Numerai & Predictoor: How AI Changed Crypto Prediction Markets - Trent McConaghy & Richard Craib 566

Transcript

and so everyone was very excited about it because it was going to be prediction Market but the special thing about a prediction Market is it can actually make an article the money going in equals the money going out uh among the different people making predictions so it's zero some among them in prediction of feeds it can be positive some because there's new money coming into the system from the people making money trading 500,000 transactions per month about $13 million um dollar worth of volume per month uh We've spiked as high as um I think like 500 million per month data scientists sign up download the data you can say they're actually the consumers of the data but we're the consumer of their predictions which is based on our data and the models that they built upon

the data and that ends up being um very highly predictive and also lower volatility than individual models welcome to epicenter the show which talks about the Technologies projects and people driving decentralization in the blockchain revolution I'm ANS and today um we have a very special episode because I'm I have not one but I have two fantastic guests uh namely Trent mccon the founder of ocean and Richard crepe um the founder of numerai and we're here to kind of talk about prediction markers and prediction feeds so before we get into that let me tell you about our sponsors this week if you're looking to stake your crypto with confidence look no further

than course one more than 150 ,000 delegators including institutions like Bido Penta capital and Ledger trust cores one with their assets they support over 50 blockchains and are leaders in governance or networks like Cosmos ensuring your stake is responsibly managed thanks to their Advanced me research you can also enjoy the highest staking rewards you can stake directly from your preferred wallet set up a white label note restake your assets on I or symbiotic or use the SDK for multi-chain saking in your app learn more at chorus. one and start taking today this episode is proudly brought to you by nosis a collective dedicated to advancing and decentralized future nosis leads Innovation with circles nosis pay

and metry reshaping open Banking and money with hashy and nosis VPN they're building a more resilient privacy focused internet if you're looking for an L1 to launch your project nosis chain offers the same development environment as ethereum with lower transaction fees it's supported by over 200,000 validators making nosis chain a reliable incredibly neutral foundation for your applications nosis da drives nosis governance where every voice matters join the nosis community in the nosis DA form today Deploy on the evm compatible nosis chain or secure the network with just one gno and affordable Hardware

start your decentralization journey today at nosis doio hey trendon Richard it's good to have you both on it's really good to come back it's been a long time and yeah goad we we just checked this out and Richard was on seven years ago which is like 70 years in like regular years so uh yeah and Trent I mean you've been on regularly since yeah my I guess pretty common but it's great to be on again and you know always speaking about something different and crazy and it's great to join Richard today too so looking forward to it super nice so Trent I think you probably don't need an intro introduction on this podcast but let's do one anyways so you

are the founder of ocean protocol um which is a decentralized data Marketplace in abling data sharing while preserving privacy and you recently a year ago or so launched a pror um so tell us about that yeah um predictor is uh crowdsourced time series prediction so um there's basically two types of uh users there's uh data scientists that run um scripts or call them simple Bots that basically predict um will Bitcoin go up for down yes or know five minutes from now and specifically this would be say Bitcoin usgt pair and binance and when they predict they actually stake

against it they make money if they're correct and they lose money if they're wrong basically and so it's sort of this game among these different data scientists and on the flip side um Traders can come along and buy these signals and um use them for Alpha in their trading for different trading Bots Etc and this is really meant for people you know R running um scripts and Bots and that sort of thing on both sides on both the prediction side and the the trading side um because it is every five minutes or every hour so that's what predict for is a vot ultimately it's a vote time series prediction we have focused initially on defi because we see you know it's got It's high volume and it is um you know low latency uh but we have interest in exploring time series prediction in other domains too weather

energy and more so that's ocean predictor yeah super nice and Richard you founded numai um while back so basically it was the first time I kind of heard about homomorphic encryption I remember reading up on this before the podcast that we did seven years ago apparently and I was like ah this is fantastic this is so cool so basically the ideas that kind of you have that kind of it it kind of Prov provides like hedge fund like thing powered by crowdsourced predictions from data scientists and kind of they they kind of um the way that they kind of predict things is on encrypted data um so kind of so so as not to kind of leak um any

Alpha um and so kind of you you've kind of come to this prediction space from the trfi arena um more recently you kind of you launched numerai crypto but maybe let's start let's let's kind of stick with the traditional numerai for now so tell us about that sure uh yeah numerai that was the key innovation of numer wiise like we found a way to share this data set uh that we had that was very expensive and we wouldn't want other people to take our data and just go and start their own hedge funds and uh it actually right at the moment the the version of the data set is really more OB fiscated than encrypted uh in the early days of numerai um we were trying

to use some morphic encryption uh it was about a billion times too slow and then it got to a million times too slow and so it's still too slow but I like seeing this many entrepreneurs out there still trying to apply homomorphic encryption and crypto um but but yeah the the important thing is that no one can uh see or know what the data is on numerai even though they can still model it like all the structure is there but they can't uh can't know what the features mean or what the stocks mean and um numer has grown a lot since then we've become basically the biggest uh online community of people uh approaching this problem and there's millions of dollars

staked on models users stake their models uh and we've also traded billions of dollars uh of volume in our Global Equity Fund so it is quite remarkable how much it has kind of worked uh since you know there's definitely been a lot of um bumps along the way but it has pretty much we're doing the exact same thing we were doing probably in the last call seven years ago uh just at a much bigger scale yeah super cool I assume most listeners will be familiar with prediction markets maybe let's give like a super short um summary who who of you want to do

that I'll I'll give a bad one uh trying to give a better one but I'll scratch some of the surface area you know one of the things that's interesting with with numerai was I had just invested in and met with uh Joey Krug who started auger so you have to be maybe quite OG in crypto to remember orer uh but orre was something that that came out uh shortly before ethereum was launched and so it was like uh it was like the type of thing that was a project that was going to happen uh on ethereum once ethereum had launched and so everyone was very excited about it because it was going to be prediction Market but the special thing about a prediction Market is it

can actually make an Oracle and so it's a way of getting uh dat uh information that's offchain uh onchain and that is a really nice property of it uh because it's not just about hey people are staking and bedding and trying to win money that's not the cool part the cool part is that that market surfaces new information Orga was my first experience with that and you can see some uh indication of that in sort of being something a project that did kind of inspire numerai to have users stake on their predictions so that the quality of those predictions would go up um so brief not very brief but yeah intro to prediction

markets I can perhaps compliment that um and yeah of course there were two OG products projects on prediction markets um one uh you know in on American soil and the one on German soil or Berlin soil to be precise and that's of course with uh and was yeah no nosis with freder and Martin and and the rest of course um and uh being in Berlin we we got to know the the no folks quite well and still do of course um and uh you know similar to augur they were going for prediction markets and kind of the way that I see prediction markets are there's a broad variety but the way you know generally um I'll describe by example let's say you want to um know what's the chance of Biden winning as

not Biden that's zero these days probably what's the chance of Harris winning as president versus Trump right and then what if you can put your money where your mouth is so um and so you put you know you can bet on this you can put down $100 and maybe a bunch of other people have put down you know $100 here and there for up and saying you know Harris and other people put down their money 100,000 whatever dollars for um for Trump and of course uh if here's a key thing in prodiction markets if more and more people are betting on Trump then uh your potential amount that you can win if Trump actually uh does win is lower whereas let's say 10% of people are going for Trump and You Bet On It On Trump then you could win a lot more

right so um you can view it sort of like there's two tokens one token for betting on Trump and one token for betting on on Harris U as anation and the value of those tokens fluctuate with time and that also the market cap of the tokens total tells you the relative chance of them winning right and it sort of put your money where your mouth is but um the signal itself of whether Trump or Harris will win is is public and people don't have to buy it um and so it's basically this sort of zero some game among people um predict Trump versus Harris and they can speculate and buy on um buy and sell along the way so it is an oracle like you say and um things you say and um and it happens to have some of these

characteristics it tends to be for one time predictions um you know who will win as president uh and um and then with betting along the way all the time with the price of the bets adjusting based on the supply and demand and eventually it resolves so that's kind of how I see it overall um probably we can elaborate more later probably also long-winded but uh hopefully it helps to set the stage as we start talking about prediction feeds fantastic so uh I I I will uh not uh make make a third maybe I'll make a third one uh a third explanation and you guys can synthesize so in a in a way it's kind of a so a brick Market kind of is a trading value for athetic assets um that kind of um refer to usually future

events where the payout is one if it's true and zero if it's not true um and then kind of you have market dynamics along the way that kind of reflect um the probability of that event becoming true or not becoming true so that that that was kind of like uh the the compliment I hope so let's let's talk about um prediction feeds H how do they because kind of in a way they are similar to prediction markets um in in a way that kind of um they help you anticipate future outcomes how are they different yeah they they are very different um that but there's definitely a lot of similarities

um numerai receives uh Equity predictions so it's not like anybody is saying I think apple is going to be uh worth more than $110 on this date and sort of doing it per stock uh on numerai uh it's it's 5,000 predictions on every equity in the world um that is tradeable with enough volume and so it's just sort of set up not on a per asset basis on numer and it's also uh happening every day so uh it's it's really like we're getting a feed of predictions a stream of predictions each day on these different assets and so it makes it a bit

different than the event-based type prediction markets and I I can perhaps elaborate so definitely the the feed is probably the key to finding characteristic it's St these oneoff it's um over and over every 5 minutes every hour every day whatever um besides that uh at least in the implementations that numeri and ocean have uh of prediction feeds compared to prediction markets prediction markets the the feed itself is public and it's a zero sum game among the participants that are investing in prediction feeds um the the feed is private unless you pay for it in the case of numerai right now there's exactly one customer for that feed and that's num the hedge fund

in the case of ocean people have to come along and pay money to get access to a feed Ty a 24-hour subscription to access a feed um and that tends to be Traders so basically that part of the feed uh prediction feeds the feed is public prediction markets that signal uh of you know win versus lose Etc is public and no one has to pay for it um and then this also means that uh if it's private you know part of the reason it's private is so that people can make money on the other end or you get them to pay for it of course right so um in the case of numerai hedge fund I it's implicitly paying for it as part of the overall flow in the case of ocean I mentioned the pay for um so but yeah the second part here is is zero something so

imagine uh with a yeah a prediction Market it's the money going in equals the money going out uh among the different people making predictions so it's zero so among them some people who predict well make money other people lose money in prediction feeds it can be positive some because there's new money coming into the system from the people making money trading so um just if it was just edding among the predictions every five minutes there sort of you know people getting money people getting slashed Etc back and forth among the the predic uh the predictors making predictions but when you have this Revenue coming in from uh you know people paying for the feeds or paying B paying you based for your feeds based on the profits then uh that flows back to

the predictors and then they make money right so it's a positive sum game with um for the um the data side is the B making predictions within prediction feeds so yeah to summarize then it's uh prediction feeds are feeds not onetime events they are typically paid for and private not public and it's uh positive sum game among the people making predictions versus a zero sum gain in the case of prediction markets but you can you can actually turn pre markets into positive sum games the same way right kind of you could pay W them and kind of have a bunch of experts kind of place money on on things and kind of then incentivize the experts afterwards depending on kind of like how

they did right so I mean so basically I think that's kind of like maybe an arbitrary distinction and that kind of that's how they're used but it's not it's not imminent to kind of what they are is a little bit in maybe in that uh to know the price that the current odds are uh to to know if you're going to uh participate in this prediction Market against Harris you would really need to see the current price and so you you maybe wouldn't want to like uh bet blind without or or have to pay to see the current price before you actually participate but and and to be honest I would see like you you can you know view prediction feeds as iterated prediction markets but you have to change three

characteristics right so generally when you change one engineering artifact one technology artifact by several factors if it's sufficiently different then it probably is helpful to have a new name so as not to it with other things so we thought long and hard when we were launching predictor and asked you know is this a prediction market and you know it chatted with Robin Hansen and with nois folks Etc and realize it probably does make sense to call this something unique right and talked it over with Richard and stuff too um and that's that's why we've just um you know really focused on using this label prediction feeds but we're not saying you know we're not going to be um religious about this saying uh it's definitely not a prediction Market it's just much easier to think of them as two different things with those characteristics as the

defaults um and and then you know there can be a blurring uh as you toggle characteristics one way the other uh and probably the main distinction is the one offs versus the the feeds yeah yeah so kind of one of the things that we haven't talked about yet um is that for prediction streams the data is often actually generated by algorithms or AI models and I mean this can also be the the case for prediction markets as we see with kind of like nosis AI stuff but um it's it's the norm for prodiction feeds right I would say that's right yeah um it's definitely the normal numerai you can't Pro provide any predictions that's

like based on your impressions of uh apple or or something you really have to build a model um on all of the equities in in our data universe so definitely very much uh leaning towards AI models and there's no value to human ins ites onto stocks on M so preure markets the idea of them has been around for a while and they never really took off until very recently and we can talk about why why that is so um maybe a little bit later um tell us about prediction feeds in traditional Finance are they things that kind of existed pre uh numerai and kind of uh

web three yeah I would say they have I mean in some sense even signing up to Bloomberg it's a subscription to a feed of data it happens to be the live prices but it's it's kind of an estimate uh no one has all the uh details of the live prices of stocks because there's dark pools and there's all these small uh details many data vendors would have different price numbers and things like that so that's uh in some sense the same type of model and then uh many data vendor would sell signals and they say we've made this thing we've used machine learning and we've used the data sets uh we don't have a hedge fund but we are

going to sell this for $50,000 a year and so that was kind of common but uh in some ways those often were very lowquality products and so why would you sell something if it was so good why wouldn't you trade it yourself and um and what's so nice about the staking mechanisms in and numai is that it's saying the you know this actually is high quality because I've staked a lot on it and you can see it's high quality because look at this track record I've developed and so that's where it starts to make uh particular sense for blockchain use case so if you look at um systems that kind of consume prediction feeds um can

you kind of give us an idea which domains they span is it just Financial or is it also so you I could I could imagine things like weather forecasting prediction fees like C can can I travel a specific passage with my shipping vessel or um I could imagine kind of like Insurance related um things that kind of could be predicted and tracked and so on so is it most def Financial or do we also see them in other venues I could talk to that sure um yeah so you know when we were uh exploring uh building predictor we actually looked at a whole bunch of different domains and

um I yeah we we found that there were uh prediction feeds in many domains and um typ quite often they had different labels in different domains um and different ways of consuming so for example if you've got an iPhone you open up your Apple weather app and it tells you you know what the forecast is for you know Sun versus rain Etc 1 hour from now two hours from now and so on so you can do that a prediction feed if you want right and it's rolling so um 1 hour from now it's going to tell you um the feed 1 hour in the future and two hours in the future from that right um and under the hood um there's a lot of data and modeling going on right um weather prediction is uh um a huge sub Ministry but we actually see that it's it's

mostly um via governments right so governments um they pay uh National forecast Services um things like no OAA as part of the USA and and more um and then that gets uh you know that outputs um predictions as API that get consumed by things like apple weather and stuff and apple might have its own AI too so that's one example within the weather domain in other domains um it's uh much more um private so for example in energy forecasting to out there's an 80 billion industry for management of um energy um so this is basically the software and systems around it to manage energy um you know balancing the grid all of this sort of thing and a subdomain within that is a forecasting of energy demand

and energy prices um but it's highly regulated this is where energy trading is you know of Enon Fame for example but it's highly regulated so you don't hear much of Oda and it's very opaque uh very hard to see you know if you talk to people in the energy space they can tell you a bit about it um but it exists and um so if you look around you know this exists in various places uh usually very very different shapes and sizes you know Bloomberg criminals versus your Apple iPhone versus some rest CPI right and uh we realize that um given that we're in the crypto domain uh whether the application is crypto or trafi or otherwise um it made sense there's a lot of uh you know prediction markets are well-known term there so it makes sense to have a term that compared and

contrasted what a prediction Market is versus um this stream this feed right and that's where we why we chose that label on that way um as opposed to just calling it forecasting because there there's different stuff going on too right you know if you call it forecasting it's like okay well how how does that differ from traditional web 2 approaches etc etc so it's useful to sort of put a stake on the ground for this uh qualitatively new thing in many ways you can view it as a even you know a new um crypto building block a new Atomic primitive if you will that can be used in crypto whether it's um you know shaped the way that numai has this technology or Ocean Technology or otherwise can you give us an idea of kind of how the ocean technology is used so give us an idea of kind of what the

things are that are being full casted who are they being full cast by and who are the consumers of these feeds yeah so our Focus uh for starting is is defi like mentioned we did look at a lot of different Industries and you know talked with a lot of different people and we saw in terms of the different Industries our criteria were um what is the um amount of volume uh that can go through and U what's the latency you know if you predict well how um how quickly can you turn around and make money from those predictions you know so for example if you're um predicting um medical um interventions you might it will take you five years seven years for those predictions to get SD approval Etc so you're talking like a you know a clock rate of one clock cycle

of five years say uh versus if you're doing say say gaming and you know marketing for gaming you know prediction of of Market marketing metrics is actually super useful maybe you can get that down to a month or maybe even a week um but then if you start thinking about trading um tradire defire otherwise you're you're looking at you know U minutes if you want right uh and you can even go to to sub minutes if you want to get in the realm of highspeed trading or M extraction right the M Searchers so um it's but the point is it's very high value flow and it's uh very low latency so uh from that we we given that uh ocean had a pretty thorough web three stack we had a lot of stuff around um you know decentralized data sharing all that you know tokenized

data Access Control including selling data feeds so we had a lot of off the shelf Tech then um we we you know started with we used that as the starting point and then um basically built an app SL mini stack on top called predictor so at the heart of it you know whereas uh yeah basically the technology is at the heart of it it's a smart contract running on a confidential evm Jane So a privacy preserving evm Jane Oasis Sapphire it uh when we went live it was the only one in production and it's uh you know it's been in production a about a year and a half and quite stable and the O Oasis team has been great to work with so that is privacy preserving uh it happens for using te right now uh until sgx but you know that

keeps expanding into or is on path to expand to other Technologies so myself as a predictor a data scientist I submit a prediction say that Bitcoin us pair and binance will go up five minutes from now I stake say 100 ocean and you know maybe Richard uh he predicts that it will go down and he Stakes say 200 ocean okay and maybe you fto R you stake 100 ocean going down so now you've got 300 ocean going down 100 o going up for five minutes from now Bitcoin usgt and binance anyone who has bought that feed can see that they'll say see that it's a 25% chance of going up 75% chance of going down and they'll see the total stake up and total stake down right 300 down 100 up so that's uh what happens and then when the actual uh true signal

emerges five minutes later then everyone gets paid out or slashed accordingly so let's say goes down then I get slashed I lose my 100 ocean and you guys get my 100 ocean Richard would get two-thirds of that Frederick would get 100 so Richard would get 66 Frederick 33 and let's say that there's sales from Traders buying this and maybe it's another 100 ocean worth of sales then Richard would get another 66 ocean from that and Frederick another 33 um and you would get your stake from back from before to you're not slashed so to summarize I predicted wrong I lost money I was slashed you guys predicted right you made money so you got my slash at stake and you got your um cut of the revenue and you know how much you got

from me from my stake and from the revenue that's Pro to your estate so so that's how ocean works at the very heart that's all on a smart contract relatively small I think uh you know a few lines of code um all running on Sapphire as a E20 solidity smart contract and yeah it's ec20 too so it behaves like an ec20 uh token too in some ways uh which is pretty neat and that's actually for the for the subscriptions right so you buy a subscription when you buy that you get 1.0 uh data tokens basically uh that gives you access to this um feed right so that's what ocean predictor is at the very heart and then you're the only because of the Oasis privacy Tech you're the only one once you buy this uh you're the only one who can see it correct yes

at first yes and then uh that that's exactly it so there's handshaking with with Oasis uh to make this visible to the person who's bought it and there can be more than one buyer right now right too um you know we'll see in the future because if too many people come along and buy and take advantage it might diminish the the signal of course but we haven't seen that yet uh we expect to we hope to you know it should happen Game Theory wise um explain to me why I can't resell so kind of like say I buy this token that kind of lets me see the feed yeah that's a great question so um you know you could uh but this would add a lot of complexity to the system and give if if you um get closer and closer to you have to submit your prediction 5

minutes are greater than when the prediction is due right so if you start wanting to speculate and buy sell and resell in the 4 minutes 3 minutes before then the prices would change Etc and that gets you closer to prediction markets and that's fine but we decided to keep um it very simple and straightforward where it's just um you know every five minutes boom every five minutes boom and you could have a rolling too it could be every every minute you have five minutes ahead every minute you have five minutes ahead rolling like that right but but but I can I can only um break five minutes ahead I can't prict say 24 hours ahead uh we actually have it right now where it's 5 minutes ahead and 16 minutes ahead um and we didn't we you know right now you know it's this the systems a

year old we've been um focusing our energies towards making sure that everyone in the ecosystem can make money so we haven't bothered expanding beyond that um we will see once you get Beyond 24 hours then it makes much more sense where humans might want to intervene if it's every five minutes it would be very tedious for humans to intervene Like Richard was hting at right so robots getting involved but you know beyond 24 hours I um it's I'm not sure how much value there is and prediction feed versus a prediction market so prediction feeds are in my view tuned better for less than 24 hours uh basically you know they can be much shorter time scales and by not having the speculative aspect in between it um it makes it really clear how every how all the incentives work

etc but who knows maybe in the future it will be better to have that right just the system the designed we have right now doesn't have that how does this material materially differ from a Futures exchange right so kind of say um I if I think about say darab it um and kind of there's kind of like the prediction for I can see that as a prediction um of the uh usdc bitcoin price 24 hours from now whatever the intervals are that they offer um couldn't I just kind of use that as a basis and then hedge yeah so uh absolutely so they're very complimentary if you squit you could even call it a future uh you know a sty a type of derivative that happens to be Futures is

but I would say it's different and probably the best example of uh how to to illustrate is simply you can use um uh predictor prediction in order to buy to make a Futures BET right and in fact we actually have a lot of successful experiments in jally where it is I'm betting on buying ads Futures exchange Etc right so um if you squint at when when I'm making a prediction I am making a bet on something happening five minutes from now I would say but um it's uh it's a slight you know the the shape of that that is different than it the the shape of a traditional future of course if you say you're just doing derivatives of stuff in the future then it's completely wide open right so you can view that this that the prediction feeds we talk about are a specially

shaped type of derivative right um but it makes it different than the Futures as well um yeah that's how we see it uh and I guess the other thing and I know Richard you can probably comment on this in the way that it's shaped it makes it very easy for non traders to engage because they having to put um traditional trading money in to trade against the price itself for the future price they're in sort of a more direct way it's sort of indirect allowing uh data scientists to engage on you know was was my prediction accurate yes or no without having to engage in the trading more directly and in some in some countries some jurisdictions you're you're limited from doing that too right so um and for many people it's just much less scary to operate on predictions you know people that like to participate in

a data science competitions like Kel and so on you can view this like a data science competition with teeth right just before we pass back to to Richard tell us about the people or the AI kind of um issuing the predictions and the Traders consuming it do you have kind of like a user profile yeah uh partly I mean we have the user profile we targeted uh but we we don't know exactly who's doing it because it's you know uh fully decentralized and all Anonymous Etc right so this is all one chain we we get Stu and uh if you go to DAP Rader we have sta ocean predictor so we see that there's um 76 weekly active users so you

can view it as um you know 76 wallets that are um making bets any given week um and maybe some users are running 10 walth maybe it's all completely individual we don't know and then the volume it's about um uh 500,000 transactions per month about $13 million um dollars worth of volume per month uh we've actually spiked as high as um I think like 500 million per month so poly Market is getting a lot of um press around this we're not really pushing ourselves to get the Press we just know that it's growing and part of the part of the reason for that is goes back to the Target profile which is we really are targeting the um the defi traders who know their way around Python and data scientist data science so maybe

you're you know an AI Pro or data scientist by trade and you know a bit of crypto or maybe you're a a crypto Pro by trade and really know Python and are dabbling in data science um it's so you basically have to have some python skill and some crypto skill to get into this but we make it really easy from there you know going from you know you go to predictor a on the top right there's a button run Bots it takes you straight to a read me and you know 20 minutes you've got um a python script running that's running your Bot on on testet right and then another 10 minutes you can have it running on Main making predictions so um it's really meant to be low friction but targeting people um that demographic right um people who uh aren't scared of uh you know some degree of

professionalism in crypto and in Python but we also want these people not just to run the bot straight out by default um you know then you can make some money but if you want to make really good money you have to have more accurate predictions than the next person right because then you can be basically grabbing their you know slash stake more often than they grab your SL stake and make money over time from that right so that's the demographic in terms of the Traders it's probably some of the same folks but uh even more so it's uh you know there's a lot of professional hedge funds out there that quietly operate that you never hear about sometimes One Man shows often teams of five 10 15 people and you know being crypto for so long we know quite a few of these different teams

right we talk to them and and get feedback back and forth and how how we're making sure we do a good job on all this is we're dog food fooding it internally so you know internally uh in Ocean I I'm running a team where it's a bunch of Engineers several of them know data science very well they all know python well um and we've got you know some web capabilities in one. two and but uh our number one kpi inside inside that team is make money trading right uh and and then we pass that off to our users and by doing that you know we we can have very good empathy for the users but it's also um yeah and a strong incentive for us to just keep getting better better better right what about you Richard so I know who the consumer of your data is namely numai hedge fund

um I remember when we talked about this seven years ago kind of you were targeting data scientist how how has that panned out yeah so it's still data scientists data scientists sign up download the data you can say they're actually the consumers of the data but we're the consumer of their predictions which is based on our data and the models that they built upon the data so everybody downloads the data they train the models they submit them and then they stake them uh and what we get is uh thousands of models that are all ensembled together and that ends up being U very highly predictive and also lower volatility than uh than individual models and so we're really relying on

the ensembling of the different models to have it uh make it make a good hedge fund uh at the end of it and how successful has this H hedge fund traded over the last seven years uh yeah it's been really uh good I mean the we're not supposed to really talk about performance um but but uh you know you can say we can say some general things uh we we we have very very good um start to the track record even though the company started a while ago um the actual investing side of the business took a while because we need to build the community we need to get a competitive amount of data uh and we need to kind of pioneer uh staking uh and our NMR token and so um once we got

all those things right in about 2019 uh the fund uh did very well and grew investor interest a lot of the investors in our fund are you know endowments or Pension funds or people like that the fund is really an Institutional uh product and so we had a yeah we had a very very good period of of the last five years where I think all but all but one year has been very very strong performance it's been one uh down year um there's all sorts of qualifiers there you talk about the volatility and the sharp ratio and all that type of thing but not going to get into that uh during this podcast yeah absolutely um so it was

very clear when when Trent just talked about kind of like how pror is set up that everything is super decentralized right um and I think this is less of um a focus point for numerai um so tell us about kind of what are the trust assumptions using numerai and um why why do you think it's fine as is yeah well it's uh it's there's some definitely good things and there some important things that that are required I mean we were not hoping for thousands of individual people to start their own institutional hedge funds uh to do so would cost like $30 million each and so in fact I think a lot of society's uh capital is wasted on uh new marginal

hedge funds sucking up Talent uh when really you just need one hedge fund that makes it possible for any talent to contribute and that's really always being numer thesis so we like the fact that if you have NMR we cannot take it back from you right because it's on a blockchain that's a nice Tool uh for us it's much better than having points in a database having the actual cryptocurrency um we like the fact that for the majority of users uh when they're uploading their predictions they do not have to divulge the IP uh used to create their model so their IP really is theirs we can't uh take it from them uh

we like the crypto asymmetry of we we we can give you this incredible data set that you can train your model on but you can't steal it from us and use it for your own trading so we're kind of using all these different elements of crypto But ultimately it's much more efficient to have the execution and operation structure of a hedge fund be kind of uh centralized and uh and that doesn't mean you know all these other pieces can't be be decentralized I mean many users for example our users put together a website called numer Bay it's like eBay for numerai and you can buy and sell numerai

models there's nothing we can do to shut that down and we don't want to shut it down like we think it's cool but we didn't think about it and and that just kind of popped out of the community uh and there many other examples of that too where the communities made open- Source things there's some called numer API that a lot of users use to connect to our API we didn't even write numer API it was written by the community so all of those aspects uh it definitely makes for quite a healthy decentralized Community but the core investment product you can't trade equities on a blockchain legally it has to be uh centralized hedge fund let's talk about the elephant in in uh the room and that's uh AIS and uh AI

models and agents what roles do they play in your in your systems well that's yeah Trent and I had a lot to uh to talk about when we first met I think it was maybe Fred urom from coinbase uh who introduced us and um we just had a lot of similar background we were both machine learning people we we we happened to get into blockchain but uh we both we both knew Ai and I think we all we both saw a lot of what was coming in Ai and to me the the magic piece is some is actually the intelligence uh being built in these systems not the user interface or the

all that stuff's good but it's really like did it actually work to predict uh this very hard to predict thing and um and with more investment in in AI there's more data scientists in the world than there were when numerai started in 2015 and so uh there's just so much more going on there's so many new architectures some numeri users for example are using Transformer architectures uh in the development of their models and so it's just like somehow it's kind of all all been uh it's kind of been a long time coming but it's somehow all being proved out and I think a lot of the things Trent had done

early on and numera had done set us up for this kind of time and yeah maybe I can add to that too um yeah it's uh you know Richard and I it was great to be introduced right when both ocean and numai were getting going and we were swapping white papers and giving feedback and so on we've been in touch ever since um so you know there's I think there's a lot of really cool um quiet sort of collaborations that people never hear about but that actually do affect you know what gets built right and that's that's that's really nice on your question uh you know the elephant of the room AI yeah I so in predictor I didn't mention this in the onboarding when you go through this read me uh under the hood uh the script that you're going to be launching and running

is uh building uh Mei model of fly right you can have super simple models that are you know not even AI just like a linear regression model or a linear classifier but then you can get fancier and fancier with things like boosted trees XG boost this sort of thing or goian process models U and we have all that out of the box for you to just uh put in and set your own parameters um and then you can get fancier yet with you know large neural networks Transformer type stuff and so on all you want and that's pretty easy to add in um with your own without changing the architecture much and that's nice uh part of the reason like we we went for a prediction was you know in Ocean we thought a lot about um data marketplaces of course we built them over over the

years and often the things that we built we saw that people you know it was chicken and egg uh do people want to you know do you try to get lots of supplies data first and then get people to can buy it or do you want to get you know create people to buy it first and then Supply or do you want to have sort of like one ecosystem where you've got buyers and sellers somehow together and we realize yeah you do want an ecosystem of buyers and sellers together and then you need to make it really tuned towards um that ecosystem and then we realized like it was a key constraint and maybe this is obvious to any entrepreneur but it kind of hit hit us in the head like uh uh B of rocks you know you have you have to be able to make money right so

we we we uh so you put this in a loop and it's you know debt data you know you spend money to buy data or create data from that you build a model you know and then from that you um from that model you run inference and you get a prediction and the prediction is the final piece of data from that prediction you take action to make more money and then you loop around you make more money create more data build another model make prediction um take action from the prediction make money loop loop loop taking the money made to buy more data that closes the loop and overall then we call that the data value creation Loop and it's actually been sort of a a core thesis um for us overall saying okay you know how how much money is potentially flowing through this Loop depending on

the domain how quickly can you go through the loop um and that's what led us to um well predictions as a first cut why because it's much more valuable to be at the tail end of a Supply Chain versus at the beginning it's better to be at Starbucks than a CO a coffee Farmer at St and Costa Rica right you know and I grew up as a wheat farmor in Canada right where you know you're getting a dollar for um five Bush for a bushel of of wheat you know that's the equivalent of say yeah 5 gallon pale so 25 lers Worth right so for a dollar right and then you go and pay buy a loaf of bread and it's $3 and you know you can probably make I don't know 100 loaves of bread from from that wheat so overall the point is it's much more

profitable to be at the tail end of the supply chain and that in data land that's predictions right and that supply chain is both Ai and data it's not just a data supply chain it's Ai and data and so so basically numai and ocean I think part of the reason we're having success with these is that we focused on the high value stuff first and then the rest of the supply chain implicitly gets filled in people are willing to spend the money to do centralized things or you know locally centralized to themselves but if you have 50 or 100 of them then it becomes decentralized to to you know to get the data to build the models Etc and you provide you know low you you know you give them some guidance to make that low friction we do see that as time goes on um the rest of this

supply chain will thicken out right just like um initially you know you had basic coffee supply chain back back in the day and you you needed coffee buyers you know people who want to drink coffee but then over time the whole supply chain thickens out and I see this now as this is how the data supply chain thickens out over time so working backwards you know you can have decentralized model inference you can have decentralized model training and so on but it's not a prerequisite right there's a lot of great work happening in that I'm paying close attention and um you know know we've even thought a lot about doing that ourselves in Ocean but our Focus right now is still focusing on the predictions simply because that's the tail end and if we do a great job on that everything else kind of half takes

care of itself right uh so so that's uh Ai and the D supply chain and of course this will extend um where I'm most excited up but what this I can extend is weather prediction imagine predicting weather for every one of the 500 million square kilometers on the planet you've got 500 million square kilometers times about seven key weather metrics um temperature precipitation humidity Etc that's 3.5 billion feeds so if you start training a model for that you're going to have a huge ass model it's going to be super intelligent just not super intelligent in hum shaped intelligence but super intelligent as in sort of all knowing a lot of the Dynamics of the world at sort of a weather time scale right so to me that's super exciting and you know fors of artificial intelligence that the world hasn't seen before but

can be extremely beneficial to society and uh that's what I really hope where we can take this technology and along the way you know help to grow a decentralized DAT economy do you see um the quality of AI predictions kind of leveling off if it were possible to kind of make perfect predictions kind of that there wouldn't be a market left right kind of the market kind of thrives on the inefficiencies to a certain extent um how so basically reframing this how good do you think these models can get kind of where kind of the minimum is kind of the noise that's kind of inherently there that kind of you you can't you can't get to

vanish yeah well that's a very good question um it's hard to know it'll be hard to even write a proof about you know this is the this is the ceiling uh what you could do and I think obviously it's dual with data though so that's being the story of of numerai if you kind of look at num's back test results you might say okay at this date 5 years ago we're seeing sharp of one and then it went to 1.2 after we added some more features and now you just keep adding features and uh suddenly you add a sharp of of a back test a CHP of three or something and so that's the type of uh growth you can see uh even by holding the model constant so if you hold the

model parameters constant and just increase the data that scales the performance but you can also scale them both and so you can have the model get 10 times bigger and the data get 10 times bigger and then that uh you get so basically I would say like it's impossible to say there's no better way to make a model um but uh it is remarkable to me how much uh we've been able to improve on you know we would have very say tough periods in the market say March 2020 where there's so much hedge fund crowding and uh maybe some of the numeri models all went down at the same time well it doesn't seem like if you had those models now

upgraded to the latest data that that would happen again and so there's there's always that type of thing happening but of course the same story might be happening inside some other hedge fund so they would have told you that 5 years ago the sharpest one and then they improved it and so there is always the uh Spectre of competition and but one of the special things that we like about numerai is that the the returns are quite uncorrelated I mean the times that we have found to be difficult for us have often been the times the Market's doing well and factors are doing well and all the other hedphones are doing well and somehow for us that's a bad those are those are tougher periods but that's a very good thing because it's almost the goal of an

investor is to combine a set of uncorrelated return streams and so uh it's very good to us that uh we're different and different in in a way that's kind of like increasingly better do do you think Transformer models are the best we'll see for this or do you think there's going to be another step change um one of the great things is uh you know I'm uh uh I'm I'm used to be younger so I was 29 used to be I was 29 when I started numer and I felt like I kind of knew a lot of the frontier things about machine learning and Ai and then I had to become a you know as somewhat of a jack of old trades

being the CEO of a of a fund that had a crypto and a community and I actually right now couldn't easily whip up a Transformer model myself but luckily uh my past self plan for this Moment by creating an open hedge fund where anybody who suddenly is an expert in Transformers can add that intelligence to numeri to the extent that it can be added and uh so that's uh kind of uh yeah kind of how how I think about it and and I I would say that yeah the one best user on numerai right now his username is the human Peep and um he claims to be using uh neural networks

and Transformers do you think this is kind of the end or do you think um there's going to be something better than Transformers I think there will be something better uh in a way that's uh probably a lot better um and uh I do you know in some ways numerai we are betting on model Innovation right if we just had there was one best model we would just run it ourselves uh but the the we have constantly seeing model Innovation be possible not just compute and data but model uh the model parameters and the type of model can have a very large effect especially in this High signal to

low signal to noise environment what do you think Trent uh yeah so overall I I think it's actually worth giving a shout out to Richard and the numerai team for increasing their sharp ratio on all these other metrics over the years relatively consistently right so that's a really nice uh demonstrator of the thesis so really like they're not as well known in the crypto space because they really have focused on their customers that the LPS into the hedge fund but uh it's really pretty impressive um you know predictor has been live for a year and we have seen the accuracy steadily um creeping upwards um and that's a sign you know there's this AR race among the people submitting predictions and they want to make more money so that's happening in practice and um you know the the the incentive structure um allows for it

right such that there's you know more and more people um continuing to to uh try different things and whatever works better they use and you know we at the same time with our internal team we've also been making advances you know we were we didn't have great solutions to handle non-stationary data at the beginning where they underly dynamical system uh itself changes and you know we've cracked that in a pretty nice way now for example internally and that allows us to train on much more large data sets Etc it's it's a big unlock actually and I'm sure that some of our community MERS have done that done that too of the traditional time series methods worked so but I've you know I've been in AI professionally since the late 90s and I've seen a huge evolution in AI so there's no reason that you know the

current state of the art of AI will end as a baseline Mo's law is going to keep rocking along because there's so much money to be made and um you know we're going to get improvements in compute power we're going to get improvements improvements in data storage we're going to get improvements in the amount of data and we're going to get improvements in the AI algorithms themselves and each of them has their own sort of subm mores law and they all combine together and if you look at the numbers it leads to about a 10x Improvement in AI capabilities year to year to year to year right um from these different components so you know Transformers were a nice Innovation and there's some other you know things coming down the pipe you know things around um uh this is uh logic around planning um you know making

plans of um how to how to decompose a task and then work it out and there's you know some really great planning research going back decades but you know now people are putting this into an llm context um with some pretty cool results and you know chain chains of logic and trees of logic and reasoning and and reasoning as synthetic data uh rationalis synthetic data sorry so there's a bunch of stuff coming in the pipe right now you know um that sort of freshes in the last half a year and and finding its way into the latest greatest llms but that's still only a piece of it right we're going to see more and more and more um H in the next few years and it's just not going to stop right until we hit you know AGI which is basically power te across the board and the beyond

that too and um because in in finance trafi and defi there's lots of money to be made um for sure you know a lot of these models are going to find their ways directly into here more and more of too so I'm pretty bullish for AI I think it's you know there's lots of money to be made in many many domains and we see that it definitely will happen here and there's a huge design space to be explored for shapes of AI um systems and we're still only at the tip of the iceberg we've got you know we've only cover of fraction of what we're going to see the world's going to change in huge ways yeah I I I tend to agree let's hope uh let's hope that uh most of them are positive Richard you recently launched numerai crypto uh which kind of veers

into the direction of pror so so tell us about that and kind of how how it's similar and different to what trend is built yeah so when Trent was building a predictor he's like by the way we we're going to be doing something uh like this and then when I was building uh numerai crypto I was telling Trent you know we were thinking about doing this so we definitely like Tren said we do feel like this is a collaborative space at crypto it's amazing how much it works that way um but we just had with numerai uh a lot of of data scientists who've signed up over the years and simply because we've paid them in cryptocurrency they've learned about cryptocurrency uh many of them had when

we first paid them and so we kind of onboarded a few thousand uh very bright uh people to ethereum by giving them NMR but the question was okay well numeri is getting very hard why is it getting hard well you have to be you have to add intelligence to a system that's already very intelligent and so it becomes harder and harder the more uh models are being added to the Ensemble and we thought well let's uh create numerai crypto which is exactly like numerai except the universe of stock the universe of of stocks of 5,000 stocks which is what numerik normal has uh is

now just 500 cryptos uh that are the highest market cap cryptos uh that add D duplicated and and a few other things and we we made that universe and we said you know what we don't have any crypto data uh so we're just going to say bring your own own data uh go and find data you can use news data you can use any data you want uh but you just have to submit predictions on this universe of 500 stocks now where it's different from predictor is predictor is these five minute predictions and uh many many firms operate on that Horizon including you know like a Renaissance there would be many models on this shorttime Horizon but numerai had basically specialized in

making one month predictions now it's not because we like hate short predictions or something it's simply because uh one month predictions have higher capacity for hedge funds uh so most people don't care if they can turn a million dollars into $2 million they're happy but uh a hedge fund that can only turn a bill they can turn a billion dollars into one billion And1 million that doesn't actually move the needle very much so if something's got a stressed capacity um then usually small cap cryptocurrencies have stressed capacity then uh it's not so valuable but if you can trade over a few days and it doesn't matter how long it takes to get into the cryptocurrency well then

you can make uh quite a lot more money and so you have higher capacity so we decided to create this thing called numer crypto you've got these 500 stocks everybody started joining and submitting and staking they're still staking to say that they believe their predictions will work and we launched it in July uh the prediction Horizon is 20 days out and we did the other another weird thing which we've never done before which is we give away all the predictions so no one uh you don't you can't we don't use the predictions because numerai doesn't trade cryptocurrency we just said we'll just give them back to the users that generated them but we're going to give

them the Ensemble of all the predictions add it together and the incredible thing is just how well it's worked um there's being almost no uh sort of one or two week periods of of uh the correlation of the models being negative uh with subsequent returns and so um that doesn't mean it's going to be positive forever it's very likely to be negative it's some point and this is really an educational interesting thing we wouldn't recommend betting all your money on these predictions but you can go to our website and download the live predictions and you can start playing with them yourself and see uh how they work um and see oh well yes you did uh

The Meta model did predict this stock right but it's a very low volume stock so maybe that's not that valuable but it also got Bitcoin right and so that's a very high volume stock so I could have made a lot of money on that and so um ironically by the way any Bitcoin Maxis out there but right now the best uh coin ac across all 500 is is is Bitcoin the second best is ether I think so is a skewed towards High market cap uh cryptos right now um whereas sometimes it's not it's very different so that's what it is and it's just been really fun to watch because it's been a kind of um it's just been kind of full fun um we

wanted to have another use case for NMR and so now $600,000 per day is staked on these predictions and um it's just been a really cool thing to watch and it's nice to see people using them and thinking about these predictions and trying to maybe turn them into a trading strategy but for now nuai doesn't have plans to do so ourselves but if I get this correct then kind of currently you kind of pay paying for these predictions that you're not actually using so how is that sustainable yes so we pay them uh from our Treasury and uh and uh it's really to seed the whole thing and and see what happens uh we don't pay that much you

know the vast majority of our payouts are going to the equity's predictions but um we are paying some to the crypto predictions to to seed it and see what happens and I think it benefits numai if we have more stuff for users to do more reasons to stake NMR instead of uh sell it um and we can always do anything we want with these in the future um there many crypto funds that have say recently maybe bought NMR or something now they they might trade uh they might trade cryptos and they could ask us uh you know hey can we get this exclusively or

or or something like that now I hope it doesn't go that way I wouldn't want it to just be a kind of exclusive thing but um right now it's just uh it's just very interesting to watch and and it's amazing how much how good it is I mean for a perspective um it's very good to get sort of 2% correlation with uh equities um but this has 9% correlation uh with the subsequent returns of crypto so it's just a very very much much sort of like worked way better than we thought it would and it almost yeah kind of gives me an indication why Trent is excited about what he's doing because he's

probably seeing similar things on the shorter Horizon yeah I can jump in um certainly and I'm very excited for the the results you guys have had with numerai on these longer time scales and yeah we've seen you know in the five- minute Horizon actually it's it's hard to get a lot of signal but once you get to one hour four hours eight hours it's it gets uh really nice you know probably the you know start be really nice at about four hours right um so but even on um you know five minutes you can get on the order of uh you know Baseline 52% but as high as 65% depending on on the token and stuff too uh accuracy that predicting up down right um and so uh yeah I mean overall I think we fully we both agree that

there's a lot of um uh maybe not low hanging fruit but medium hanging fruit for predictions in in Defi and what we're doing you know it's it's complimentary on purpose you know in at least a couple Dimensions one is time scales numerai is focused on longer time scales and the second dimension is just the structure and incentives of numai versus ocean predictor you know um so numai is a a hedge fund that uh is that single customer for those feeds and so you get paid in that that way that way in Ocean's case it's um a feed with many many buyers and so it's you know just different payoffs you could have I I'm pretty sure that there people that are participating in both the numeri

competitions and the ocean predictions you know for um you know the 20-day stuff as well as the 5 minute stuff and everything in between uh that's great right um and uh I think those are the probably the two main dimensions of of difference and then uh overall though um and I guess you'll get to this Frederick but you know what what are the there's many potential points of collaboration and we've started going down this path more so right so um around the same time numerai uh crypto was launched I think a week later uh we launched with numerai um a data challenge uh called crypto facture modeling data challenge so this is basically some of the uh stuff for Richard was describing but where people can submit what they did into um this ocean data challenge describe what they

did you know because with if you're just pure numeri you get to stay opaque but if you want you can put this into a PDF describe what you did and then um how well you do as well as your description Etc can lead to prize Wy and it was on the order of 20,000 um dollars worth the prize 20y and came pretty interesting right um and that's yes it just uh it just ended so yeah it was a really nice uh kind of collaboration everybody knows the factors in equities like you know it's sort of like in textbooks it's like size momentum value uh beta all these factors uh are determin some of the stock returns but in crypto it's not actually really known what the factors

are and so by kind of uh doing this collab this sort of competition uh to find factors uh in partnership with ocean we got uh uh a great deal of new interest in that field and if we get good at that uh it helps us may do all kinds of things including have better scoring on numerai where we neutralize Factor risks from the from predictions and so on so um yeah it's been a been a good start to to doing something together because we always talk about doing things together but we haven't really done it properly yet yeah and one of the things like you know uh freck you had mentioned you know and pointed that a lot of nerize infrastructure is

centralized so Richard and I talked about this and realized for their business model there's less incentive to have it fully decentralized it just doesn't matter as much right and that's fine and if they if there was specific aspects where decentralization is really useful then cool now you've got you know the ocean predictor contracts that you can basic you know numerai can deploy and we can work with and that's a an obvious point of collaboration but interestingly because of the the slightly different uh business models and incentive structures it's not necessarily needed as much which is kind of interesting right and but that also points to then um um you know complimentary user Bas at least on the on the consuming side right and on the supplying side I think it's actually um you know people who become experts at

one can cross over their expertise to the other and vice versa right so um and that's that's I see what what what you know is happening for sure and at least at least one or two people and probably more um and will happen more in the future too so um yeah and but because that you know ocean is now doing stuff uh in the last sh with predictor much closer to what numerai is doing and numerai is also you know dabbling in in crypto it's the perfect time to you know find excuses to work together and and then then you know make make um extra high values yeah what tell us more about uh the the potential future collaborations that can of you Enis envisaging there's so yeah I would say

that trend has uh has uh stayed ahead of of multiple things so for example there's no one at numerai who's who's an expert at o Oasis uh which predictor users and you could see you know versions of numerai crypto or something like that where we might want to use that and therefore uh working with TR team on a project to do with that could be very very useful um and we would be sharing the right skill sets in some sense numer team used to be you know maybe half blockchain engineers and half website engineers and now it's like 90% uh trading and hedge fund uh execution uh engineers and and

researchers and so it's sort of like our the makeup of our company is designed around our customers which are our investors which we care uh the most about and have a fiduciary duty to be you know to put them first whereas some of the crypto stuff especially because we've been yeah yeah like sometimes we get excited about doing something in crypto and then the gas prices would be 10 times higher than we could we could do manage it and so on uh but I that's that's where I think it could work we he basically his team knows more about blockchain we might know a little bit more about Quant Finance yeah you guys certainly know a lot more about Quant finance and but we're learning and we're bringing in we

find ourselves bringing in more trading and try toy people too just to to buff up our own expertise um I think a good example of of collaborations and where this can go I had mentioned earlier in this podcast about prediction feeds as a new type of uh crypto building block right so we've got oracles chain chain link style where um you know the traditional definition of Oracle is actually a future prediction right but in cryptand uh it happens to be that uh a chain link style Oracle is uh simply a feed typically of crypto prices or of existing data right so um if you want to have forecasts or prediction on top of that perhaps you call it a future Oracle if you want to think about it as a crypto building block or if not future

Oracle then simply prediction feed right but once you have that then there's lots of um crypto products that can built on top for example there are quite a few um decentralized derivatives projects right where um people are building different sort of features trading strategies options and so on all of that could um could uh leverage uh decentralized predictions quite well right um especially once you know predictor extends to have continuous valued predictions and volume volatility which is one of the directions for us of course um beyond that too if you're you know doing a um a loone protocol um you could you might be able to reduce the collateralization ratio from instead of 180% maybe down to 120% um if you can uh predict um volatility better right

because then you need less time to react and similarly for St coins that are based on collateralization and then also for dexas uh where you're do the LPS are you know adjusting their their weights in things like balancer or adjusting their threshold in in things like the you know sudo order book he set up that unisoft has and so on all of these could leverage predictions in order to do a better job and of of course if you can run this stuff all as its own new protocol on top that combines these existing protocols plus this new building block of uh prediction feeds then you can get you know basically um much more interesting Behavior you know optimize all of existing defi infrastructure like I mentioned whether

it's uh derivatives or uh loans or stable coins or dexas and and also put the opportunity to to New Primitives that are higher levels so that's an example um yeah and you know ocean can do that on its own but then we with numerai we can pull in stuff from um you know imagine the numerai trafi stuff and the numerai expertise so there's a lot of back and forth there or you know numerai itself building some of these um things on top right so there's just um basically because of where predictor is the numeri is we've increased the surface area of possible uh technology artifacts right uh increased the size of the adjacent

possible if you will and from that you know ocean can reap rewards numer can reap rewards and any other um adventurous Builder so final question prediction on prediction feed in the next two years what's going to happen what are we going going to see that may be surprising I uh yeah I mean I must say I haven't followed um is it poly Market poly Market yeah I I was sad that orer uh didn't end up working but I'm very very happy that someone's uh taken the taken and they seem to be doing unbelievable job but I wish I knew more of the technical details uh but I'm very happy to see that because it's always

been for me it's like um uh stable coins and prediction markets and prediction feeds you know oh that's very very 2017 of you I know but it's like it's like what what's better than those things um and uh and so you know because those things actually catch AI in their web and that's what's so cool about them so um yeah that would be my prediction is that there's more there's a lot more of this and um it is the vector for AI to kind of Express uh block chain yeah no I I can explain to you how a poly Market Works uh because it actually uses nosis uh conditional token framework and we built it uh years ago

no way uh but that yeah but that that's that's a story for for another time okay fine there's probably another podcast about that one Trend uh what about you prediction on prediction feeds yeah I mean I think overall prediction feeds will find themselves playing a role in the broader world of AI you know mainstream has you know learned to what AI within the last couple years in in Spades right you know me and my friends would hang out all the time going back to the 90s talking about it right so it to me it's wonderful that a lot more people talk about it and what the future holds with AI and the philosophical implantations and societal risks and societal gains potential right and as these um uh AIS uh you know more and more powerful AIS get built um some

will be centralized some will be decentralized and both the centralized and decentralized ones are going to want to be using um the prediction feeds to to make better decisions right so it's not just for pure trading you know that's the N Stuff but for any sorts of decisions right already we're seeing you know like chat gbt and stuff hooking into different um apis to do what it does well there's nothing stopping it from looking into you know this API this stream of ocean predictions and num predictions potentially you never know um so there's that and but this will actually unlock a lot more power right because you can have ai systems if you have like super awesome weather prediction uh which I hope you know maybe not you know Global on a two-year

time frame maybe but 3 four five who knows um and incentives are a powerful thing um then uh with super aesome weather prediction it will you know have big impacts on um improving safety whether you're flying or driving otherwise and a whole bunch of other stuff we probably can't even predict but just sort of I macro level I see that prediction feeds will catalyze the AI space that's probably the biggest thing and I also see that um they can help to catalyze the defi Space by uh increasing the capital efficiency um and uh also the try space as Richard and the numeri team are showing and you know so that's some of the basics so improving AI uh generally uh improving the the FI the

five spaces fi and then beyond like I mentioned earlier you know what weather prediction Etc and one of my pet dreams is to have a super intelligent AI emerge by uh that is actually you know a single large model trying to predict the weather dynamics of every square kilometer on the planet right I think that'd be really cool to see um and by the way there's going to be another big side effect which is part of the point of predictor which is this will be one of the leading horses pulling along the whole decentralized daa economy right um you know now there's um already uh there's good money in making predictions right doing it at numeri or or ocean um but um then working backwards just like Richard had talked about there's this

descent this Market numer Bay for the models uh we can see more and more of that stuff and thinking thickening out the rest of this decentralized data supply chain um and so overall all of this stuff is going to catalyze decentralized AI um for really the large scale models and that's pretty exciting so so yeah that's where we're headed fantastic so where do we send people to find out more about predictor and numerai so numerai uh if I would the one I was talking about crypto I would go visit that one first it's uh crypto numeru doai uh SL meta hyphen model that's will give you what the model is predicting right now and you can

download the predictions uh and you'll also find a way to sign up there but there's also numerai uh for main numerai if you're a machine learning person or a data scientist without any data then that's great because you can download all data from there what about uh predictor yeah so uh predictor doai so predictor has two o's um and from there you'll see a table that looks a bit like coin gecko or coin market cap and you know scanning the table and looking at it you can sort of quickly grock um how predictor is behaving and so on and you actually get the Bitcoin feed or five minutes are free so um play with that too and then like mentioned earlier you

can go to the top right click on run Bots that will take you to a read for running um prediction Bots and actually trading Bots too so you can have your own trading bot running live on binance Etc um you know so so that's that's the main way to get going on predictor U going to predictor doai you Bitcoin wins free yeah so yeah maybe that's interesting I mean maybe someone's already using our API to predict on yours or vice versa and that's very interesting yeah I could see that Co thank you both for coming on that's been super interesting um and uh yeah have have a lovely evening thank you very much thanks for having us