2023 State of Crypto Report with Eddy Lazzarin, Daren Matsuoka, and Robert Hackett

Transcript

all right hi everybody thanks for joining and welcome to the a16z crypto live stream I'm Robert Hackett an editorial partner here at a16z crypto and today we're going to be doing a deep dive into our latest state of crypto report which published this morning this is our second search report and it's chock full of data highlights and takeaways about everything that's been going on in web 3. we're going to start that deep dive in just a moment after that presentation we'll also be showing off a new interactive tool we just launched the state of crypto index now this index combines a bunch of public data sources from across the industry it Blends them together and it spits out a picture of the industry's health it's really cool go play with it if you haven't already

now I know many of you are super curious about the findings in the report about all the nitty-gritty nuances and the methodology how the index Works what it means so let's get into it if you've got any questions comments or thoughts please share them by replying below shooting us a comment in the chat box wherever you are if you're on YouTube Twitter LinkedIn wherever just shoot us a note and while you're at it go ahead and subscribe to our a16z crypto YouTube channel there you can stay up to date with all of our latest talks videos and resources I would also encourage you to follow and engage with us on uh Twitter and forecaster at a16z crypto and while you're at it subscribe to our podcast web3 with a16z

we'll be airing an episode separate and different from this live stream about the report and the index there soon you can catch us having a conversation uh we'll be joined by Chris Dixon there who's the founder and Lead partner of a16z crypto so you can hear his thoughts there okay let's get on with it uh say hello to my hosts uh give them a wave a GM whatever you like I'm here with Darren matsuoka our lead data scientist and spear header of this report and index and we've also got Eddie lazarin the chief technology officer at a16z crypto both of whom spent many many hours putting this package together for you all right gentlemen thanks anands take it away you guys

sure so yeah I'm Eddie lazarin and uh I'm gonna take us away on the report for now so first uh just a general overview where's crypto today a lot has happened over the last year there's been some significant progress and some significant setbacks and like it is every year crypto is volatile and fascinating and the technology continues to evolve in new and exciting ways there's challenges ahead like on the regulatory front uh something that we've been investing a lot of time and attention into uh but we're very hopeful about particularly in the US and most importantly of all there's incredible

opportunity and the last couple years have made that even clearer than before our overall view of crypto and that's important to situate all the things that we present in the report our overall view on crypto is that it's more than a financial movement crypto creates a new kind of computer and it's an evolution of the internet we'll say a little bit more about that blockchains are more than ledgers their full programmability means that more sophisticated and complex things can be expressed in an environment where computation can be verified and this isn't very interesting consequences so why does web3 matter there have been three kinds of networks

that have unfolded over the history of the internet in different phases and these three phases are very distinct in terms of their Technical and economic properties particularly ownership control privacy verification of compute these properties radically change how those networks can be used so these properties have immediate economic consequences this changes take rates it changes the ability of users to exit and avoid rent collecting it affects what kinds of businesses can be built on different types of networks it changes the ability

to remix and recombine them and the composability of these systems so the network design has economic consequences Eddie I think this is an important point because a lot of people they hear that web3 has low take rates but they kind of think that maybe you know as soon as a platform gets big enough they'll just jack up the rents maybe you could just touch on yeah that's a great thing to assume it's what's really interesting is that examine why it's possible for a system to start to take higher take rates as it's as it accumulates a network effect I consider one of the social media networks or consider one of the application platforms what are the

fundamental reasons why they can increase their take rate or avoid including the users of the system in the economics at all it's because they can constrain people's ability to exit the valuable things it's because they can constrain access to the underlying code there's all kinds of ways that they can create extreme defensibility of their Market position but the way that web3 works is you can guarantee the ability of users to exit with a valuable thing you can guarantee interoperability you can design a system that remains unified but also open and can be something that people can build on taking for granted that it can't be evil because of the power of

decentralized consensus so that Network design has economic consequences for the people who are building on it for their users who use it and so on and so forth and we're seeing how those how the early experiments in these new network designs are playing out it's a really good point as Chris Dixon likes to say uh quoting the Google motto you know they say don't be evil but in web 3 it's can't be healed yeah that's right and that's because of consensus because of network design exactly so one of the uh one of the consequences of a network design as we've seen and we just discussed can be the extreme consolidation of Network Power

uh like in web 2 or the inability to fund development or other critical features like in web one crypto offers a third way to design networks that can resolve these issues and we're in the earliest days of understanding how those networks can be used but it is a genuinely New Path along many important dimensions and although crypto can provide Solutions and we're seeing examples of that it can also create a casino and it's because creating a casino only requires the free transmissibility of assets which is one of the most one of the simplest things that that a crypto Network can provide among many other

possible things so uh with that I'll I'll hand it to Darren for this next section thank you Eddie um as kind of Eddie was just talking about with the crypto Casino versus the crypto computer uh we really like to focus on the technical technological progress in in crypto uh but the reality is is that the market has both what we call Financial Cycles uh as well as product Cycles uh and in crypto we think these things actually reinforce each other and the end result is consistent uh long-term growth uh in in the space this isn't a New Concept it's actually something that our firm has spent a lot of time thinking about and writing about

in the past we call it the price Innovation cycle and the bottom line of the analysis is really that we believe there's a feedback loop between prices and Innovation uh in in crypto in fact we think that prices are often a leading indicator of that technological progress and the reason for that is that prices uh kind of naturally lead to more interest in this space that interest leads to new ideas those new ideas become startups and projects and those startups and projects build great stuff that contributes to the development of the industry and then ultimately an increase in in the prices that we see um and you know this might make sense in

theory but we can also really support this by looking at some of the data in more granular views of the industry metrics on this slide you can see the global crypto market cap compared to social media activity as measured by Twitter mentions developer activity as measured by GitHub stars and funding activity as measured by funding rounds from pitchbook you'll note that the x-axis here is all the same showing the growth since 2011. and it's it's maybe a bit hard to see in this particular view but if you look closely at the data you'll actually see that the developer activity the startup activity the social media activity that all happens and and

really becomes more active in the months shortly after the big price swings um and I think that speaks to this price Innovation cycle that we're we're talking about it's happened consistently across the four major cycles that we've seen in crypto since bitcoin's Inception um and again just really speaks to that uh that feedback loop that that we've mentioned um and then you know the result of course is that consistent long-term growth as measured by um you know the the very kind of positive uh Cagers we also wanted to really double down on the distinction between Financial cycles and product Cycles uh throughout history uh you know great products have been built in all Financial Market Cycles you

can see a consistent founding of the most important companies uh in Tech each and every year look at companies like Facebook and YouTube that were founded in the Years shortly after the.com crash as well as companies like uber WhatsApp and Bitcoin that were born during the global financial crisis remember Financial Cycles are highly volatile very hard to predict but product Cycles are much more consistent in terms of you know what that means for where the crypto industry is today of course the prices have fallen but great products have launched as Builders continue to to build and uh with that maybe I'll pass it back to Eddie to talk about some of this development that we've seen and

some Trends to watch into space yeah so if there are products I vehicles and financial Cycles two different forces at play that relate to each other in some way but ultimately unfold independently then what are the key things that we can track to watch the development of products over time although prices are the most spectacular thing there's actually much better indicators specific things that we can look at underneath the hood that show us how technological progress is compounding so this section we're going to go over some interesting things you can see almost all these are public data and things that you can find for yourself but we've put together for you so first

there are maybe one of the most important Grand themes to me and I can say a little bit more as to why is that blockchains are really going down the path of scaling and exploring many different new avenues we have new layer ones that have different fundamental architectures that try to scale radically through choosing different parts of the trade-off space and adopting new technologies and new research uh we have new we have convergent evolution in terms of what some might call you know application specific chains some some would call execution shards it's interesting to see the nomenclature and the space evolve but

I'll say in this top right corner category CK saying Cosmos optimism polygon polka dot and others have all adopted a sort of very high level similar approach where applications or groups of applications group together and perform computations together and then settle that shared State somewhere else and of course they have very different choices but this model is playing out and looking very very promising within that category we have optimistic Roll-Ups zero knowledge Roll-Ups which are two different very different approaches to scaling that have will show in the next slide actually have achieved significant adoption over the

last year and we have some fascinating developments in data availability which is a core premise core component of the way that blockchains uh guarantee different properties of that data and its availability so maybe you can zoom out for just a second and say why scaling is important at all yeah it's it's a great question this could be debated and I'm happy to be wrong and it's a really interesting topic to me but my my firm belief maybe in you could do an analog with like the early stages of the internet like dial up and Broadband right if you told someone that what you were going to get out of broadband was loading a specific website faster that probably would be nice but that wouldn't

really be that exciting or profound because the kinds of things that you'd anticipate are kind of buried they're a little opaque when you frame the question that way but if instead you say well if everybody gets broadband and fast internet we can have YouTubers and twitch streamers and MMOs and all these new types of applications you know uh the all the things that we've seen on the internet today then it becomes a little bit more clear how scaling one key Dimension or several key dimensions in computers many of those have scaled over the last two decades can allow for experimentation and Innovation new types of applications blockchains are no different just sending a token or transferring an asset

is a relatively simple computational task but being able to do it in a huge way where you're reaching a massive number of people or more subtly if you're mutating and doing complex complex things underneath the hood that can require significant scalability and my genuine belief is that if we radically scale the throughput of blockchains and improve the user experience and a variety of other things that all relate to the Technologies maturing then we'll create the opportunities for waste and experimentation that people need in order to be able to discover what the native new applications are and we're starting to get the signs of that since blockchains are just starting to be able

to scale so thanks for thanks for that Robert here so I alluded to this slide this is a really cool one of my one of mine Darren's favorites we talk about this one a lot we've been we've been tracking it for some time this is the percentage of block space that is being consumed on L1 by all the l2s so of course L1 like on on ethereum ethereum has a specific gas budget that's the amount of compute you can think of it or storage it's like kind of a mixture of a few things but it's uh the amount that can happen per block and this is the percentage on average of those blocks that are consumed by what's happening on l2s this is going way up there was a recent pop you can see uh

promised we didn't time we've been working on the report for months so it couldn't have anticipated couldn't have anticipated this but you can see uh ZK sync uh era recently launched which is radically increased uh this measure this shows that more and more activity is happening on l2s and l2s are finding it worthwhile to continue to park more and more of that data on L1 this is about as concrete evidence as you can have that l2s are starting to scale right and by the way we shared this not this same the same exact uh chart last year in the 2022 state of crypto report and the current uh percentage of ethereum fees paid by L2 Roll-Ups was

1.5 so it's since more than quadrupled and this is actually a number that we think can go even higher and so it'll be interesting to see what uh what this number looks like in in next year's report and and even in the past year it's pretty amazing we we actually have a question from the audience too related to l2s um Stefano Andres on Twitter uh he said Layer Two blockchains have seen significant growth in terms of fees on ethereum how do you foresee Layer Two Solutions evolving in the next few years and what potential challenges could they face in terms of widespread adoption that's a great that's a great question uh in terms of obvious developments I mean the first one I think is on the top

of everybody's mind on the tip of everyone's tongue is uh proto-dink sharding EIP 4844 which could as a result of creating a separate little pocket of data specifically used for Roll-Ups and things like that could radically decrease transaction fees even further it's a really nice economical use of space and this is a proposed ethereum software upgrade just for yeah exactly exactly right I think that that could make a huge difference but maybe more importantly is going to be the story of interoperability if we have a bunch of l2s and uh interconnecting l2s the story of how they interconnect and the types of

things that they they share like their shared State their shared Bridges or sequencers those are kind of the prerequisites of uh of uh interoperating them uh that story hasn't been told yet they're just starting to work it's just starting to become very easy to deploy in L2 we recently saw coinbase launch the base uh L2 on ethereum which is really exciting uh but uh the story of how they interoperate has not been told we're just getting over the first stages of scaling and I think that there will be a lot to learn in terms of the way they interoperate with ethereum the way they interoperate with each other and other blockchains there's also the story of the user experience just how is it

that a user should reason about where their assets are and where their objects are and where they're you know where the games they're playing are all those different types of things at first glance it is a little bit challenging then again we've seen in the rest of computing and the internet people have found ways to abstract the underlying infrastructure that people are using in a way that maintains its usability and allows people to retain choice but still uh you know still not worry about what's going on underneath the hood so we'll have to see we'll have to see there thanks for that question awesome so so speaking of infrastructure

we've been talking a lot about that uh since we released the last state of crypto report we saw ethereum uh successfully complete its transition to proof of stake which is uh super exciting for for all of us I think it will go down as one of the most significant upgrades in the history of Open Source software development and the result as you can see from the slide here was a 99.9 reduction in energy consumption uh you know Eddie and I were just talking about this the other day it does seem like most of the environmental objections to using ethereum have rightfully gone away and I think that's fantastic for for the industry um but in addition to the kind of

massive reduction in energy consumption uh proof of stake also brings us a couple of other benefits one is more efficient security so one dollar of the ethereum security budget now goes about 20 times as far it's also more fair and decentralized in the sense that you no longer need specialized Hardware or technical skills to to set up a validator it's easy to to stake and then lastly we also now get this property where finality is no longer probabilistic which is actually a very important property for many of the scaling solutions that we're building today and so in addition to the uh in addition to the the energy consumption

there's also a lot of really great benefits that proof of State stake brings us and you know we're very excited to see that that transition happened successfully Darren you described it as one of the most significant events in open source history just maybe you could elaborate on that like why why was it so significant why was it so momentous well part of the reason was that they were kind of improving the plane while it was in Flight right there was significant kind of real world activity that was happening on ethereum and they had to not only kind of make this upgrade but do it without breaking all of the existing applications and tools and products that were built on top of it

and so it was a very and in a kind of fully open source transparent way right and so I think a remarkable achievement for the ethereum community that they should be incredibly proud of totally totally agree it was a momentous day I'll never forget it it was totally uneventful and uh it's almost outrageous to me how it was it has disappeared into memory just a few short months ago which is great right because that means it was totally seamless I bet some people don't even know what happened no news is good news right yeah totally yeah totally sweet

um I guess if we go back one slide if we want to further contextualize the energy consumption of proof of stake ethereum just because I think this is something that a lot of people are interested in we can compare it to other well-known uh products and industries for example ethereum now consumes uh .001 percentage of the energy that is consumed by YouTube and you can see just how small that is in comparison um of course I I will kind of make the the caveats here that there are nuances when it comes to making these comparisons for example you know the energy consumed from watching a video on YouTube might have otherwise been sent spent on getting in a car and driving to

the movie theater the world of course is very Dynamic but I think critically here we have removed what was really by definition wasteful energy consumption from proof of work ethereum which I think is a big win for the crypto industry I also want to make this point that uh you know a lot of people think staking is just about earning money and there's some apy that you can get from staking your if but of course it's so much more than that for ethereum it's used to secure a 400 billion dollar economy of eth tokens and nfts um you know one metric that we like to

track is the ratio of total value uh secured to total value staked which today is about 13 x um you know it's not wrong to say that a lower ratio implies more security but to really fully understand the implications of that you need to analyze several different factors this is something that we've talked about internally with our research team and I'm happy to share more thoughts on that if people are interested or have questions but uh in the interest of time maybe we can can move on yeah here's two two of uh our favorites I know we talk a lot about uh zero knowledge zero knowledge has been

around for a long time it has a deep academic history I recommend you look at our zero knowledge Canon on the a16c crypto site it's fantastic but uh blockchains have given it new life and that's because there's super interesting things you can do with it in very in brief uh ZK allows for scaling because it creates very succinct representations of facts and it can be used for privacy because you can very finely control what kinds of things you share and proving specific facts that we're dealing with underneath them so what things that are unpacking recently will show more in the next

slide Hardware costs are radically decreasing for zero knowledge because the economic demand is materializing tons of cool courses there's the ZK learning online mooc it's amazing some of the members of our research team participate in that super incredible resource and high level programming like programming language is that take advantage of them are starting to mature and so we're seeing the space unfold of many of you who are deep in in crypto I've heard a lot about zke VMS and there's some interesting and contentious discussion between them those are signs of growth those are signs of advancement in the underlying Tech so it's very exciting to see one use case that we didn't even go into

on this slide uh one of our uh Partners at the firm Elena Berger just wrote a great piece about how zero knowledge can be used to apply checks and balances on artificial intelligence and machine learning um AI is obviously a super hot topic right now um maybe Eddie you could just provide a little bit more information on that yeah yeah just uh maybe a very brief piece Atlantis piece goes into detail on this that's totally worth worth reading I'll say that in short if you're running an AI model uh besides trying to reason about whether you're dealing with a human you probably want to be able to reason about what model you're using and uh what you know are you using the right model are the

inputs to the model something that you can control and keep private like there's so many fascinating questions and we will probably require verifiable computation so the ability to know whether an inference or output of a model is what you expected and what you requested or whether or not a model was trained on data or in a way that respects the underlying privacy of data creators or data providers or or the people who have trained the model or put data somewhere on the internet verifiable computation is going to be a big part of the story of AI we could probably spend a whole podcast talking about at some point Robert so I'm kind of glossing over it but it's a it's a fascinating one and but to do that of

course we will need zero knowledge to scale and uh part of there's a there's a whole bunch of things that are part of the scaling story one is academic Publications right advancing the actual science which have been accelerating uh in recent years partly as a result of blockchain's revitalizing interest in the field uh GitHub stars are increasing at a rapid clip more ZK proofs are being verified on ethereum this kind of plays into the slide we were showing earlier about the consumption of L to block space and and critically in the bottom right corner we can show that the keys are the technical parameters the performance of zero knowledge systems are also rapidly

scaling prover time proof size and verifier time and I'll say a little bit very briefly what each of these are prover time is the amount of time it takes to generate a zero knowledge proof there is overhead in the process of generating this and that is coming down the size of the proof is also a key parameter because if you can prove very very very complicated large things in a very small amount of space that's ratio is how you decide it's one of the key pieces in deciding how much it can scale and the last is verifier of time how much time it takes once you receive that proof to know that it reflects what it claims to reflect right so because these

are being used in production because these industries are maturing they're scaling incredibly rapidly and I think it's uh incredibly exciting so this slide mentions uh Moore's Law which is the idea that um you know performance of semiconductor chips doubles every roughly two years um it's interesting to see such a rapid Pace here in zero knowledge is that a trend that we can expect to continue into the future what are the smartest brains kind of thinking about how that's yeah that's a great question and I'm weary of prognosticating about it right on the one hand uh you know the past is uh the past is pretty is Preamble right and that this is this is

how stories unfold in technology is rapid compounding advancements so that feels like a pretty solid empirical ground to take on the other hand you never know different very complex Fields often advance in fits and starts so it's hard to predict but there is an interest although the Moore's Law comparison is certainly it's a it's a little bit of a metaphor I think it's an interesting particularly interesting metaphor because part of the process of Designing zero knowledge systems is in designing what are called zero knowledge circuits and circuits are a way that we create a representation of a program in zero knowledge using what looks like a little computer but it's

appearance soft it's purely software and human beings have been very good at advancing and scaling computer programs despite how hard it is to do so with Hardware so I think that if the space that we're optimizing is just software space I'm bullish so by the way we've mentioned a bunch of pieces related to zero knowledge this is an area we track quite closely so if you're interested in getting some of those resources from us you can sign up for a newsletter web3 weekly it is on sub stack check it out subscribe share with your friends yeah awesome now now we're going to shift a little bit here and talk about uh nfts and more specifically their role in the

Creator economy uh in the last two years nft marketplaces have paid out nearly two billion dollars worth of royalties uh in secondary sales to creators uh note that this doesn't include any Revenue that's generated from primary sales which is actually a much higher number uh but even the two billion dollars is not insignificant when you consider that Facebook earmarked only a billion dollars for creators over multiple years despite having 3.6 billion users on their platforms compared to the tens of millions in in web3 now with all of that said you can see the chart has gone down uh and Creator royalties have been a topic of heated debate over the last few months and the

core issue around that stems from the lack of a viable on-chain enforcement mechanism and because of that the industry is actively exploring alternative Solutions this is also an area that we've spent a lot of time thinking about internally as a team so if there are more questions that people have around that happy to dive into dive into it later we also think it's worth mentioning uh that many of the world's biggest brands are exploring web3 uh with nfts being the most popular entry point uh some recent examples include Brands like Starbucks which started in nft based loyalty program Reddit which onboarded

millions of users through their collectible Avatar nfts and most recently Gucci which launched a partnership with Hugo labs's metaverse project called Otherside which is a cool experience that I'd encourage people to to check out and it's very obvious when you just look at this slide here that there is a huge opportunity to onboard new users into crypto and so we're very excited to continue watching this and see what else is uh to come hey Darren on the previous slide about nft royalties I think there's an important point to be made about take rates and about marketplaces that uh let you trade nfts you know one of the interesting points uh just on the slide

immediately prior to this one is um you know when you can genuinely own a digital asset a digital item object whatever you can go to whatever platform you want and sell it there and so it creates this intense competition among marketplaces uh I think it kind of goes to show about the uh the structurally low take rates in web 3. it's a really good example of that that that's exactly right and it goes back to what Eddie was saying at the beginning of this presentation around how the network design has Economic Consequences all the stuff about having the right to exit I can take my nft like openc doesn't charge a low take rate because they're nice right they charge a

low take rate because I can take the assets that are mine that I rightfully own to another platform uh if if I want to right and so there is that increased competition lower pricing power which ultimately leads to leads to lower take rates and better outcomes at the end of the day for for end users so definitely an important point that I think we want to to hit hard is this is this a solvable problem by the way I feel like people are probably interested to know that yeah I think it is I think it is and it may not have to do with transfer royalties on transfer as we have today yeah royalties on transfer are the first way that we've really experimented that has some product Market fit and started

to grow you know this industry in the form we recognize today really only started to take shape you know late 2021 all right mid-2021 it's incredibly new there has been relatively little uh opportunity for allowing each kind of business model and each opportunity to unfold I think it will be solvable and I think we'll reach an interesting equilibrium where the underlying truth that the relationship between creators and their fans have has have has been disintermediated right in in the good sense like they have a direct relationship now and uh the consequences of that are playing out I think this will be solved but we'll

we'll see what form it takes it may not be transfer on royalty on transfer it might be something altogether new but the design space is very big a little ray of sunshine for our creators out there yeah it's just a matter of time awesome so so we were just talking about like the onboarding new users into crypto and I think uh games could end up playing a very important role in that regard to the to the web 3 Vision uh 68 billion dollars was spent on in-game purchases last year so clearly there is a huge audience to to build for um our crypto fund actually in partnership with our games fund has invested in a large number of very high

quality games teams um and the thing about about games especially the kind of high quality games is that the the good ones take a really long time to build but we know from uh our kind of recent an activity that the product pipeline here is extremely strong so we're excited to continue to watch the the space um as these new games launch in the months and years to come one thing I always like to say is that people who earn crypto playing games are going to naturally be drawn to things like defy when they you know look to generate yield or exchange their tokens or borrow um it's just a natural kind of transition for them um so again it speaks to that uh ability

for games to to onboard new users into the space and we haven't even talked yet about kind of open worlds which may initially look like games but can ultimately lead to these metaverse type experiences which are you know even bigger opportunities so a lot to be excited about when we think about web 3 gaming and the potential uh the potential a quote from Chris uh Dixon comes to mind here you know he likes to say that the next big thing starts out looking like a toy and uh what looks more like a toy than a game so exactly keep in mind that's exactly right um web3 has also been a great place to experiment with new forms of governance

and we've actually seen a lot of really cool Innovation some good examples are things like nounsdao which has created a model to generate recurring Revenue with Community Driven funding of new projects we're also seeing protocols like Lido experiment with new Concepts like dual token governance which adds more checks and balances to their governance system and continued progress in how Dao's interface with the real world through Banking and legal entities which I think is going to unlock uh even more opportunity to to Really build out the the Dao space and and I I think for this area in particular I think we're still

super early right I think we're running the early experiments of uh governance but I think there is so much more to come here and is again an area I'm really personally excited about yeah so and programmable governance is a huge design space we have yet to see all the experiments and we require a lot more ux and a lot more Tech before we can explore the whole thing and just to look at some data here we we have seen really positive momentum when it comes to participating in Dow governance uh 13 million votes casted across nearly 2 million unique uh voters and 78 000 unique proposals um and when you look at this data over

time by the way this is looking at voting activity on Snapshot which is a popular platform for for Dows it's been growing even in the the bear Market um I think there's something to be said about the user experience of not having to pay for gas on each transaction which makes it easier for people to actively participate but really cool to see that uh you know active participation is is growing and I think there's some cool numbers here to to unpack um but I'm now going to pass it back to Eddie who can talk about uh one of the most important pieces of our report which is the trends that we're seeing in the US and how the Regulatory and policy

and government will play a critical role in shaping the future of these Technologies yeah that's right I mean one data point not a not a happy one for me is that the proportion of crypto developers in the US seems to be seems to be shrinking and that's not because the number of crypto developers in the US overall has shrunk meaningfully it's just a change in the pace of growth in the US versus non-us sources I'd love to be able to get underneath the hood and to be able to prognosticate about exactly why developers may be preferring uh to move out of the U.S or you know maybe it's just more appealing to developers outside the US

but uh I think it's it's fair to say that there's a lot more that the U.S could do and that's in particular relating to uh policy and regulation of crypto which is going to be a critical in shaping the future of crypto in the US just a couple quick points and these come from our policy team that spends a lot of time thinking about these things Banning new business models and Technologies is not uh is not the American way of handling technology as far as I know uh and probably something that we need to uh reconsider if that's really the path that we're headed down clear regulation and clear rules protect consumers there's a reason why those

rules are are in place and have been success successful in fostering American economic growth in other Industries over the last decades we probably need that here but it can't be regulation by enforcement and uh opaque to would-be builders who have to select and where they're going to do their work uh crypto businesses that aren't breaking the law need access to financial services and uh you know businesses are reasonable entities to regulate that's that's the way things work but protocols and decentralized systems that don't actually have a single Source or a single party controlling them uh

probably aren't I mean that just understanding that requires understanding the technology to some degree and uh I'm hopeful that that'll that that'll unfold there's all kinds of fascinating ways that regulation is is being debated today and we highlight here some key bills to watch and some key cases that are that are unfolding super interesting stuff not my specialty but uh but nevertheless incredibly uh consequential uh for the US thankfully there are some early signs that there is bipartisan momentum as our legislators learn more uh and get deeper into crypto and kind of understand the full consequences of this type of

Technology so I'm uh I'm very hopeful and uh with that we're going to get into back into some key Market Market metrics so I hope you enjoyed the trends to watch I'll say a little bit to link us back to where we started if there are Financial cycles and product Cycles or if there are if there is innovation in the underlying technology and there is adoption of that technology we should be able to identify key indicators that show us how how far along we are in terms of their development and those thanks to the open source and public nature of crypto uh you know open

source code open data all these fantastic places we can look we can probably find those indicators and they can probably be more precise more interesting and more insightful than just looking at prices or Market capitalizations of tokens or things like that so I'm going to hand it to to Darren to walk us through some of these indicators thank you so starting with the supply side of the market which we call Innovation there's a number of kpis that we can look at and analyze over over a period of time and we've identified seven metrics here that we are choosing to to track and show uh since the since the beginning of 2016. so I'm going to

step through each of these but I'm going to do so pretty quickly in the interest of of time all of this data is kind of in the report it's also part of uh the state of crypto index which we'll talk about after this uh but let's let's maybe just jump through quickly the the different slides active developers so this is the number of unique GitHub users who have committed to or forked a public crypto repository during the month I think a cool kind of long-term upward Trend that we're seeing here bull markets bring in the developers and those develop there there are signs that those developers are sticking around which is really cool to see and speaks to that kind of price Innovation cycle that we talked about in

the beginning of the report um we also can look at a larger set of developers including not only uh unique GitHub users who have uh committed to or forked the crypto public repository but also those who have starred which is a little bit less of an active engagement but represents kind of more interest in the space and you know we're seeing now more than 50 000 developers are interacting with these GitHub repositories monthly and uh you know that that number has grown over the years which is is cool to see contracted players so this is the number of unique addresses deploying smart contracts on the different blockchains that we uh that we track and you can see

a really rapid growth in the number of unique deployers I think that can be attributed to some of the rise in nft activity that we saw note that nfts are each a unique contract and so the uh the act of launching an nft means deploying the contract we've also seen much better tooling emerge that has made this process a lot easier so cool to to see uh continued growth there we can also look at something called the verified smart contracts which is the kind of process of verifying the code that you deployed uh through a block Explorer which is uh in many ways a leading indicator of kind of the

pipeline of of product launches so really cool to to see that continue to grow even in the bear Market developers are definitely continuing to stay active um another kind of lens that we can look at this through is developer Library downloads specifically web3 and ethers which are tools to help crypto users uh interact with smart contracts that are being deployed through a front end and this chart is pretty much up and to the right showing kind of steady growth in the dev community so an interesting one we've built a kind of pipeline I did a pipeline internally to fetch data from dblp and kind of tag

metadata based on keywords and it's very clear that crypto now has become a significant area of academic research we have a research team internally that is uh helping kind of contribute on on that front but cool to to see in the data and track going forward yeah very very cool one unique I think you can also look at job search interests so this is just looking at Google Trends we have kind of a a special process for how we analyze that data but you can see that when crypto prices go up it definitely generates interest in crypto related jobs and it's cool to see that there's now kind of a legitimate industry for full-time employment in the crypto space and that

uh you know that continues to show very positive signs and just see the big picture on some of these slides um and comparing it to the slide way back when when we talked about the price Innovation cycle and you just look at the shape of these charts and you can see which areas are a little bit more cyclical a little bit more affected by market sentiment compared to you know a lot of the things related to building activity are kind of staying elevated or higher um it's really interesting to see that from a kind of macro perspective yeah I mean what like I feel like this slide is pretty representative I mean there's a couple interesting sort of archetypical

patterns here but it's clear that when crypto is at its most hot Job searches are going to go to their hottest right people think people are searching for opportunity it makes a lot of sense and the more you're hearing about it the more sort of social media waves that are made by crypto projects like the more interest people might have the more possibility there is to find a really cool job in crypto and even despite dropping we can see that where we've kind of started to level out even on Job searches and this pattern is true in any other slides we're way above where we were before right we're even we're near where we were at the prior Peak if that zigzag pattern holds then I

think that that's a sign that there are some under there's underlying maturity forming in the industry and that's kind of a key pattern that I hope that we we can pull out that's a great totally moving to the demand side of the market now we can look at what we call the adoption indicators and similar to kind of the supply side I'll step through each of them individually the first is active addresses so this is the number of unique sending addresses across all of the blockchains that we track and this is uh really cool to see that has that it has continued to grow despite kind of the volatility in the market and in the prices like we've talked about uh

you know we are finding new ways to engage with uh with blockchains and and with web3 products which is uh you know resulting in a growing number of active addresses um on a kind of similar Dimension we can look at transactions which is the number of successful transactions that have been kind of completed during the month and this has really scaled largely as a result of just more block space becoming available through alternative layer ones and layer twos that have uh reduced transaction fees for users which kind of unlocks a new wave of potential for you know transacting on chain and you can see kind of the rapid growth there we can also look at transaction fees

paid which this which which is really just the overall amount of money that people are paying for Block space across the the different blockchains now this one's a little bit nuanced because of course right as uh you know demand Rises people are going to naturally pay more for that block space but there's also kind of the supply side of that market which is you know as scaling Tech comes online as we now start to get layer twos and other layer ones that reduce the transaction fees you're going to get more block space kind of resulting in less overall kind of net transaction fees paid and so this one is kind of a interesting uh metric to to unpack but yeah I think it's still important to track yeah I I really hope we get some

thoughts on this one because uh it it's transaction fees are a key part of the story right people talk about and when crypto was in its uh you know most chaotic phases like uh like a little bit over a year ago we saw Sky High transaction fees but what should the transaction fees be in a healthy system right zero could be easily conflated between incredible scaling you know massive you know massive uh block space Supply or on the other hand total lack of Interest right and just a total lack of demand and vice versa very high transaction fees can be a very negative indicator because the marginal cost of using the

systems is too high that's obviously bad it's exclusive it makes it uh difficult to use but on the other hand High transaction fees mean that users are willing to pay and that their estimation of what makes activity worthwhile is high it's like a good analog might be like electricity right what's the ideal electrical cost extremely high seems bad if you're a consumer but it also means that the system is healthy and the economy has an incredible demand for energy which is probably a positive because it means that energy is being used on the other hand if energy costs are extremely low it could mean that we're in an economic slump everyone knows gas prices like petroleum gas

prices go down when the economy is in a recession so it's a similar pattern but overall I think that it's an important one to show and we're interested in seeing how people if people agree with our interpretation now if we think back to the earlier slide on l2s that sort of kicked off the key trends section where you have this rocketing up of usage you know seven percent of uh of fees paid on ethereum are paid through l2s yep um given that uptick that upsurge of uh people using l2s and scaling Tech does it make sense to keep putting this slide in the index like in the future is this something that we you think we should maintain given like the weird nuances around it

it's a great it's a great question I think it would be a little short-sighted to leave it out of the story altogether right because I think it's really important in showing like if you looked at a graph of like total Global energy production in joules and also total costs of energy globally in dollars right that the closest analog in crypto would be like total block space in terms of bytes or some unit of compute proxy right a little bit difficult to measure and then total transaction fees paid I think in the healthiest possible World you'd want to see the total amount of fees paid growing slowly but the total amount of space growing faster right because you want there to be a lot of

space but you want it to be economically worthwhile every time someone uses it and then to demonstrate that with a willingness to pay it's a really nuanced one but it's one that I think should be out there and it's probably more important to have top of Mind than for example mere prices so leave it at that moving on uh Mobile Wallet users so this is something that we also like to to to look at and the reality is since the the early 2022 that number has been declining of course you know I think this this tracks uh Market sentiment um I think there's also some interesting stories to kind of further analyze or unpack here especially when you contrast

it to the number of active addresses growing uh I think one you know we still need to kind of find better user experiences when it comes to kind of using mobile wallets in particular I think there are a lot of great companies trying to solve these problems but they are far from from solved I also think there is kind of this aspect of products that are creating wallets for users that are native to the application so kind of behind the scenes creation of a wallet that allows a user to kind of interact on chain but doesn't necessarily require them to bring their own wallet or more specifically their Mobile Wallet um which could be kind of explaining some of the differences here but this is

a number that we would like to to see kind of rebound and we'll be kind of tracking it going forward because uh of course uh we want uh more mobile experiences uh in in crypto it is such a weird disparity between the active uh addresses slide that you showed a little bit earlier you would think that they'd be more closely meshed I mean it makes you wonder like are people mostly transacting on desktop computers like what yeah it's a really interesting question I I maybe Darren has some thoughts but we'd have to go one click deeper in the analysis and probably Source some deeper data it's one I've thought about for a while you know I admit personally my crypto use does not use a mobile app like mine is on on my laptop uh and mine

too actually yeah yeah but and and that wasn't always true actually that has been a change for me just because I have grown to appreciate the desktop experience but obviously we're not necessarily representative so I'd be really interested in learning yeah that's exactly what I was going to say you know we've done some preliminary analysis here but far from kind of the complete picture that we would want to do and like I think we can add this to our list of things to follow up on and if anybody wants to contribute to this type of work we're more than happy to collaborate so let us know please um Dex volume so this is the total on chain volume on decentralized exchanges um you know some some really positive

signals to to look at here uh for one it's you know now trading over 100 billion dollars monthly uh We've grown for three consecutive months in a row uh following kind of the the big drop uh as a result of the the bear Market uh uniswap the most popular decentralized exchange for the last two months have surpassed uh coinbase in terms of uh trading volume uh so I I think uh really kind of seeing some signs of of stabilization here I think more organic use cases less speculative um a lot to a lot to be positive about when when we look at numbers like this and and honestly a very similar story

for nft buyers um you know clearly there was a lot of speculation that flooded the market uh during the last nft wave um I do think a lot of those speculators have exited and we're starting to see the more organic uses of nfts emerge and similar to kind of the data in the Dex volume uh we're seeing you know the last two months uh really kind of uptick in the number of unique buyers that are engaging uh in in nfts yeah I hope we have the chance to go a little deeper on this one but it's interesting because nft volumes as a total all are down in this period but buyers are down much

less and in fact I'd argue they've stabilized at a pretty high level the total number of nft buyers given the maturity of the industry and how new it is I think we've begun to see signs but these are more sort of anecdotal and you know in our in our view as as investors is it seems like there's more non-speculative use of nfts more like patronage style uses lower value nfts that are less for constant flipping and more for direct relationships and uh you know collaboration and support with artists and creators right I think that that's a little bit of a pattern we're seeing

more of but we don't quite yet have uh super clean data to share on it yet super interesting one and then lastly we have stablecoin volume um you know I think this one is an interesting one to to look at uh it's been you know growing uh even in the the bear Market I think uh it's very clear that the demand for stable coins has remained very high uh you know there are lots of kind of considerations and uh you know things to think about when it comes to uh analyzing stable coins and kind of the the potential future for them in their role in in web 3 but it's very clear that the demand for stable coins as a product is is very high and I

think we'll see continued growth in usage on chain uh going into the future just a reminder for people so make your questions if you've got any questions comments thoughts feedback Etc um and we'll answer anything that you send our way yeah so just a last couple things we so these slides these indicators we've shown on the Innovation side on the demand side we are making them visible uh and we'll update them regularly on our website this uh state of crypto index you can go to our site now to see it we what we do is we collect all the metrics that you saw on the Innovation

side all the metrics on the adoption side and we aggregate them into a graph that I will show you here you'll see a page just like this on the site and it shows the accumulation of the Innovation indicators in the bottom left that's the kind of reddish graph and the accumulation of the adoption indicators in the bottom right and the total of both at the top in blue and I think pretty clearly here this is kind of what I hope you take away these look like two different patterns these are clearly two different cycles and the underlying order is different and you know if you if you disagree with how we've added them up you'll see more about our methodology on the site uh you

can go and you can change the weights yourself because we're not trying to be totalizing and objective and claimed this is like some scientific index that shows exactly you know a perfect tiny synoptic view of what's happening in crypto we just set weights that we think make a lot of sense but you can tweak them yourself you can zero them out you can set them to 100 whatever you like and that's because each of these numbers are based on public data and uh it's a little bit subjective which matters but I think you'll agree with us as you go through that there are clearly different indicators of different types of patterns unfolding as crypto develops so go there to uh state of crypto

a16crypto.com to check it out and we have just a couple more little things to show what's next right what's next in crypto so a little bit of prognostication it's still really early but the total number of Internet users and uh is the total number of active addresses continues to track just behind uh internet users in uh in the early development of the internet right who knows if this pattern will continue to hold I think it's a little difficult to say of course just merely being an internet user is in a way a lower hurdle than actually actively transacting and being a uh you know sending a

transaction on a blockchain today there's a lot of ux hurdles and other types of reasons but so far the growth pattern has been similar in some ways and uh note that we're tracking this across a variety of blockchains but we are deduplicating uh addresses that are shared across networks so hopefully the minimum amount of double counting Eddie there's a there's a big range on here you know it says 20 to 120 million why such a gigantic range yeah so it's a great point I wish we could be more precise but you know as a result of the pseudonymous nature of crypto right where you just have public addresses public keys and not you know a name or a household like you could with internet

use or with bank account use or something like that we have to be approximate it's certainly possible that people use many different addresses I mean I know I have many for my things and people people do that in those cases of course I don't transact on all of them every month or something like that but the uh you know it's a little bit hard to do that mapping so this is our best estimate of that range and it's why because it reflects our our imprecision all right but what we're trying to highlight is the trend even if it's uh there's little reason to believe that the proportion of addresses per user is radically changing it's probably shifting but the proportion of users per address is probably within a moderately

bounded range on average which means we can probably reasonably infer the trend uh even if we can't exactly infer how many users there are hope that answers the question yes yeah so uh so then just some quick thoughts I'm not going to enumerate all of these you can take a look yourself uh if you see the report there's a lot of cool things that we think are unfolding in crypto some of them we talk about on the podcast some of them we tweet about so please let us know your thoughts let us know what things you expect to come next as a result of all the compounding technological progress in crypto uh and uh I'll leave it at that there's a disclosure slide

[Laughter] how should we uh how should we rap Robert um yeah is it possible for us to show off the index like live I know we had some slides in there of the index itself but yeah I think so pretty cool to throw that up for people to take a look at maybe manipulate some parameters yeah let me let me pull it out great and by the way this uh Trend slide of what we're expecting uh which was just up there um all those different areas are things that we're paying very close attention to um we're discussing them constantly on our podcast uh web3 with a16z which you can subscribe to wherever you get your

podcasts uh you can also follow along with us uh our newsletter comes out um usually once a week it's called web3 weekly and it's on substack right could you bring the uh my tab on stage there you go boom so here's the here's the state of crypto uh live on our site and uh you can see here what I was pointing out earlier which is the overall aggregated Index right here and then the broken out adoption indicators and Innovation indicators here on the right side you can see all of the indicators that sum together to make the total it's worth saying maybe very explicitly that what the index

shows is the cumulative net growth for the that set of indicators so a 200 here would be a net cumulative 3x growth from when these specific indicators at this point in time began being calculated if you disagree with how you we've calculated them maybe you know maybe you think contract deploys that's worth nothing forgets forget smart contract deploys instead you know I think like Library downloads those those show people who are working on the weekends that's important active developers that's really important I'm going to

double that to 30 and there you go that like changes kind of the the shape of the Innovation indicators curve you know similarly for adoption indicators maybe like Dex wait forget about that you know it really matters to me it's nft buyers right that's that's everything that's like the world is made of nfts so like there there you go that's like that's some other way of doing it you can zero them out so again not trying to be an objective scientific measure but trying to show how specific metrics on from open data sources can be brought together to show a view about how technology is progressing in crypto relating to adoption relating to innovation

and Eddie maybe you can also just scroll down to show people that we can also see the underlying metrics and you can click into each of these underlying metrics to actually get a kind of more detailed view which shows the like the description of exactly what is being calculated uh the source um and again this this data will be updated uh monthly going forward so something that we can continue to to track together so we've talked about the price Innovation cycle moving through multiple waves uh we've said that we're probably in wave four right now um and on this index however we start from the year 2016 which is well after Bitcoin uh launched uh in 2009 so why

why start the index at that date in 2016 rather than earlier yeah that's a great question happy to have you win Darren the the key point is that uh it really took until ethereum before we started to see uh a variety of these specific key indicators even materialize obviously you know contract deployers verified smart contracts and you know key developer libraries all of those require there to be something for a fully programmable smart contract platform like ethereum to launch so in order for the index to be really robust and Broad and to draw from a wide

variety of sources so that it's kind of Representative of over the overall form of each size of the space we decided to start the index at that period and you'll see that's actually when if you zoom out that's when host activity starts the the amount of activity back from the 2016 2015 and prior period uh is pretty small there's just not a lot happening in crypto really the vast vast vast majority of development of all types in crypto begun after that period makes you wonder what wave five is going to be like um Eddie maybe you could uh maybe you could refresh this View and just show

the uh the original indicators versus Innovation indicators versus adoption indicators and uh maybe just provide a a word for us on the difference between the way those charts operate uh in what they're displaying like what's that yeah of what we're seeing yeah in totality it's it's clear to me that adoption is more volatile more cyclical more linked to the financial cycle it can come and go if this adoption indicator chart went down over the next year I wouldn't be surprised no I'm not necessarily betting on that I'm not hoping for it but it just wouldn't Shock Me whereas you can see the Innovation side that's progressing much more steadily yeah

there's periods where it kind of goes sideways or kind of bounces around or has Peaks and troughs but they're much more smooth and consistent it's no question to me that these two things are unfolding at different Paces because of different underlying facts and I think I I'd love to hear from other people what kinds of open public data sets they'd like to include to make these more robust or different ways they think it should be calculated that would make it more representative more accurate yeah I would double down on that if anybody has thoughts or ideas or wants to work on this with us like the cool thing about this this is all public data like we've said many times now and so

like we can truly work on it together with the community and take feedback and kind of just collaborate in various ways so I would encourage people to just reach out to us if they have ideas there's a there could be a great opportunity to work together that's a great point you guys know where to find us on Twitter on farcaster uh LinkedIn YouTube wherever you're watching us feel free to send us a note give us your feedback tell us why you disagree vehemently with the parameter choices we've made uh and we can make this bigger and better next year we've got a couple questions in the chat that have come through from people maybe we could just go over them quickly please first one is how do you account for custody crypto

addresses which are contained in an Omnibus account like coinbase yeah so a uh exchanges like coinbase use typically multiple Omnibus addresses they'll have like I'm just you know making this up it's obviously not my business but like 10 key Omnibus addresses and they'll receive different types of tokens this that would actually be something we'd be under counting right because if a hundred people or a thousand or ten thousand people had their balances virtually reflected in a single or in 10 coinbase you know deposit Omnibus addresses we would really only count that as 10. now they would be active probably every day because my

presumption is that the exchanges would be using them but they would radically under count the number of users and their intentions and their activities that are being represented and in a way it's a counter balancing Force to the fact that you know we have imprecision around kind of one user equals one address as Eddie was mentioning earlier and so it speaks to just like the you know imprecision in this whole uh you know analysis when you look at addresses but there's different forces that ultimately I think make it a pretty decent proxy exactly um and you're sort of alluding to this uh next question that somebody asked about the inferred user ratio uh the ratio between users and addresses what is our sort of take on that

it's a really good question that I think we would maybe need to dig into um you know there's a couple things that come to mind as to how you may be able to approximate that but nothing that I have top of mind right now but something that is a really good question that we should put on our list yep okay uh one more question um in terms of companies and Technologies in the space What are some of the factors used to determine who's VHS and who's beta uh those being a few famous uh yeah yeah cassette Technologies of the past of a bygone era

yeah VHS versus beta so fascinating one I wish I knew better the history of exactly why VHS beat the data Max to to stretch the analogy further but I think uh it really depends Space by space that's something that we try to do obviously you know we try to we try to understand the key Technologies in the space that are transient or temporary or the technologies that are really going to underpin a whole new wave of Economic Development I'll tell you right now like you know ZK multi-party compute fully homomorphic encryption those are not datamax those are VHS big time so get get ready uh but as far as like a different you know specific programming

models or specific zkbms it's really really hard to say we could go specific by specific I'd say like the key the key factor is just uh when there's a new kind of Technology what specific challenge is it trying to address and how resilient and flexible is that solution to the next wave of development you know maybe in VHS versus betamax it all came down to who had the right business development right maybe it was which which company plugged the right deals and helped build more VHS players right maybe that was some key business strategic thing maybe it had nothing to do with the underlying technology on the other hand maybe there was a Kia uh you

know maybe one is much more expensive to produce than the other at the limit maybe uh maybe they have different Fidelity characteristics that were important or maybe those weren't important uh you know maybe it has to do with the duration of the tapes things like that so interesting question if I'm remembering my consumer Tech History uh well I believe betamax had a more closed platform and uh and DHS was more interested in partnering like you said some of these business development deals and kind of let their technology proliferate um and to me that kind of speaks to the difference between web3 and web 2 uh the difference between openness and being closed and uh yeah well a friend of mine

just texted me that it's because on VHS you can put more on a tape even though the quality is worse maybe it may be apocryphal knowledge thanks Josh [Laughter] so any any other historians out there who know about the battle between VHS and betamax feel free to reach out with feedback for us on our on our take here our attempt to reconstruct history um but Eddie and Darren thank you so much for this conversation uh fascinating Illuminating Deep dive um if I were to summarize three takeaways here from this presentation I would say you know the market is doing much better than prices appear to indicate um these the financial cycling product

cycle are just different things they operate on a different logic uh I'd also note that scaling Tech is Going Bonkers people are really interested in optimistic Roll-Ups in zero knowledge Roll-Ups and then finally there are so many new products and applications coming online uh on chain gaming nft related experiences and so much more so that's a wrap thank you everybody for coming in uh again if you've got more feedback to share you know where to find us uh hit us up let us know how we can improve next year's report and also remember to subscribe to the YouTube channel if you're tuning in there or our podcast web free with a16z and also our

newsletter on substack web 3 weekly on all those platforms we'll continue posting resources data breakdowns helpful materials for Builders creators and web3 enthusiasts anyone interested in this next evolution of the internet all right Eddie and Darren thank you so much for this awesome run through thank you thank you