Erik Voorhees: Private AI Is Up Next (Why Venice Is Leading)

Transcript

Welcome back to the rollup. I'm Robbie. >> I'm Andy. Rob and I met at the University of Florida in 2017 where we first found out about digital assets. >> We learned a lot along the way and we're bringing you face to face with the leaders of our industry. >> Sit back, relax, and enjoy today's episode. >> Back live at Nearcon Day 2, the last segment of the day. We've got Eric Vorhees, uh, notably the CEO of of Shape Shape Shift amongst others. >> First place I made an onchain transaction. >> Was it? >> Yeah. Eric, so story time for you is Rob and I met at the University of Florida in the uh blockchain club there, Bitcoin club in 2017. Part of our, you know, inauguration and getting started there was bringing on speakers into Zoom and teaching us things about crypto. And

with the help of a friend named Antonio and Felipe, we actually called you in. You zoomed in in 2017 and did a live Shape Shift demo for us. talked for about 30 minutes about uh censorship, resistance, kind of libertarianism, and then you were gone. >> Yeah, that checks out. >> And here we are today. >> That checks out. That was the best crypto year of all 2017. Yeah. >> Yeah, man. Well, welcome to Nearcon. >> A lot has happened since then. >> Yeah. A lot of a lot has happened. Uh industry is a big thing now. >> Yeah. Eric, one of the things that I think is super super pressing in the industry now and it's the cause of a lot of the sentiment disruption is that we've built a lot of this infra open source infrastructure at scale. Um, but

we haven't quite figured out value capture from it and there's a lot of these kind of the anti the antithesis of what we've built here being institutional funds, trady, if you will, coming on building on the rails and capturing a lot of the value from it. people have a really really bad feeling about this kind of environment. Also with the like the era of pragmatism coming about where it's much more about users customers products revenue and business and less about the the core ethos and less about why we're here. How do we as an industry go into this next era of pragmatism of institutional adoption while maintaining the core ethos of what we've built?

>> Great question. Uh I think the core ethos is um is protected by the protocols, right? Uh the ethos is built into Bitcoin. Um and that that doesn't change regardless of who's using it. So um it's the protector is really the code that people have written and that that's been the whole point. Uh I I don't take the same view that it's um so problematic that all these institutions get involved. I think that was always hopefully inevitable. The point is that uh that crypto takes over the world and to do that it has to merge with the world. And so um this should be a sign of of success. Now what would not be a sign of success is uh if somehow we all

ended up just using like custodial companies and there was no way to use anything else. >> Uh if there was no alternative, if there was no DeFi, if there was no self-hosted wallets, like if if that got lost somehow, then we'd be in big trouble and we wouldn't have created anything useful. >> Yeah. But that's not our situation. >> What do you think is the main limiting factor for mass adoption and that could come with institutions being the distribution channel to citizens and users ultimately? Is is that primarily the the vector that we're exploring or is there some other limiting factor that could bring the masses?

>> Uh I think the masses just have to happen over time. Like what what we've done with crypto is fundamentally change the way that money can work. Uh but people's behaviors take a lot of time. Like people have inertia. They get used to certain behaviors and uh that can't be flipped off overnight. Um there is so much friction and inertia in the the way businesses transact with each other and people transact with each other like that just really takes a long time. Um but if you look at sort of crypto adoption by age group uh it grows as the group gets younger. So it is multi-deade phenomenon but you know like my grandchildren probably think crypto is

pretty pretty normal. Even even my my children will >> Yeah. when you were kind of early pioneering in Bitcoin and just generally in in DeFi 2015 2013 era all the way till 2017. >> Well, there there was no DeFi in 2013, >> right? >> Pre Ethereum, pre smart contracts, >> right? >> But you you had part of the vision, right? You saw some of this or at least you saw >> Bitcoin Bitcoin, I guess, was the first DeFi project, right? It's like it did defy for money. >> Yeah. >> Yeah. What was the inspiration for ShapeShift which ultimately kind of spurted off decentralized exchanges DeFi from there?

>> Uh so I started ShapeShift um back in 2014 after Mount Gaus collapsed which was like the you know 95% of all the exchange market back then uh big centralized custodian and almost destroyed the whole industry. So I was like well how do you build a exchange that doesn't hold any customer money and doesn't require them to reveal who they are? you just protect people by the design of the of the exchange. So yeah, ShapeShift was the first non-custodial exchange. Um, and yeah, we built it, you know, two years before Ethereum came out and Ethereum came out and the whole explosion of different assets really uh caught fire. So it was good timing.

>> Yeah. Did has this played out the way that you thought? Has the has the last 10 years so 2016 to 2026 did you envision kind of obviously you could have predicted it perfectly but was this like part for the course? Was it way off? Was it close? >> Um at least for me I did not predict or understand what a smart contract was before that happened. Yeah. So it wasn't like I was like yeah someday there will be these smart contracts and everyone will be using DeFi primitives. Like that was a little alien. uh and even when Ethereum came out, it took me a while to understand the value of what that was.

But, uh, crypto is, you know, like slowly taking over the world and and, um, Bitcoin has become, you know, top 10 asset on planet earth. That is absolutely what I was hoping to see. >> Yeah. So, in some ways it feels like it's happened faster and other ways slower than I would have wanted, but it's happening. >> Yeah. >> Do you see similarities in the way AI is playing out now and the way crypto played out in the early days? No, not I think they're totally different. A >> AI had this like long academic history going back decades. >> Uh but it wasn't ever quite potent enough to break through in the mainstream and then the chat GBT moment happened and suddenly it took over everything like it's just it's been a

few years and AI is everywhere and it it became completely mainstream within two years. Crypto still is not completely mainstream. Yeah. And uh and crypto also didn't cryptocurrencies at least did not have you know like 80 years of of academic history. It was a much more recent phenomenon uh and has been slower growing during that time. Whereas AI was like you know being worked on forever and ever and ever. I mean, imagine being like someone in AI and living through 50 years of disappointments and it never achieving the things that you wanted.

>> Yeah, man. >> Um, but now it it finally certainly has. >> Yeah. Let's talk about about the work that you're doing at Venice, Eric. Uh, we're at we're at Nearcon. Kind of this convergence point between these these two industries. Uh, Venice is a part of a of a new kind of asset in crypto. there's this this idea of kind of doing GPUbacked tokens in crypto. Um you've got one called DM, you've got USDA, Venice is is you know effectively in that uh category as well doing novel kind of uh GPU experiments with tokens.

How's that been? What like what's the short and sweet on Venice? Like if you had to explain it to somebody who is it's their first time. Uh and why is this something that you're spending your time on? Uh yeah, so that's it's my it's my full-time thing now. Uh has been for two years and and basically I wanted to build a private uncensored version of chat GBT. Yeah. >> So I I loved chat GBT, but I was like this is super dystopian unless someone builds something with actual like crypto ethos in it. >> Um so it is first and foremost a consumer app for AI. Um and of course me being a crypto guy, I'm going to bring crypto elements into it. So you mentioned DM, that's one of the the tokens of of the Venice ecosystem.

>> Yeah. DM basically lets any holder of it get free compute uh a dollar per day of free compute. Um and so we built this you know kind of as an experiment to see like uh will agents um like the ability to get uh essentially like inference at zero marginal cost instead of paying per call they get a free free inference up to a certain amount every day. Um, so yeah, we're just playing around with these kind of um primitives within within AI. And I think uh the AI and crypto world are moving increasingly close together. I think we may all realize at some point in the future that like cryptocurrency was actually just being built for the AI robots >> and the human humans have always been awkward with it, but like agents and and

um machines will understand it easily and have no problem with the UX issues. So, it's interesting to see these worlds merging and uh it's been fun to build in both of them. >> What was a turning point from when you were working with crypto and then ultimately, you know, now with Venice? Was there a point in the middle there that was it the chatbt moment that kind of lit a light bulb and said, I'm going to really focus on this new >> venture. Yeah. So, we we decentralized shape shift in 2021 and I took a couple years off and um I didn't know what I would do next. Uh then the AI thing happened and um I just to me it was one of those clear cases where like someone need someone who cares about the principles that I care about needs to

build >> an alternative to what these will become. >> Yeah. >> Um no one in the AI world cares much about free speech. No one in the AI world cares much about privacy. uh I'm generalizing but like those are relatively foreign concepts to a largely previously academic community that u was very much in favor of always controlling people and just making sure the right person was controlling them. That's a totally different mentality than what we have in crypto. So um I felt like I had to do something and built Venice for that.

>> Yeah. So in the ecosystem you know I mentioned DM you you're talking about the private uh kind of uncensorable interface like I if this works like are we are we building an entire economic structure with the principles of kind of you know that you share with the early stages of crypto you know censorship resistance decentralization immutability etc building that in the AI lens with a fully self- sustaining kind of economic model inside like what is the ecosystem vision that you have for Venice. Um you know like like what can we what can we kind of like like you know tie this to or map this to in our minds?

>> I the the simple way to think of it is just like um Venice is a a consumer app trying to do a very specific thing which is like provide inference that is private and that is not censored. >> Yeah. >> And uh everything that we do is built to support that that mission. So, we're not trying to do everything or create we're not really trying to create an ecosystem. We're just trying to build a a really important product and infuse within it the the ethos that I find so important. >> Havachi is fast and private per trading secured by ZK with Celestia underneath.

Go to habachi.xyz to start trading today. >> Looking for yield that stands out? Infinify gives you just that. Deposit stable coins and earn far more than the current savings rate. Infinifi is a battle tested reliable DeFi platform that is ready to go. Head to infinifi.xyz. when when you see some of the foundational labs um censoring at their model level um does that give you more valition around what you're building to essentially provide that uncensorability to whether it's claude or trasht they can people that were being centered can now use venice as a vector to explore those models >> yeah so like you know those are private

companies and they can set whatever rules they want on their models right that's that's fine Um what's not fine is if that's the only alternative and what's not fine especially is if governments come in and basically um demand that that kind of censorship must happen. >> And the risk with those companies like OpenAI and Enthropic is that they will and have lobbyed governments for regulation uh to enshrine the way that they do things as as the correct way. Um, so I find that problematic and you know, Venice is a consumer app as an alternative, but the the amazing thing that's happening is that like Chinese labs are actually the greatest bastion of uh >> open source

>> open source uh models here. It's so it's super ironic uh very weird and strange >> but um yeah as as long as open source models are uh competitive and you know widespread which they're getting increasingly powerful then we're good you know and it's uh it's all fine. Go anywhere onchain instantly with relay.link. Relay is closing in on 10 billion transferred. Go to relay.link to bridge your assets now. What do you make of the geopolitical evolution there being the Chinese open source models and the closed source private AI companies in the US?

>> I don't I don't know. I don't know. Um, you know, and it's always it's always hard to it's always hard to figure out like to what degree things are centrally coordinated like is the Chinese government like dictating everything that's going on. It's easy to think that way because people tend to think of nation states as like personified entities that act with pure focused attention. But um it's all just a in my opinion like a messy phenomenon that no one is really sure how it's going to unfold and there's lots of incentives going on all at the same time.

>> Yeah. Eric, uh last thing that I have, you had a very famous uh interview with SPF uh before he uh blew up. Um you you know you were right notably um and recently SPF's gotten his phone back and has been on Twitter. >> Great. >> Um >> today Trump said that he's not going to pardon. >> Correct. Trump is Trump is confirmed that even though that SPF is quite literally on his knees on Twitter for Trump. He's not Trump will still not pardon him. Just wanted to give you a chance to comment on that. >> I mean it does not seem that he is uh worthy of a pardon. No. Uh No. Uh >> he's trying to say that he that he was always liquid, that he was always

solvent, and that due to external forces, his insolveny was forced upon him rather than and that he's been this great investor in AI. >> I just don't think anyone believes that story. He has been a great investor in AI, that's for sure. He made a $500 million investment in anthropic. Of course, did it with other people's money. Um, amazingly that investment turned out to be the thing that is making most of the anthropic creditors whole or mostly whole or uh >> FTX um yeah, FDX creditors whole but that was that was luck. Um and uh yeah, I don't think SPF deserves anything but his sentence and he doesn't really deserve the attention anymore. I I'm so sick of all the scammers in crypto that

have just like caused such damage to the industry. There's been so many of them. Yeah. >> Um, >> yeah. >> And it's been a bummer because, you know, the rest of us have to deal with the consequences of that. >> Yeah. Especially, I mean, in AI, we've seen it as of late more and more where there's just this complete separation and just desire to not be affiliated with crypto at all. >> Yeah. And it's it's understandable. Like from the outside perspective, you can be forgiven if you think crypto is a scam because if you just if you read articles once in a while and you have a couple friends that got ripped off, like what else are you supposed to think?

>> Yeah. um you know and I can't sit down with everyone and have a 30 minute conversation about the philosophical principles that undermine you know like sovereignty in the realm of money you know so it's a it's an uphill battle. >> Yeah. Eric last thing I have for you we're at New York. You had a chat with Ilia on stage and I'm I'm curious any takeaways that you had from that chat and uh how you see Venice and Near working together in the future. >> Um yeah I just I was glad to be here and to have a chance to talk about privacy in in AI. I mean, Ilia is one of the very few people that is that has deep credibility on in both worlds, right?

Like almost no one can actually claim that. I certainly have no credibility in AI. Like I'm a I'm a tourist. I'm learning about it in the last couple years, but um he he has both. And so this is uh it's very cool to see the this event coming together as a fusion of these two worlds. And I hope it gets um bigger and more I I want more of that integration to happen. So he's he's one of the few like figureheads that could actually do that I think. >> Yeah. >> Eric, as we said, we you've inspired us since 20167. First time on on on the show. We'd love to make it a recurring thing. Thanks for coming on, Eric. Hope you have a great rest of your conference.

>> Cheers. Cheers, man.