1 00:00:00,000 --> 00:00:04,340 Mark, thank you so much for coming on the 21 and 21 show today. 2 00:00:04,660 --> 00:00:05,760 Yeah, excited to be here. Thank you. 3 00:00:06,060 --> 00:00:21,800 Yes, of course. For those of you that don't know Mark Suman, he's the co-founder and CEO of Maple AI, a privacy first AI chat built on OpenSecret, which is a developer platform his team also created to enable encryption and end to end security for AI applications. 4 00:00:21,800 --> 00:00:27,720 Before Maple AI, Mark worked at Mutiny, which is a web-based self-custodial Bitcoin wallet. 5 00:00:28,060 --> 00:00:35,140 And prior to that, he spent several years at Apple as an engineer focused on AI, machine learning, and privacy-preserving systems. 6 00:00:35,300 --> 00:00:36,000 Yep, that's correct. 7 00:00:36,360 --> 00:00:37,040 You've done a lot. 8 00:00:37,820 --> 00:00:41,920 Yeah, been working in the tech industry for, gosh, like 20 years now or so. 9 00:00:42,840 --> 00:00:48,020 I also did a few startups before going to Apple, largely focused on education technology. 10 00:00:48,920 --> 00:00:51,220 A lot of people have used our stuff in university. 11 00:00:51,220 --> 00:00:56,040 we built Instructure Canvas, which is learning management platform, and then also worked on some 12 00:00:56,040 --> 00:01:01,940 apps that helped kids in speech therapy who had speech stutters or impediments. And then the way 13 00:01:01,940 --> 00:01:05,980 that I really got into privacy, though, was my first job right out of school was an online backup 14 00:01:05,980 --> 00:01:10,960 software where you would back up your entire computer to our cloud. And you would like all 15 00:01:10,960 --> 00:01:16,260 your photos, all your documents, everything. And I was helping someone. I was on the developer side, 16 00:01:16,260 --> 00:01:21,320 a customer service escalated somebody to me to say, hey, can you help them restore their files? 17 00:01:21,400 --> 00:01:25,400 They had a catastrophic event with their computer. And I got in there and I could see all of their 18 00:01:25,400 --> 00:01:29,660 photos. And I was like, I feel very violating right now that I'm walking around inside these 19 00:01:29,660 --> 00:01:36,060 people's family photos. And that's when I realized this is what all of our cloud infrastructure is. 20 00:01:36,440 --> 00:01:40,380 It's like just this database where employees can see what people are uploading to it. 21 00:01:40,900 --> 00:01:46,100 But something that we offer to our customers was we let them have a private encryption key if they 22 00:01:46,100 --> 00:01:51,300 wanted to. But only it was so hard to use that only about like five to 10% of our users actually 23 00:01:51,300 --> 00:01:55,880 use that and encrypted all the data away from us employees. So that was my big eye opening moment 24 00:01:55,880 --> 00:02:02,320 that I've kind of kept with me my whole career so far. Wow. Yeah. I mean, it's definitely like a 25 00:02:02,320 --> 00:02:09,160 rabbit hole privacy that is just I mean, obviously, we're in the Bitcoin space. That's something I 26 00:02:09,160 --> 00:02:16,580 know and care a lot about but ai becoming just so much everywhere it's the average person is so much 27 00:02:16,580 --> 00:02:23,320 more affected yeah definitely or more aware but anyways i have a bunch of questions for you um 28 00:02:23,320 --> 00:02:26,480 and you are kind of like i'm excited to talk to you because you're kind of like the perfect 29 00:02:26,480 --> 00:02:32,000 candidate of like a of the show in persidia bitcoin where you have bitcoin history you have 30 00:02:32,000 --> 00:02:38,300 big tech and also ai because we're really interested in the intersection of bitcoin and ai 31 00:02:38,300 --> 00:02:38,700 Mm-hmm. 32 00:02:39,380 --> 00:02:40,280 So I'm curious. 33 00:02:40,520 --> 00:02:45,060 Well, first off, can you maybe just like tell our listeners a little bit more about Maple 34 00:02:45,060 --> 00:02:46,760 AI and OpenSecret? 35 00:02:47,220 --> 00:02:47,660 Yeah. 36 00:02:47,700 --> 00:02:48,760 How they're connected? 37 00:02:49,360 --> 00:02:49,980 Sure, definitely. 38 00:02:50,440 --> 00:02:55,700 When we pivoted away from Mutiny Wallet, because it was a difficult space to build in, we took 39 00:02:55,700 --> 00:03:00,920 the privacy ethos of Mutiny, which was every user has their own private encryption key. 40 00:03:01,420 --> 00:03:06,240 And so when their data gets synced to our Mutiny cloud, if you will, it was like a private 41 00:03:06,240 --> 00:03:07,260 data vault per user. 42 00:03:07,260 --> 00:03:10,960 but we put the burden on the user to manage their private key. 43 00:03:11,180 --> 00:03:13,400 And it was very difficult to do. 44 00:03:13,700 --> 00:03:15,260 We had big scary warnings on the screen, 45 00:03:15,340 --> 00:03:16,720 don't lose this key or you'll lose your money. 46 00:03:17,340 --> 00:03:19,980 And so we sought out to build a better platform. 47 00:03:19,980 --> 00:03:21,880 So we want to build an app developer platform 48 00:03:21,880 --> 00:03:23,520 that managed private keys for the users. 49 00:03:23,700 --> 00:03:25,380 And we discovered secure enclaves, 50 00:03:25,720 --> 00:03:28,320 which are basically a way to do encryption, 51 00:03:28,440 --> 00:03:29,880 hardware encryption in the cloud per user. 52 00:03:30,240 --> 00:03:31,900 So that's really what OpenSecret is. 53 00:03:32,200 --> 00:03:33,920 Anybody who builds an app on our platform, 54 00:03:33,920 --> 00:03:37,380 each user gets their own private data vault 55 00:03:37,380 --> 00:03:39,940 and the app developer cannot see what that data is. 56 00:03:40,860 --> 00:03:43,840 So part of that is we decided to build a proof of concept 57 00:03:43,840 --> 00:03:46,460 that turned out to be an app that lots of people wanted. 58 00:03:46,660 --> 00:03:48,080 So that's what Maple AI is. 59 00:03:48,520 --> 00:03:50,320 And it's effectively ChatGPT 60 00:03:50,320 --> 00:03:53,060 but every user has end-to-end encryption with their chats. 61 00:03:53,160 --> 00:03:54,600 So we can't see what you're chatting about. 62 00:03:55,040 --> 00:03:56,800 No AI company can see what you're chatting about. 63 00:03:56,920 --> 00:04:01,240 Nothing goes back to ChatGPT, to DeepSeek, to any of them. 64 00:04:01,240 --> 00:04:04,920 It's really just you and an AI privately in a room chatting with each other. 65 00:04:06,060 --> 00:04:06,460 So interesting. 66 00:04:06,640 --> 00:04:08,940 I was just playing around with Maple AI a bit before this. 67 00:04:09,140 --> 00:04:09,220 Yes. 68 00:04:10,600 --> 00:04:20,500 So I guess like your biggest focus right now, is it on Maple AI or is it on OpenSecret or is it focused on both? 69 00:04:20,720 --> 00:04:21,960 Yeah, it's definitely on Maple right now. 70 00:04:22,420 --> 00:04:24,740 We found a lot more interest in Maple. 71 00:04:25,080 --> 00:04:30,080 There is definitely interest in OpenSecret, but to build out a developer ecosystem really takes a lot of resources. 72 00:04:30,080 --> 00:04:32,860 You need a lot of documentation, a lot of developer support. 73 00:04:33,060 --> 00:04:35,060 You need developer evangelists to go out there. 74 00:04:35,740 --> 00:04:38,540 And so we decided to kind of let that simmer. 75 00:04:39,200 --> 00:04:46,140 And then Maple is actually helping us build out OpenSecret because as it grows and needs to scale, then we find things that we have to add into the OpenSecret platform. 76 00:04:46,360 --> 00:04:48,340 So we're basically dogfooding our own technology. 77 00:04:48,900 --> 00:04:54,940 And at some point in the future, as Maple really hits scale, then we can decide to open up OpenSecret to more developers. 78 00:04:54,940 --> 00:04:59,540 We have one developer who's actually here in Procedure Bitcoin that is building on OpenSecret. 79 00:04:59,540 --> 00:05:02,880 and we might selectively open it up in the future. 80 00:05:02,980 --> 00:05:04,540 But for now, we're just really cooking on Maple. 81 00:05:05,140 --> 00:05:05,420 Okay. 82 00:05:06,200 --> 00:05:08,600 And how did you come up with this name, Maple? 83 00:05:08,820 --> 00:05:09,100 Yeah. 84 00:05:09,800 --> 00:05:11,420 You know, it's kind of interesting. 85 00:05:11,680 --> 00:05:12,660 So first and foremost, 86 00:05:12,660 --> 00:05:15,380 we wanted a word that was just really like easy to say 87 00:05:15,380 --> 00:05:18,380 and people had a positive connotation with it, right? 88 00:05:18,960 --> 00:05:20,880 So, because we thought about names 89 00:05:20,880 --> 00:05:23,280 like secret GPT or private chat. 90 00:05:23,280 --> 00:05:25,960 And while those are very descriptive of what Maple is, 91 00:05:26,300 --> 00:05:29,080 it's kind of off-putting to the mainstream general audience. 92 00:05:29,540 --> 00:05:43,305 And then when we decided to pivot off of Mutiny it was a pretty difficult time to be honest for all of us And so we took a week and we went out to a beach house And we said let just have a hack week and let see if we can build something together 93 00:05:43,845 --> 00:05:45,225 And that's really where Maple came from. 94 00:05:45,225 --> 00:05:51,465 And while we were there at this beach house, the part of the country we were in had a bunch of forests that went right up against the water. 95 00:05:52,245 --> 00:05:58,205 And I started kind of, we were driving around, I was looking at these trees, just admiring these forests that I love out there. 96 00:05:58,205 --> 00:06:27,545 And I started thinking about species of trees and how a few of them send their roots down into the ground and communicate with each other privately. And maple is one of those species. And so what they do as a forest, if there is some kind of threat in any part of the forest, they are able to send messages to each other and share resources to help weather this harsh environment. And so we thought those are great metaphor for us as humans trying to use AI as a companion to help us kind of make it through this life that we live in. 97 00:06:27,545 --> 00:06:32,205 that's so interesting also completely random but my mom has recently gone down the rabbit 98 00:06:32,205 --> 00:06:37,505 hole of trees really okay yes she's super into it and books and she was telling me all about it so 99 00:06:37,505 --> 00:06:41,245 i'm really excited to tell her so she can tell you all about fungal networks and all that stuff 100 00:06:41,245 --> 00:06:47,105 yeah she's very interested so i'm excited okay something i learned today from able ai is about 101 00:06:47,105 --> 00:06:55,225 maple trees yeah um well i love it uh so maple ai definitely fundamentally has a different approach 102 00:06:55,225 --> 00:06:58,925 than like, you know, big AI companies like ChatGPT. 103 00:06:59,285 --> 00:07:01,645 Can you maybe tell me a little bit more about that 104 00:07:01,645 --> 00:07:04,785 and maybe why someone who's less educated 105 00:07:04,785 --> 00:07:08,465 about privacy around AI should care? 106 00:07:08,905 --> 00:07:11,685 Yeah, the existing AI technology that we use now, 107 00:07:11,765 --> 00:07:14,105 the bigger ones, ChatGPT, Grok, you name it, 108 00:07:14,425 --> 00:07:16,225 they are all built on the same paradigm 109 00:07:16,225 --> 00:07:18,305 of the cloud world that we've lived in 110 00:07:18,305 --> 00:07:19,245 for the last 20 years, 111 00:07:19,525 --> 00:07:21,785 especially social media and other SaaS, 112 00:07:22,025 --> 00:07:24,005 you know, and then the whole SaaS product market, 113 00:07:24,005 --> 00:07:26,845 which is users give their data over 114 00:07:26,845 --> 00:07:28,985 and then the companies monetize that user data 115 00:07:28,985 --> 00:07:30,025 for their own benefit. 116 00:07:30,525 --> 00:07:32,565 And then either you get the AI for free 117 00:07:32,565 --> 00:07:34,145 or you get it for a subsidized rate. 118 00:07:34,505 --> 00:07:36,525 So most of these shops are not making 119 00:07:36,525 --> 00:07:37,685 very much money per user. 120 00:07:38,245 --> 00:07:41,345 You know, in fact, when they try to start to raise prices, 121 00:07:41,465 --> 00:07:42,925 users get mad because they don't realize 122 00:07:42,925 --> 00:07:44,325 that they're actually paying a lot less 123 00:07:44,325 --> 00:07:44,905 than they should be. 124 00:07:45,605 --> 00:07:48,305 So we've kind of flipped that on its head 125 00:07:48,305 --> 00:07:50,885 because we saw so many problems with social media 126 00:07:50,885 --> 00:07:52,285 where we gave over our data 127 00:07:52,285 --> 00:07:53,465 and we gave over our privacy 128 00:07:53,465 --> 00:07:55,545 and we gave over a lot of our rights and our freedoms. 129 00:07:56,425 --> 00:07:58,905 And so we don't want to repeat that mistake with AI 130 00:07:58,905 --> 00:08:01,245 because frankly, I think that there should be, 131 00:08:01,625 --> 00:08:03,745 I'm hoping, I'm really hopeful for all of us 132 00:08:03,745 --> 00:08:05,565 that there's a line in the sand that people draw 133 00:08:05,565 --> 00:08:07,865 that say, we gave you our social media 134 00:08:07,865 --> 00:08:10,225 because we were curating the information we gave out. 135 00:08:10,605 --> 00:08:11,585 But as we use AI, 136 00:08:11,765 --> 00:08:13,385 really we're dumping our whole brain into this. 137 00:08:14,005 --> 00:08:16,125 And I want people to draw a line there and say, 138 00:08:16,225 --> 00:08:17,465 we're not going to give up our privacy 139 00:08:17,465 --> 00:08:18,805 of our thoughts and of our brain. 140 00:08:19,305 --> 00:08:20,505 And that's where Maple comes in. 141 00:08:20,505 --> 00:08:24,465 And Maple lets people have their data encrypted end to end. 142 00:08:24,925 --> 00:08:32,045 And really, it's a place where you can think freely and not have somebody that's watching you approving or disapproving of what you ask the AI. 143 00:08:32,425 --> 00:08:39,485 Because we are people who just think through a lot of ideas and we're constantly throwing away ideas, right? 144 00:08:39,985 --> 00:08:44,565 But there's nobody in our head that's saying, hey, I'm going to flag you for a thought you had. 145 00:08:44,565 --> 00:08:49,985 And I'm going to restrict your access to thought tomorrow because of what you thought about today. 146 00:08:50,505 --> 00:08:52,665 that should not exist within the AI environment. 147 00:08:52,805 --> 00:08:54,425 We need to be free to kind of ideate 148 00:08:54,425 --> 00:08:57,425 and settle on the right answer eventually. 149 00:08:58,465 --> 00:09:02,045 So right now, basically like how MapleAi, 150 00:09:02,145 --> 00:09:03,625 because right now you can either like use 151 00:09:03,625 --> 00:09:05,625 ChattrPT or something similar in the cloud 152 00:09:05,625 --> 00:09:06,865 that's, you know, public, 153 00:09:07,505 --> 00:09:11,125 or you can like download locally on your machine AI, 154 00:09:11,285 --> 00:09:13,705 but that's kind of like hard and difficult to do. 155 00:09:13,745 --> 00:09:17,105 Is MapleAi kind of like an in-between solution 156 00:09:17,105 --> 00:09:18,445 or how does that work? 157 00:09:18,545 --> 00:09:19,605 Yeah, I mean, yeah, 158 00:09:19,605 --> 00:09:24,705 you kind of just drew a little spectrum there of like chat GPT is the least private and then local 159 00:09:24,705 --> 00:09:31,565 AI is awesome. It's but it's difficult to use and you sacrifice these models are smaller. So they're 160 00:09:31,565 --> 00:09:36,725 less powerful. They're a little bit slower, that kind of stuff. Maple tries to be as close to the 161 00:09:36,725 --> 00:09:41,245 local AI privacy as possible. But we are trying to give people the exact same experience to get 162 00:09:41,245 --> 00:09:46,845 with chat GPT. And the way that we do that is through this secure enclave architecture. And I 163 00:09:46,845 --> 00:09:50,325 I mean, that's like a whole other episode to dive into that technology, but it's really awesome. 164 00:09:50,745 --> 00:09:53,885 And it lets people have a private end-to-end encrypted experience. 165 00:09:54,805 --> 00:09:58,945 And the biggest part of all of this is that we make our AI verifiable. 166 00:09:59,445 --> 00:10:02,445 And what verifiable means is the code is open source. 167 00:10:03,065 --> 00:10:08,925 The confidential computing stuff has mathematical proofs that our open source code matches what's running on the servers. 168 00:10:09,005 --> 00:10:13,005 So you don't have to trust us when we say we're encrypting your data and we can't see it. 169 00:10:13,005 --> 00:10:17,845 you can actually go look at the open source code, inspect it, and know that how your data is being handled. 170 00:10:18,545 --> 00:10:23,965 And we think that is essential to the promise that we give our users, is that they can verify it themselves. 171 00:10:25,005 --> 00:10:25,285 Interesting. 172 00:10:27,025 --> 00:10:28,285 Okay, one question I have. 173 00:10:28,485 --> 00:10:33,045 So, I don't know a ton about how, you know, these AI models work, 174 00:10:33,125 --> 00:10:38,665 but roughly, like, the models kind of, like, learn via information, and that's how the data and training. 175 00:10:39,345 --> 00:10:42,665 So, how does private AIs work? 176 00:10:42,665 --> 00:10:48,845 Do they kind of train in similar ways of something like Rock ChatGPT or is the model learning a bit different? 177 00:10:50,425 --> 00:10:55,965 So right now we use open source models that have been trained off of big, large data sets. 178 00:10:56,205 --> 00:10:58,025 And then we don't modify them at all. 179 00:10:58,205 --> 00:10:59,465 So we give you access. 180 00:10:59,885 --> 00:11:03,045 For example, ChatGPT open sourced a model a few months ago. 181 00:11:03,145 --> 00:11:04,345 It's called GPT OSS. 182 00:11:04,605 --> 00:11:18,390 So we have that running in one of our enclaves and you chat with it and we just give you like raw access to it Basically we give it to you as if you had this laptop that was capable of running it at full speed but you don And so we encrypt you a channel 183 00:11:18,390 --> 00:11:20,250 into our data center to let you run it there. 184 00:11:20,870 --> 00:11:22,350 It's not learning from you yet. 185 00:11:22,450 --> 00:11:23,790 That's something we want to build into Maple. 186 00:11:23,910 --> 00:11:25,190 We would love to build this concept 187 00:11:25,190 --> 00:11:26,910 where Maple is understanding 188 00:11:26,910 --> 00:11:28,750 how you chat and how you think, 189 00:11:29,270 --> 00:11:31,950 but do it in a way that's positive for you. 190 00:11:31,950 --> 00:11:34,310 So we would build it open source, of course, 191 00:11:34,710 --> 00:11:36,390 and then you'd be able to inspect the data 192 00:11:36,390 --> 00:11:37,250 that it knows about you. 193 00:11:37,590 --> 00:11:38,870 Whereas right now with Chat2PT, 194 00:11:39,290 --> 00:11:41,070 they build this whole biography. 195 00:11:41,210 --> 00:11:42,690 If you use Chat2PT for like a year, 196 00:11:43,170 --> 00:11:44,490 they're going to write this entire biography 197 00:11:44,490 --> 00:11:45,530 about who Haley is, 198 00:11:45,970 --> 00:11:46,710 what she thinks about, 199 00:11:46,770 --> 00:11:48,310 and also how gullible is she 200 00:11:48,310 --> 00:11:49,150 and where is she gullible. 201 00:11:49,570 --> 00:11:50,550 And then if they ever wanted 202 00:11:50,550 --> 00:11:51,390 to convince you of something, 203 00:11:51,470 --> 00:11:52,890 they know like how to talk to you 204 00:11:52,890 --> 00:11:53,590 and convince you of it 205 00:11:53,590 --> 00:11:54,510 without you realizing it. 206 00:11:55,010 --> 00:11:57,390 Whereas we would want to let you see 207 00:11:57,390 --> 00:11:58,890 that whole biography we have on you. 208 00:11:59,350 --> 00:12:00,970 And as you're chatting with private AI, 209 00:12:01,050 --> 00:12:02,210 you know what's going on, 210 00:12:02,210 --> 00:12:03,610 what the relationship is. 211 00:12:03,610 --> 00:12:07,190 so I guess it's kind of a roundabout way of saying 212 00:12:07,190 --> 00:12:08,830 not yet, it's not learning about you 213 00:12:08,830 --> 00:12:10,390 but that is something we really want to build 214 00:12:10,390 --> 00:12:12,270 in the upcoming 2026 year 215 00:12:12,270 --> 00:12:13,290 Interesting 216 00:12:13,290 --> 00:12:16,930 Well, I guess do you think that 217 00:12:16,930 --> 00:12:18,890 people have to make any sacrifice 218 00:12:18,890 --> 00:12:20,350 on like the information quality 219 00:12:20,350 --> 00:12:22,950 or amount of information they can get 220 00:12:22,950 --> 00:12:24,510 from using a private AI 221 00:12:24,510 --> 00:12:26,030 versus something that's bigger 222 00:12:26,030 --> 00:12:28,650 and getting more of your information 223 00:12:28,650 --> 00:12:31,090 like you're sacrificing privacy for information 224 00:12:31,090 --> 00:12:33,210 Yeah, that is the trade-off 225 00:12:33,210 --> 00:12:34,230 that we want to avoid. 226 00:12:34,470 --> 00:12:36,270 Right now, people are. 227 00:12:36,550 --> 00:12:38,650 I think we're probably like 90% of the way there, 228 00:12:38,710 --> 00:12:39,650 95% of the way. 229 00:12:40,170 --> 00:12:43,190 But the frontier models is what we call them. 230 00:12:43,270 --> 00:12:44,430 Those ones are still the best, 231 00:12:44,530 --> 00:12:46,550 the closed frontier models of OpenAI 232 00:12:46,550 --> 00:12:47,750 and Anthropic and others. 233 00:12:48,570 --> 00:12:50,410 Open source has really closed the gap a lot. 234 00:12:51,030 --> 00:12:52,790 And then if we add this memory feature in, 235 00:12:52,790 --> 00:12:54,450 that also helps close the gap even more. 236 00:12:54,690 --> 00:12:56,270 And then I think there are things that we could do 237 00:12:56,270 --> 00:12:59,270 to kind of do local, like anonymize your data 238 00:12:59,270 --> 00:13:00,970 and then pass it over to ChatGPT 239 00:13:00,970 --> 00:13:03,670 to take advantage of the benefits that they get. 240 00:13:04,150 --> 00:13:05,870 But frankly, I think that a lot of people 241 00:13:05,870 --> 00:13:10,970 actually don't need the most powerful model all the time, right? 242 00:13:11,010 --> 00:13:12,650 There's a lot of times where you just need something 243 00:13:12,650 --> 00:13:14,950 that gets you 90 to 95% of the way there. 244 00:13:15,410 --> 00:13:18,050 I love to make the analogy of a car. 245 00:13:18,190 --> 00:13:20,570 You don't need to drive a race car to work. 246 00:13:20,630 --> 00:13:21,990 You just need to drive a car that gets you there. 247 00:13:22,730 --> 00:13:24,930 And so that's where we're at right now. 248 00:13:24,930 --> 00:13:28,430 But I think the only way for us to win in the privacy game 249 00:13:28,430 --> 00:13:31,430 is to make experience as good or better 250 00:13:31,430 --> 00:13:33,590 than the one that is invading your privacy. 251 00:13:34,390 --> 00:13:36,690 If not, people are going to choose convenience every time. 252 00:13:37,550 --> 00:13:38,830 Yes, yes, I know. 253 00:13:38,930 --> 00:13:42,370 It's like we're all getting so used to this easy information 254 00:13:42,370 --> 00:13:44,230 where it's like a few years ago, 255 00:13:44,370 --> 00:13:46,270 it's like I actually had to put more work 256 00:13:46,270 --> 00:13:49,290 into learning or writing something. 257 00:13:50,870 --> 00:13:51,730 Good and bad. 258 00:13:54,550 --> 00:13:57,310 So I guess as you think about 259 00:13:57,310 --> 00:13:59,590 kind of like what are you thinking most right now 260 00:13:59,590 --> 00:14:03,710 what the future of AI generally should look like 261 00:14:03,710 --> 00:14:05,210 maybe in like five or so years? 262 00:14:06,210 --> 00:14:07,330 Oh yeah, that's a big question. 263 00:14:07,450 --> 00:14:09,590 Okay, how big vision do you want me to go here? 264 00:14:09,610 --> 00:14:10,350 As big as you want to go. 265 00:14:10,350 --> 00:14:13,410 Okay, I really see the need for us 266 00:14:13,410 --> 00:14:15,850 to have an entire ecosystem for AI 267 00:14:15,850 --> 00:14:17,710 that is open and verifiable. 268 00:14:18,250 --> 00:14:19,530 And I would equate it to the internet. 269 00:14:19,770 --> 00:14:23,030 I think that the internet originally started proprietary 270 00:14:23,030 --> 00:14:24,830 and it wasn't growing very much. 271 00:14:24,830 --> 00:14:31,670 It wasn't until we built the Internet on open protocols that it just exploded and flourished all over the world. 272 00:14:32,570 --> 00:14:33,930 AI needs to have a similar moment. 273 00:14:34,030 --> 00:14:35,250 Right now, it is proprietary. 274 00:14:35,670 --> 00:14:36,190 It's closed. 275 00:14:36,510 --> 00:14:37,690 It's a privacy nightmare. 276 00:14:38,770 --> 00:14:44,530 And the problem is, is I think it could take over the world that way. 277 00:14:45,630 --> 00:14:51,970 And we're seeing it with like other governments that are being more oppressive and using AI in certain ways. 278 00:14:51,970 --> 00:14:54,350 so I would love for us to 279 00:14:54,350 --> 00:14:56,730 we really need to get going on this now 280 00:14:56,730 --> 00:14:58,410 but have 281 00:14:58,410 --> 00:15:00,610 five years from now that the popular 282 00:15:00,610 --> 00:15:02,250 AI is all built on open protocols 283 00:15:02,250 --> 00:15:04,510 and open source and open ethos 284 00:15:04,510 --> 00:15:06,450 and privacy protecting and encrypted 285 00:15:06,450 --> 00:15:08,790 and it's not just AI chat 286 00:15:08,790 --> 00:15:10,830 that's the current incarnation of it 287 00:15:10,830 --> 00:15:12,550 it's going to be inside the car 288 00:15:12,550 --> 00:15:14,510 that you're riding in, it's the robot 289 00:15:14,510 --> 00:15:16,770 that's in your house, it is the wearable 290 00:15:16,770 --> 00:15:17,630 that you have on 291 00:15:17,630 --> 00:15:20,750 it's every single service you interact 292 00:15:20,750 --> 00:15:25,570 with online, like everything's putting AI inside of it. All of those need to have some sense of 293 00:15:25,570 --> 00:15:31,210 verifiability within their stack so that we know where our data is routing, how it's being processed 294 00:15:31,210 --> 00:15:39,310 so we can avoid, you know, effectively censorship, surveillance, top-down level control over the 295 00:15:39,310 --> 00:15:44,950 general populace that's using AI. You seem to be using a lot of the same keywords that 296 00:15:44,950 --> 00:15:50,190 us and the Bitcoin. Oh, yeah. Well, I know you were in Bitcoin before, but there seems to be a lot of 297 00:15:50,190 --> 00:15:55,510 overlap. And I mean, obviously, I work in like the open source space in Bitcoin. Do you see like the 298 00:15:55,510 --> 00:16:01,570 AI open source space kind of like growing in similar ways? Anything they can learn or collaborate 299 00:16:01,570 --> 00:16:06,790 with Bitcoiners in the open source space? I think so. There seems to be this thing going on where 300 00:16:06,790 --> 00:16:11,210 Bitcoin and AI keep getting brought up in the same conversation, even though technically there's 301 00:16:11,210 --> 00:16:19,510 no overlap at all between the two technologies. But the Bitcoin and AI both have a similar 302 00:16:19,510 --> 00:16:21,690 need for really good privacy, right? 303 00:16:21,710 --> 00:16:23,670 And they will succeed the most for humanity 304 00:16:23,670 --> 00:16:25,830 if we have privacy built into them. 305 00:16:26,390 --> 00:16:28,070 And so I think that's one reason 306 00:16:28,070 --> 00:16:29,030 why it keeps coming up. 307 00:16:30,450 --> 00:16:31,510 I forgot your question. 308 00:16:32,230 --> 00:16:32,710 That's okay. 309 00:16:33,230 --> 00:16:34,590 Similar ethos, I guess. 310 00:16:34,590 --> 00:16:37,470 Similar type of people who care about AI for good 311 00:16:37,470 --> 00:16:39,390 that also care for Bitcoin. 312 00:16:39,390 --> 00:16:40,170 Yes, yeah. 313 00:16:40,290 --> 00:16:43,010 And really what we're seeing at Maple is, 314 00:16:43,770 --> 00:16:53,795 I super grateful for the Bitcoin community We part of the Bitcoin community but they have been our most early adopters and our biggest supporters And Maple would not be as far along today as it is 315 00:16:53,795 --> 00:16:55,915 if it weren't for the Bitcoin community 316 00:16:55,915 --> 00:16:56,835 who cares about privacy 317 00:16:56,835 --> 00:16:58,315 and shares a lot of those same values. 318 00:16:58,555 --> 00:17:01,875 So it's been a great spot to kind of start and build. 319 00:17:02,375 --> 00:17:04,935 We really feel like we're building the AI for Bitcoiners. 320 00:17:05,555 --> 00:17:07,795 We're building, similar to the Signal app, 321 00:17:07,855 --> 00:17:08,715 what they did for messaging 322 00:17:08,715 --> 00:17:10,135 and what Proton did for email, 323 00:17:10,715 --> 00:17:12,755 where Maple is going to be that for AI. 324 00:17:13,255 --> 00:17:19,235 Yes, you said in your presentation the other day, Signal and ChatGPT had a baby. 325 00:17:19,335 --> 00:17:19,995 Was that what you said? 326 00:17:20,215 --> 00:17:21,335 Yeah, that was Matt O'Dell. 327 00:17:21,475 --> 00:17:22,135 Oh, okay. Sorry, Matt O'Dell. 328 00:17:22,135 --> 00:17:25,155 He said it on a podcast recently, so I just quoted him. 329 00:17:25,515 --> 00:17:27,955 Okay. That was funny. I liked it. 330 00:17:28,815 --> 00:17:32,155 But someone in the audience was like, I said that to you yesterday. Why don't you quote me? 331 00:17:33,135 --> 00:17:33,595 Everybody's thinking it. 332 00:17:33,595 --> 00:17:34,915 Pretty funny, right? I guess so. 333 00:17:34,975 --> 00:17:35,975 Everybody's thinking it. 334 00:17:35,975 --> 00:17:43,295 is there anything else that like is top of mind for you right now 335 00:17:43,295 --> 00:17:46,955 or any rabbit holes you've recently gone down? 336 00:17:47,875 --> 00:17:48,855 That's a good question. 337 00:17:49,295 --> 00:17:51,435 I wish I'd been thinking of that a little bit more. 338 00:17:53,055 --> 00:17:54,855 Rabbit holes I recently went down. 339 00:17:55,495 --> 00:17:56,755 I mean, I'm thinking a lot about education. 340 00:17:57,195 --> 00:17:59,735 I have kids who are in school right now 341 00:17:59,735 --> 00:18:05,735 and AI is kind of definitely affecting that industry. 342 00:18:05,975 --> 00:18:22,315 And there are good and bad pieces to it, right? And so I'll start with kind of the good side, which is I have some teachers of my kids are embracing AI and giving them new assignments that are tailored to AI tools and teaching them how to use them. 343 00:18:22,315 --> 00:18:36,335 So for example, my daughter came home the other day and said, hey, dad, I'm supposed to produce like a three minute, a five minute video of a short story that I wrote and I have to produce like an audio podcast with like a video element to it. 344 00:18:36,635 --> 00:18:38,935 And we're specifically supposed to use AI tools. 345 00:18:38,935 --> 00:18:44,675 So we hopped on 11 labs, we generated an audio book out of her script or out of everything that she wrote. 346 00:18:45,095 --> 00:18:47,475 And then we hopped into Chagikati and generated some images and stuff. 347 00:18:47,475 --> 00:18:53,175 So it was pretty cool to like work with her on these cutting edge tools that I use every day for work. 348 00:18:53,775 --> 00:18:55,995 And so, and yes, I use ChatGPT. 349 00:18:56,155 --> 00:18:57,515 We don't generate images in Maple again. 350 00:18:57,855 --> 00:18:59,155 So we had to use that. 351 00:18:59,655 --> 00:19:01,135 So I think that's really cool. 352 00:19:01,215 --> 00:19:02,695 There's a lot of promising stuff to happen there. 353 00:19:02,775 --> 00:19:06,515 I think kids need to learn how to use AI in a good way. 354 00:19:06,835 --> 00:19:13,095 And then the flip side of it is a lot of school districts are signing up for AI and just giving it to kids, right? 355 00:19:13,095 --> 00:19:17,355 So my kids use Google Classroom and I wish they would use Canvas. 356 00:19:17,475 --> 00:19:20,795 But they use Google Classroom and it has AI right in front of them. 357 00:19:21,395 --> 00:19:30,035 And we had a lot of parents over the last few years who went to school board meetings arguing over the books that were being assigned in school, saying that the content wasn't good. 358 00:19:30,535 --> 00:19:32,715 But now we have AI being assigned in schools. 359 00:19:32,715 --> 00:19:36,755 And I'm not really hearing a lot of people talk about, oh, which AI are you choosing? 360 00:19:37,115 --> 00:19:38,075 What are the biases in there? 361 00:19:38,315 --> 00:19:39,515 What are they going to learn from it? 362 00:19:39,535 --> 00:19:40,155 What are the restrictions? 363 00:19:40,835 --> 00:19:47,395 And I think that's actually this, I don't know if Trojan horse is the right phrase, but that is something that really needs to get more attention. 364 00:19:47,475 --> 00:19:53,595 is like what AI is being sit in front of all of our kids across America. 365 00:19:54,315 --> 00:19:57,895 I don't know if you've ever talked to Matt Velez on the Spiral team. 366 00:19:58,155 --> 00:19:58,935 No, no, I haven't. 367 00:19:58,935 --> 00:20:07,475 But he has four kids and he's also very much thinking about how his kids learn to use AI in a responsible way. 368 00:20:09,235 --> 00:20:13,355 I just think a lot of these kids that are young right now are probably going to be super geniuses. 369 00:20:13,355 --> 00:20:19,935 If they learn how to use AI well, responsibly, and the right tools at like a young age, hopefully. 370 00:20:20,615 --> 00:20:20,735 Yeah. 371 00:20:21,175 --> 00:20:27,455 If they learn how to use it well, because there are some reports coming out right now that shows that their brain kind of atrophies. 372 00:20:27,815 --> 00:20:37,315 And so, which is good for us to learn now so we can figure out like, okay, how do we adjust the usage of AI so that they are flourishing instead of going in the wrong direction with it? 373 00:20:38,635 --> 00:20:41,555 And there's more studies that need to be done on that. 374 00:20:43,355 --> 00:20:58,535 Well, I have lots more questions, but we'll just, we have one more minute left, so I'll leave it with the final question. If you could leave our listeners like a single thought about AI, Bitcoin, anything else, like what would it be? 375 00:20:58,535 --> 00:21:14,215 Okay. I guess I'll leave with the thing that I tell a lot of people, and that is I'm not here to tell people to uninstall ChatGPT. Use the best tool for the tasks that you have. But I would encourage people to find a privacy-preserving AI that's out there. 376 00:21:14,215 --> 00:21:38,635 I prefer Maple, obviously. There are other ones. But download and use something like Maple and have it side by side in your toolkit of AI. And then as you're using AI and you start to run into some kind of constraint where you're like, I don't feel totally comfortable sharing this right now with ChachiBT, hop into Maple and talk about it there and see how you feel. See what your vibe is. 377 00:21:38,635 --> 00:21:41,035 and we have a lot of people who tell us 378 00:21:41,035 --> 00:21:43,215 like I felt liberated when I went 379 00:21:43,215 --> 00:21:45,255 and used Maple and I was chatting in there 380 00:21:45,255 --> 00:21:47,115 and I finally feel like I'm free 381 00:21:47,115 --> 00:21:49,235 to be myself. I'm free to think. I'm free to talk 382 00:21:49,235 --> 00:21:51,235 how I want to. So I just 383 00:21:51,235 --> 00:21:52,815 encourage people add Maple to your toolkit. 384 00:21:54,035 --> 00:21:55,275 You know, you can have five AI apps 385 00:21:55,275 --> 00:21:56,835 on your phone. Just make sure Maple is one of them. 386 00:21:57,255 --> 00:21:59,255 Well, I have it. So I'm taking 387 00:21:59,255 --> 00:22:01,295 you up on it. Okay. If our 388 00:22:01,295 --> 00:22:03,495 listeners want to follow more what you're doing 389 00:22:03,495 --> 00:22:05,195 or Maple, where should they go? Yeah. 390 00:22:05,335 --> 00:22:07,095 Try Maple.ai's website 391 00:22:07,095 --> 00:22:09,255 and then I'm on X 392 00:22:09,255 --> 00:22:11,355 and Noster X is Mark Suman 393 00:22:11,355 --> 00:22:13,875 and then Noster I'm Marks at Primal.net. 394 00:22:14,655 --> 00:22:15,495 Check us out there. 395 00:22:15,955 --> 00:22:17,455 Well, thanks so much for coming on the show. 396 00:22:17,675 --> 00:22:19,815 I hope you enjoy the rest of your time in San Francisco. 397 00:22:20,395 --> 00:22:20,895 Great. Thanks. 398 00:22:21,175 --> 00:22:21,755 This was a pleasure. 399 00:22:21,975 --> 00:22:22,575 Oh, good. Thanks. 400 00:22:22,575 --> 00:22:22,775 Yeah.