some benchmarking, you know, just like some tables that 719 00:46:28,000 --> 00:46:31,700 tell you, this is good at that. This is performance for this, 720 00:46:31,700 --> 00:46:36,100 this solution for this, kind of set of problems to me. 721 00:46:36,100 --> 00:46:43,100 That would be shining a light where it needs to be shown. 722 00:46:43,200 --> 00:46:46,900 So it's a Yeah, but I think the things have to come live. 723 00:46:46,900 --> 00:46:51,100 They have to happen before and then we have to see and then 724 00:46:51,300 --> 00:46:56,300 like I think I hope it was like, it might be a best best team 725 00:46:56,300 --> 00:47:00,600 wins or there might be this really interesting multi L2 726 00:47:00,600 --> 00:47:04,700 scenario with with awesome. ZK Bridges to L ones and they're 727 00:47:04,700 --> 00:47:08,200 all like they have their Niche and their use cases that like 728 00:47:08,200 --> 00:47:10,200 fit for exactly their architecture. 729 00:47:10,700 --> 00:47:12,800 Yeah, it's going to be exciting to see how it comes, how it 730 00:47:12,800 --> 00:47:16,400 goes. Thank you both for coming on. 731 00:47:16,600 --> 00:47:20,700 Thank you for giving us so much of your time and I wish you a 732 00:47:20,800 --> 00:47:25,700 super happy holiday season and get into the new year very 733 00:47:26,100 --> 00:47:29,900 safely. Yeah, you too, always a fun. 734 00:47:30,500 --> 00:47:32,100 Thanks fredericka. Thank you. 735 00:47:34,300 --> 00:47:36,200 Thank you for joining us on this week's episode. 736 00:47:36,500 --> 00:47:38,100 We release new episodes every week. 737 00:47:38,700 --> 00:47:41,400 You can find And subscribe to the show on iTunes Spotify, 738 00:47:41,500 --> 00:47:44,600 YouTube, SoundCloud or wherever. The podcast. 739 00:47:44,900 --> 00:47:47,700 And if you have a Google home or Alexa device, you can tell it to 740 00:47:47,707 --> 00:47:50,700 listen to the latest episode of the epicenter podcast, go to 741 00:47:50,700 --> 00:47:53,800 epicenter, .t V /, subscribe for a full list of places where you 742 00:47:53,800 --> 00:47:56,400 can watch and listen, while you're there, be sure to sign up 743 00:47:56,400 --> 00:47:59,200 for the newsletter so you get new episodes in your inbox as 744 00:47:59,200 --> 00:48:01,700 they're released. If you want to interact with us 745 00:48:02,000 --> 00:48:04,300 guests or other podcast listeners, you can follow us on 746 00:48:04,300 --> 00:48:07,300 Twitter and please leave us a review on iTunes helps people 747 00:48:07,300 --> 00:48:10,400 find the show and we're always happy to read them but thanks so 748 00:48:10,400 --> 00:48:12,700 much and we look forward to being back next week.