From Bits to Atoms: How DePIN Is Making Crypto Real | Shannon Wells - Livepeer | Rob Solomon - DIMO

happy to see all you here in the crowd thanks for showing up this morning thanks for being here after a really solid day one of content programming here at e Denver day two also looking pretty good here at the Atlantis main stage we got D5 we got rwas we got really strong privacy Block in the afternoon we have deepin to kick things off that is a thing that I heard I heard it's going to be deep in summer this summer I was out last night and someone said deep in summer is a thing so to dig in into that and emerging Trends in the decentralized physical infrastructure Network space I'm going to welcome our moderator to the stage Adam wasney of Akash he's going to welcome our panel let's hear a nice big round of applause for Adam and the Gang let's do this all [Applause] right hello hello hello good morning eth Denver day two eth Denver how's everyone doing today woo we're having a good morning everyone's fully caffeinated ready for some deep pin talk awesome well thank you guys for joining today's panel um my name is Adam wosy I head up Community for overclock Labs the founders of the aash network um I'm honored to be the moderator for our panel here today um in the landscape dominated by centralization oo can we all oo Collective shiver um deepin offers a new paradigm it challenges Norms empowers communities and fosters Innovation so needless to say I'm very excited for today's conversation before we get into any of that I wanted to introduce our guests so first up we have Rob Solomon Rob is the co-founder of demo which is an open platform for connected vehicles and developers who want to build smart contracts for them he's definitely a blockchain guy he wrote that I just got to put that in there he's definitely a blockchain guy who prior to demo worked for consensus for four years thank you for joining how are you doing today rob you look great by the way you look fantastic doing great thanks for coming at 10 appreciate you guys being here absolutely uh next up we have Shannon Wells Shannon is the head of ecosystem at live Pier she's an award-winning policy and governance expert we were talking about that earlier before we got on stage creative systems thinker and entrepreneur passionately building a future that empowers people and places on live Pier thank you for joining Shannon uh we were talking about this briefly but you're in Canada right you live in Canada I am yes excellent oh yeah you can pull a little little closer yes I live in Canada excellent um I've been to Canada before but I've actually never been any kind of crypto events in Canada what's the general crypto vibe in Canada uh I Canada was one of the first places to start hosting crypto events I'd say the early Bitcoin Community was actually out of Toronto um I mean vitalic was obviously out of waterl founder of Cosmos actually is um from a place just outside Toronto as well so I like most um Technology Innovation in Canada were really good at at develop at developing it commercializing it not so much so we've seen most of the growth um come outside Canada but um I I think Canada's been um also interesting on the regulatory side I think that there's been a little more Innovation around sandboxing um for defi projects in Canada um there's actually the the first Bitcoin ETF was actually in Canada last year so people think Bitcoin ETFs actually just um came to life recently but they were they've been in Canada for a while there's also a theum ETFs in Canada already as well awesome so Canada has crypto in short excellent thank you for joining um and finally we have Mustafa Kazi uh Mustafa is a research engineer at six-man Ventures where he publishes research on cryptoeconomics he previously worked with the ethereum foundation researching EIP 1559 snaps to that um Game Theory uh welcome Mustafa uh how's your eth everment going it's good it's good it's a lot more I mean like popping I think than last year and even the side events are like two three times as is Big so it's like it's great to see yeah excellent um cool so I wanted to give you guys a little bit of time to briefly describe your project so everyone's kind of familiar with like how it's relating to deepin so maybe starting with Rob with demo give us a little introduction to demo yeah so as you mentioned demo is a platform for cars we allow anybody to turn their cars into smart devices and connect to the world's first open Mobility platform that happens to be built on blockchain kind of needs to be for us to actually be able to deliver this uh what does that mean to have a smart car uh if you have a car now it's probably more like a flip phone or a landline in terms of how smart it is you have a Tesla it's kind of in that Blackberry territory um but we're here to usher in that iPhone moment for cars where connectivity data identity uh and a kind of a shared digital infrastructure for vehicles leads to an explosion of apps and use cases you can think of like we've been talking about web 3 Uber since I got into crypto years and years ago we make that possible defi auto loans actually the original uh idea for demo was you pull up to a parking garage the sensors in the garage talk to your car tell it where to go park pulls your Ford in you know pulls a BMW out to pull your like just that kind of shared digital infrastructure to build these types of applications that's what we're building and it could only really work with blockchain uh I don't drive a Ford so it's not pulling my Ford I have sub sorry shout out to um I'm not getting paid by them thank you for that description Rob um and let's move on to Shannon uh give us a quick description of live Pier what should folks know sure so live Pier is a decentralized video compute Network and we think of what deepin is it's a lot of different things it can be uh compute networks it can be storage networks it can be bandwidth um and at live Pier we're really focused on on video and the infrastructure stack um needed needed for that and so the most important part of compute regarding video is actually transcoding and so we're a distributed Global Network of node operators who provide CPU and GPU um compute through encoding and decoding trips on CPUs for transcoding and and why does transcoding matter um if you upload a video if you record a video on your phone and you want to broadcast that to anyone anywhere in the world on any device no matter what their internet speed that file needs to be uh repackaged it needs to be um kind of compressed and put in multiple different file formats so that's really compute intensive and it's really freaking expensive um YouTube sold to Google in 2008 because it couldn't cover its infrastructure costs twitch sold to Amazon um because it couldn't generate enough Revenue to cover its infrastructure cost um and that's all related to transcoding um so we have built software on top of this distributed um infrastructure that is 10 times cheaper than traditional cloud service providers whether that's AWS or Muk because we don't run any data centers or have any um capex overhead uh for the software we build right on thank you for that yeah it sounds like you got a lot going on over there it's really important stuff so thank you let's move on to Mustafa um hoping you could give a quick introduction to six-man Ventures yeah so six-man Ventures was founded by Mike dudus and serge cassarian about four years ago um we're an early stage web 3 fund uh preced seed series a and and we invest in a lot of deepin projects um I do research so we pride ourselves on doing more data driven research we've done some pieces on token unlocks we looked at over 5,000 unlock events and gave some uh recommendations on decisions currently working on one for airdrops so um doing a lot of digging through data for that finding and verifying a large data set um to kind of give Founders better idea of what they how they should structure their airdrops um but yeah we uh we also wrote some articles on dpin we have an in-house agent-based simulation on dpin that we use for uh writing those articles um yeah so we're excited for to keep investing in the space and keep growing the space very cool well I'm I'm really excited to have you guys here I think we have very unique perspectives on the topic of deepin from what each of you are working on um so I thought a good way to kind of kick off our conversation is just a high level description of deepin and maybe you guys prob a little color to it um it's actually hard to find like a clear description of dpin so I kind of half created it half pulled what I found online but um I want to get your guys' opinion on this so dpin is decentralized physical infrastructure networks uh they're blockchain protocols that build maintain and operate physical Hardware infrastructure in open and decentralized manner at a high level would you say that's pretty accurate or would you guys like to provide any color to kind of your perspective on what dpin is yeah I think one thing that's really important about these networks is that they're deployed by a diverse group of people you know it's like it's not like it's four company in a Consortium it's often they're very popular Global permissionless networks you can contribute to yeah that's fair yeah I think that decentralized nature is is the most important thing and um you know it's the crypto incentive layer on top of that distributed Network I think that matters most when we talk about deepen in this space um you know there is a lack of supply of compute needed for you know the increasing demand for things we want to do on the internet most notably AI I think ai's really been driving this this current boom of deepend and I think it's probably if we do get to deepend Summer um what will continue to sustain that there's just an insatiable demand for people um to use you know Chad gbt um micro Microsoft and open AI just really sore their text a video platform and so that's um there's a real Monopoly today in terms of who who can provide that um if you look at Nvidia which is you know producing like over 90% of the world's gpus um you know it's it's actually their software it's a Cuda software that you have to use in order to accelerate those gpus um and it's why they have such a strangle hold on the market they you know their data center revenues grown 400% over the last year their last quarterly results for data center Revenue were like over 18 billion it's crazy so um with crypto you know you can really um create an incentive for infinitely scalable Supply um of this compute you know there's a lot of issues around we still rely on you know Nvidia GPU chips but I think it's that incentive layer of being able to reward um you know smaller players who can put to work idle compute resources you know folks on the live pier network also validate and provide resources to other deepend projects including storage um so I think it's that coordination layer um across different you know discret aspects of um specialized Compu or specialized infrastructure that's really neat in web 3 right and Mustafa like from your Vantage Point like how do you guys frame kind of deepin at six-man Ventures yeah I think um a good mental model for people who are brand new to deepin is kind of like the franchising model for like fast food um instead of because you know capex is hard when you're a large um Geographic spread um and you have to deploy Capital there um the franchise model allows you to just create a CER standard that you deploy to other individuals so you kind of do that with infrastructure right you do that with um phone service Wi-Fi with gpus um so that's kind of how we see it from a very high level I think it's an easy way for people to understand what this is and identify the sector um but then you go down the rabbit hole of of people saying everything is deep pin and then you can't have a why not why not do it I think we can figure it out together if we put our brain power behind it maybe after this talk we'll figure it out thank you for that so I kind of want to start around just like the state of deepin like where it stands today so maybe going down the line we can go from Rob here um from your Vantage Point what is the current state of deepin and would you say deepin is really ready for Mass adoption or do you think there's still some things that need to take place yeah I think it's it's absolutely ready for Mass adoption a lot of the deepend projects are inspired by helium uh it was one of the first and biggest and they've been going at it for a little bit longer they if you don't know what helium is anybody can contribute hotspots to a cellular network they started off with loroan and now they're doing 5G and you can actually you can pull out your phone right now if you have a new iPhone you can download helium mobile and have a second eim with a helium number and you're walking down the street you'll be using 5G helium hotspots there's gaps in the networks they use T-Mobile which is a cool thing that they got that partnership but I think even more interesting recently massive Telco in uh they're announced in Mexico state City in Waka telica that uh people can roam telica customers be walking down the street they're like swiping right swiping left uh using Uber whatever and they're passing their data over helium hotspots uh in in Mexico City so you're starting to see uh a deepen a blockchain deepen Network being used by people who don't even know what blockchain is or care about it and there's nothing technical or sci-fi about how you scale that it's just needs to mature more the the the you know the life cycle of their adoption it's just kind of pushing up that curve so there's nothing holding us back um a lot of deepin projects are very early they're still in the uhu a lot of deepin projects have found product Market fit in terms of people will do things for tokens but they need to find that other bit of it to really get that flywheel going of the demand on the network that they're building and so hels a little further along we're getting there and everyone else is trying to get there but it's uh there's nothing stopping it yeah yeah please yeah live live Pier was also one of the first deepend projects before it was called deepend the project launched in 2017 the token launched in in 2018 uh we've been playing a leading role in uh coordinating other web 3 infrastructure projects we're a founding member of the web 3 index which actually reports on usage um of of these projects and I think that's a really important differentiator when we talk about deepen and what makes it different from other parts of the crypto space this is really not about speculation it's not about like launching a token to you know um I don't know a lot of a lot of the point is a lot of token launches don't have much behind them but we're really you know launching utilities um and these tokens are used to incentivize people who are providing an actual internet service whether that's computer storage or bandwidth um and so I think when when we look at you know what's blocking um or what could block you know the the continued development of Deep deepen I do agree it it is it is demand um and I think there's just so much generic deepen coming online right now you know masari's put out a really great report I encourage people to read it it came out in January there's something like over 600 deep pen projects that have been launched in the last year who are you serving with that compute um and if you don't have a good understanding of the end consumer use cases for the infrastructure that you're providing you're probably not going to last very long and at live Pier we've been laser focused on video and video workflows and media workflows to understand where are video and media consumer applications going what are the infrastructure needs for that um we've recently announced that we we've launched an AI video subnet where we're adding um generative video processing capabilities to the network because we just know that AI is justup in video content creation and consumption and we're looking at what are the specific AI job types that we need to be adding to the network whether that's basic text to video also upscaling um scene classification all of these different um AI processes that are going to add value for video so if you if you're looking at deepin projects right now if you're an investor um I would be thinking really critically about what is the end user and and use cases that these projects are serving yeah whenever there's Buzz there's always charlatans that jump into the mix right so I hear you Mustafa I love to your thoughts yeah um I think I agree with everything that was said from a like tokenomics perspective I think we're still in the stage of like um I so to preface this I think dpin has the uh biggest potential for Mass adoption um that I see right now in web 3 um but before we get there I think what the phase we're in right now is figuring out how to use cryptoeconomics levers the proper marketing how to start flywheels Etc we're figuring that out um and it's there's a lot of research I mean I think there's going to be much more research um as the sector grows but I think that's kind of where we're at right now that's what I see right now and um I think it's really good I think it's better to you know test that now before we have a something that really needs to scale if we're going to actually have proper massed option right so I kind of want to talk a little bit about some of the advantages of deepin over like traditional kind of providers um what do you guys see as the primary advantages of deepin over kind of traditionalized systems like for me personally like at a Kos I think of edge compute and it's a topic that kind of became popular a few years ago and I think it's going to start kind of coming up again because if your next door neighbor has server capacity and you can get that from them latencies goes down to zero so it's just simple science guys um so I would love to hear your guys' perspectives on uh you know the primary advantages over of deepend over kind of traditional Services cool so we get to build demo in more of a Grassroots way we get to go directly to drivers onboard drivers the automakers aren't even kind of part of this equation yet if we had to go to if the only way to build demo was to sign individual Partnerships with every automaker insurance company car dealership like it would be absolutely impossible so the fact that you can go directly to users drivers that's excellent sort of similar in the same theme of that is the alignment there's this unfortunate dynamic in building traditional corporate networks right where company raises millions and millions from investors uh spends all of it on marketing and you know initial sign up bonuses or whatever you build this huge Network you have investors on one end who own it and control it and then you have the users on the other end who have no owner they they they've spent their credits they were swayed by some marketing but they're the ones who really make the network special they're the ones who care about it but the company and the the product is beholden to the investor and a lot of times they both want the product to be good but they really want to extract value from the users so you can build a more aligned Network when the to economics allow you to distribute the control to users stakeholders in our case automakers drivers that kind of thing so that's excellent and then also in the the in the theme of alignment the security and reliability guarantees you can get with blockchain are really special you know our Network too it would just would not work if we were this centralized company that could turn off API access at any point and pose a 30% Apple app tax kind of thing it needs to actually be infrastructure in order for automakers to trust it Banks to trust it insurance companies to trust it so again I mentioned earlier like we couldn't do this without blockchain like for all those reasons like without any one of those three things we demo could not exist and that's why it hasn't existed historically and why it will soon you know mature and exist in a big way for people very soon right yeah I I look at this from a really Global macro perspective a Global macroeconomic Perspective the five largest companies and the most profitable companies in the world are internet service providers Google Amazon apple and now Nvidia um there are very few people who benefit from the returns to the infrastructure that these companies have a have a monopoly over so that's why I entered the space and I was uh I I had a career doing a lot of other things before I entered crypto um and so that's ultimately what I think the the potential is here but other than being I think a more fair um Equitable and Democratic way of of doing business on the internet I think there's like three really practical um advantages that that deepend projects have over the centralized web 2 internet companies today um one is just the infinite scalability so if you're um if you're Google or or Nvidia even um you're relyant on a on a few um a few providers of that infrastructure there's just huge capex and there's huge startup cost to be able to provide the infrastructure needed to power um the appli you want to provide at the end of the day um with crypto networks you can um just bootstrap uh smaller providers in a Global Network um permissionless Le so anyone can can come into that Network and get rewarded um there's also the resiliency and the reliability in decentralized networks that you don't have with um one central point of failure and you know one or two large data centers that are powering Google um and and we see that in our networks if one node goes offline there's there's hundreds of others um that can come in and and take up those jobs um and I think that ultimately it's about cost efficiency as well you can provide software developer tooling that is you know 10 times cheaper than what you would be paying to you know Nvidia to use Cuda because it's the only software um that's going to accelerate their Hardware right and and just to add to that before I hand it over to Mustafa at aosh all product features updates go through Community governance and I can't think of a time that you know some other nameless providers of compute uh have gone to their community and say will you vote on this and see if we can actually do it can you please vote on this so going through the community is like a way it's a benefit right it's a value add to actually making our Network that much stronger um but yeah we' love to hear what your thoughts on our advantages of kind of um dpin over centralized Services Yeah we actually uh wrote a piece on this very topic um so we kind of outlined three things that we we thought were um for the generalized deepin model better than centralized first one is risk tolerances right so um you kind of mention this I think like when you have just a centralized entity they have a single risk tolerance that changes based on what stage they're in whereas when you have this Grassroots kind of thing where there's a bunch of different people with different risk tolerances then you can do a much more Dynamic Capital formation um that's just a better way to bootstrap a network um on top of that the second reason is uh that you can get access to Future cash flows like now um so the providers who are getting paid for their investment get access to that those cash flows as soon as their services are being used um so we think that's a a benefit and something that can actually you know incentivize more providers to get into the into the network and lastly your profits and this is where the token comes into play I think a lot of people wonder why do you need a token um we think you need a token because you need something that grows in value with the network so um your past profits will grow um even more as you know the network theoretically grows in value um and as more people join as your service gets better um it's just you know all of those things I think combine make like a perfect um incentive scheme for fast fly to to Kickstart flywheels and to get like good Dynamic service um out there right um so I've time for maybe a couple more quick questions here so I want to talk very briefly about a privacy and kind of data management on deepin networks um it's actually one of the things that I've seen kind of come up in conversations many times when we talk about things at AOS and I'm sure you guys have dealt with these things too how do you guys perceive kind of deepend solutions addressing concerns around data management privacy um and kind of also ensuring that you're we're making sure that there's usability and convenience for these users using deep in net netw works I I think we're privacy's Last Hope when it comes to technology being useful but also being privacy preserving so with demo we have uh we represent your vehicle with an nft that's its digital identity and it's programmable and importantly on that uh nft there's permissions where you can specify who can access the data for your car so you as the the owner of the nft are the only one who can set those permissions as to who can access the data the nft also points to where the data is being stored so right now we're the only host of vehicle data on the network but in the future just like email you can choose Google Microsoft you can store your own emails if you want right email's not encrypted you know we will be we are uh but you'll be able to choose where your data is stored be able to CH choose who can access it and you will not have to compromise rise on usability the for a developer building on the network you log in you connect your wallet the the vehicle ID points to where the data is being stored they use the the storage provider runs a common open source client that makes sure that whoever is requesting the data VI API has the right permissions for it has you know paid demo tokens for it and they can retrieve it so you can kind of imagine that allows you to build really rich applications all the benefits of normal connectivity but without any of the I just have to trust Google Apple Microsoft Amazon whatever with every secret about my life and oh there's data breach oops now I need credit monitoring oh oh God you know so we can be the solution for that very cool yeah I think data provenance and authenticity is really important um when we think about especially in in a new AI context so how do you prevent against deep fakes um how do you ensure that people who create content can be rewarded for that um you know as it's shared live per has been doing a lot lot of work around um verifiable video um and content authenticity and just like a couple of use cases on this so um there are a lot of independent journalists in the world who work really hard to put stories online a lot of news is now consumed on social media platforms that are using large language models to just scrape user data so you have uh the hard work of an independent journalist that's get getting shared over hundreds of platforms without any attribution um or ability to get rewarded for that cont content so um this is an area that like web 2 companies have um also been been working on but they haven't yet leverage public blockchain Solutions so we're really excited um at live Pier to contribute to that we are part of the content authenticity initiative and the c2p standard that's being led by Adobe and others um and I'm excited for um blockchain Solutions like atest stations Creator sign content um where we store those on public blockchains I think that'll be really important right yeah um I think that you know this is the best the best equipped or the most well equipped industry for privacy um it just requires projects being open- Source you know uh being transparent about their data um and like I like um you know I think it's possible that early on in a project you're going to have a centralized entity that's storing data but as long as you know there's open protocols there about how that's being stored um whether it's on chain or not um that's all really important so um I do think we're in a good place for that and like I don't think we've had any major problems with with that in specifically in dpin knock on wood yeah um cool so quick little lightning round we have about two minutes left before we have to go um this has been fantastic thank you guys for kind of doing this today looking ahead what are some of the emerging Trends or Innovations you guys anticipate in the world of deepend what does it look like in the future I'm excited to see more composability between projects uh you know we're an automotive one would love to see defi auto loans open SE style nft marketplaces people registering their car in the DMV that's going to become really magical for folks AI just AI yeah I you know there's so many different angles to this there have been great events this week on on the the intersection of aan crypto right now you know everything from um open sourcing of model train um and you know data that goes into that to you know what we're focused on at live Pier which is providing scalable compute needed for this growing demand um for AI applications on the internet I think the decentralization of that the open sourcing of that the cryptoeconomic incentives that can just I think really unlock a whole new wave of innovation um around the internet very cool snaps huh come on that's good I like that go ahead Mustafa um I'm pretty excited about AI obviously I think GPU that's like the most pressing need um but also we've seen some cool GPS stuff we've seen some cool energy stuff um I just think there's you know anywhere that you see this kind of infrastructure I think deepen can tackle it and make it better so um yeah very cool and my final question with the 30 seconds we have left how can folks help your projects you can you can connect now uh drivve demo.com to learn about uh the different devices you can plug in your car if you have a newer car two uh that has like a login you can log in with that in Deo mobile so just download that in the App Store and get started follow live Pier on Twitter in our Twitter bio you will find a link to join our Discord join our Discord and hop into our AI video channel where our community is just um going really hard on on this whole AI processing AI application world so jump in yeah and you can follow six-man Ventures follow our research um and we're investing if you're building cool thank you guys and if you want to check out AOS we have a booth thank you guys for joining this panel thank you all and enjoy your e Den we'll see you in the [Applause] metaverse all right thank you my deepin friends