Projects
![]()
No projects found
Investors
New
DAO Treasuries
DATs
Lend / Borrow APRs
Chains
Categories
Spaces
Ocean Predictoor live on Sapphire Mainnet: Oasis × Ocean × Celer
Recorded: Oct. 12, 2023 Duration: 0:45:49Space Recording
Full Transcription
Hey, hey. Good day, everyone. Welcome, everyone, to our Twitter spaces. I'm Michna, community
leader at Oasis, and I'll be today's host. And with me, I have three amazing guests from
Ocean, Seller, and Oasis Network. And together, we'll go over a very important launch on Sapphire
Mainnet. And that's Ocean Protocols Predictor. Predictor is an on-chain privacy-enabled AI-powered
dApp for making and verifying predictions. And during today's event, we'll be talking
about the way Predictor makes use of our smart privacy feature, how Seller integrates with
everything and much more. We have a very engaging discussion ahead, and I encourage everyone
to participate. So feel free to ask questions and share your thoughts in the chat or on our
community channels, as we'll be answering your questions live. So grab your seats, get comfortable,
and let's dive right in. And before I make introductions, I'd like to start off with some exciting updates
from our side. As you might be aware, our team recently had the pleasure of attending SmartCon
in Barcelona. We had the opportunity to meet with some fantastic projects interested in either
deploying on Sapphire or integrating with us via OPL. And we're thrilled to have made so many new
connections in the Web3 space. Some hot topics were privacy and account abstraction. So it's
always great to see the enthusiasm for what we're building here at OASIS. In other news, we have a
very exciting announcement to make. And that's that OASIS is supporting the integration of OPL
with the help of Ourai Chain into the Cosmos ecosystem. And this is super important because
bringing smart privacy primitives such as OPL over to Cosmos marks a significant step towards enhancing
confidentiality tools for both builders and users alike. So we're looking forward to the positive
impact this collaboration will have. Another recent development for our ecosystem is our
collaboration with Thinther. And that's transforming our native token to an omni-chain asset for any ecosystem
in the crypto space that presents a huge improvement for our ecosystem. And during Q4 of this year, Thinther
will be launching CIROS, which is a synthetic iteration of ROS, our native token. And this will be available,
of course, as additional liquidity on Illuminex. Illuminex is the first DEX built on OASIS Sapphire.
So we're very thrilled to partner with Thinther on this exciting venture. And last but not least,
our privacy for World 3 Hackathon has concluded. We've had a whopping 1200 registrations.
And now the judging period is underway. And of course, we can't wait to announce the winners later this month.
So we had incredible ideas and innovations from the Hackathon participants. And that's truly,
truly inspiring to see. But enough updates. I don't want to bore you again with updates. So let's get to
the heart of today's event. As I said, we have some distinguished guests joining us, representing three
incredible projects in the Web3 space. And they'll share their insights and perspectives on everything.
So without further ado, I would like to kick off with introducing our own Matei, BD Manager at OASIS.
So Matei, if you'll take it from here.
Yeah. Hey, guys. Hey, Meghna. Always a pleasure joining these spaces. So yeah, today,
we were joined here by Trent and Wanli to kind of kick off the mainnet launch of Ocean Predictor on
Sapphire Mainnet. So I'm hoping that there will be questions, you know, from the community to see what
they're interested about, you know, what are the maybe the questions that they need answers to,
how things work, why Predictor, you know, needs Sapphire, how it intertwines with
seller and all that. So maybe Trent, you can give us a bit more insight into into that.
That's it from my side.
Great. So I guess I will jump in. So hi, everyone. I'm Trent McConaughey, co founder of Ocean Protocol.
And I'm also been the one driving the team inside Ocean Protocol, building Predictor. So it's the same
team that has built ocean data farming, the VE Ocean stuff and whatnot. Ocean down in the past.
And yeah, for the last many months, the better part of the year, basically, we've been building
Predictor. And as of Monday of this week, October 10, we shipped Predictor. So I guess it's Tuesday,
sorry. So today, it's my pleasure to share with you guys, what that's about. When I say shipped,
I mean, on mainnet on Oasis Sapphire mainnet, four weeks ago, we shipped on testnet. So and given that
I'm mentioning schedules in four weeks and two days, we have one other release, which is starting
kicking off data farming for prediction feeds. So if you are submitting predictions in this crowdsource
setup, then you can view it as extra baseline revenue, on the order of $10,000 plus worth of ocean and
ROAS tokens. So I'm going to just jump into what Predictor is about. And then there will be a follow
up with our good friends and colleagues at Oasis and Cellar. So Predictor is motivated by the goal of
having prediction feeds. Prediction feeds is something maybe most of you haven't heard about, but it's
a time series that is always giving a value in the future. For example, you could have a prediction
feed saying, every five minutes, it will predict that Bitcoin will go up five minutes from now or
down five minutes from now. And there's an amount staked against it. And that's an example of, you
know, binary feeds up, down, etc. And by the way, that's what we've launched with is crypto prices,
top 10 tokens by market cap, up down every five minutes, as well as every 60 minutes. So that is
the baseline. But of course, we, this is the beginning of a much broader, bigger vision,
we will have continuous value feeds, going for other for DeFi tokens, as well volatility indices,
volumes, all of these, as well as going beyond crypto land into things like rain prediction,
wind prediction, etc. And, you know, we envision 1000s, 10s of 1000s, hundreds of 1000s of feeds
throughout the world, for various verticals from weather to climate, which is long term weather,
to energy price predictions, energy usage predictions, logistics, and more. So that's kind of
the vision. And at the heart of it, we believe that predictions are the most valuable type of data,
because once you have a prediction, then you can take that prediction, and immediately make action
against it, typically to make money. For example, in crypto price predictions, if you know that the
predictions are accurate, then if you know, the price, it says ETH goes up in five minutes,
then you could buy ETH five minutes later and sell it. And then on average, you'll make money,
right. And so it's alpha, in a sense, for predictions, I described a naive trading strategy,
of course, you can get fancier, fancier, with futures and other things, but this is a baseline.
So that's what predictor is about, where do these predictions come from? This, it's crowdsourced,
actually. So just like cowswap or Uniswap X or Numeri or other things, there's a group of people out
there, typically data scientists, who submit values to the protocol that then the protocol evaluates and
then pays based on how well you do. In the case of predictor, what this means is, every five minutes,
one or many predictors, people out there making predictions, they submit a prediction for, say,
the price of Bitcoin. Will it go up or down five minutes from now? And they also stake against it,
you know, one ocean, 10 ocean, whatever. The more they stake, the more confidence they are, confident
they are. Five minutes later, the true value comes along. And if they predict it up and it's up,
or if they predict it down and it's down, then they make money. They get a cut of the sales buying
that feed. If they predict it wrongly, then their stake gets slashed. So that's the baseline. And then
from the other side that people can, people can buy feeds. And this is sort of ocean's bread and
butter is these data assets. So in this case, it's data feeds. So you can come along, buy a data feed.
It's a ERC-20 data token. And in buying that, then you can start consuming it. And it's a subscription
to this feed for 24 hours. And then, you know, every five minutes, you'll see what the prediction is
for five minutes from now. And then you can take action as you want. So that's the summary. The
two main actors in the ecosystem are predictors. These are the data scientists making predictions
every five minutes. And the traders who are using these predictions, the aggregated versions,
the stake-weighted sum of individual predictions, the traders are using these predictions as alpha
for their own trading. And in both cases, typically people are running bots for this, right? So
inside Ocean, in the code, we supply off-the-shelf bots for both predictors and for traders.
But of course, people who want, they could integrate their own trading bots, et cetera,
as needed to. So that's the heart of predictor. We're very excited about it. This is,
you know, we're still at the beginning. Just this week, we launched it in production. But we are very
excited for the future of predictor and where it will take us. A couple of key things related to
this AMA before I pass it off, of course. When a predictor submits their prediction, they can't put
it onto a public chain because then it would be transparent. And then the calculation of the multiple
individual predictions to the aggregate, that also can't be public because, once again, others would take
it can use that without, you know, value accruing. And then finally, that final aggregated value,
every five minutes, it can't be public as well. Once again, you know, we want value to accrue. So
in all three cases, we need privacy. We need privacy for the individual predictions that are being
submitted. We need privacy for the calculation of the aggregate. And finally, we need privacy for this
final aggregate value. And the final aggregate value, it does need to get revealed to the person who has
subscribed, or the people who have subscribed. So that has a special affordance. So those are the
three privacy needs, needs we identified in Ocean. We actually did a lot of research exploring various
different approaches to this, looking at homomorphic encryption, differential privacy,
ZK, a lot of other stuff, too. And in the end, what made the most sense for Predictor was the approach
that Oasis Sapphire offers, which is Trusted Execution Environments, TEE, to enable a privacy-preserving EVM
chain. Everything in Ocean is EVM, mostly Solidity, some Viper. So that's why we went with Oasis. And,
you know, we've been iterating with the Oasis team for many months now on this. And they've been
really great to work with. And, you know, we successfully shipped on Testnet four weeks ago,
and two days ago with Mainnet. Finally, one other thing, I had mentioned that Predictor,
in order for the predictions to be accountable, you need to have some skin in the game. If you don't
have skin in the game, then people could make predictions, and it doesn't matter, you know,
you're going to have people gaming at all this, right? It's like the weatherman, right? If he's
wrong when he predicts whether it's going to rain or not, it's sort of no skin off his back, right?
But in this case, people, you know, we're in a very Byzantine environment, a dark forest. And so
everything will be, anything that can be gamed will be gamed. And so it's really important that
the token dynamics makes sense. What I mean, to summarize, we need to stake, predictors need to
stake in order for, so that the predictions aren't gamed, such that they, if they're wrong,
they're slashed, if they're right, they make money. And that happens every five minutes. So with that,
because, you know, we're Ocean Protocol, all that, it made sense for us to use the Ocean Token.
So then we needed the Ocean Token on Seller. Sorry, we need the Ocean Token on Sapphire Mainnet.
Because the, you know, Seller has some really great characteristics, notably that, unlike almost
all bridges out there, all multi-chain solutions, unlike most, almost all of them, it is actually
decentralized. And that makes a huge difference, given, you know, all the different hacks, et cetera,
we've seen on various bridges over the last couple of years. So the decentralization aspect is really
important. And by the way, a lot of those hacks were related to the multi-sig. So the fact that it's
more decentralized than these, in that aspect, helps to improve our confidence in the security and,
you know, the long-term decentralization. So, so the Oasis team with Sapphire had already went with
Seller. And for us, we are very happy with that choice. And so we're very happy that
Ocean Token can be bridged from its root chain, which is ETH Mainnet, to the Sapphire
Mainnet chain. So that's the two things that we're very happy to have these collaborators with,
as we have shipped Ocean Predictor, the privacy-preserving event of Oasis Sapphire, and the
decentralized bridge of Ocean Token from Ethereum Mainnet to Sapphire Mainnet, thanks to Seller.
So with that, you know, I think the future is bright. There's lots of opportunity for people
to make money using Predictor, either as predictors, data scientists making predictions,
or as traders, people who are using these predictions to trade and using the predictions
as elephant. With that, I'm, you know, done with my intuitive predictor and, you know,
the role of Oasis and Seller, from my perspective. And I hand back to the host and to our friends at
Seller and Oasis. Thank you.
Thank you, Trent. And we do love your vision. I've been recently reading your Medium article. So,
yeah, exciting stuff to come.
Yeah, let's go over to Wang Li Zhu, who's the Head of Ecosystem at Seller. Wang Li, could you give us a
bit of an overview of Seller and the details of the partnership with Ocean?
Yeah, definitely. Thank you so much for having me here. So, my name is Wang Li. I'm the Head of
Ecosystem from Seller Network. And as mentioned before, Seller is an interoperability protocol,
and we have supported both generic message passing, as well as asset bridging for the Oasis ecosystem.
And for the case of Ocean, we have supported token bridging for the Ocean token between
Ethereum mainnet and Oasis Sapphire mainnet. And we're also exploring whether there will be some
crushing messaging use cases for the Ocean protocol. But in general, like, we will strongly support the
interoperability layer for Oasis Sapphire to be connected to a wide range of blockchains that we
support. And as mentioned before, also, Seller is basically one of the most secure bridging solutions
in the market. And we're fully decentralized. We use a proof-of-stake layer as the message
relayer for crosschain messages. And there are currently 21 validators on the proof-of-stake
blockchain that we run. And we have facilitated over $15 worth of value in crosschain transactions
across more than 40 blockchains. So, we're really happy to have supported Oasis Sapphire as well as
Ocean. And we're looking forward to building more use cases and have users or projects utilize Seller
Interchain Messaging, the generic message-passing protocol to build their privacy-focused use cases.
Yeah, thank you, Wanli. If I can just jump in real quick. So, I really like, you know, the emphasis
that both Wanli and Trent didn't mention it today, but he mentioned in previous topics about
decentralization, right? So, you know, Predictor could use a survey solution for the privacy
aspects, but I really like, you know, that they support the ethos of, you know, permissionless,
decentralized solutions. And currently, yeah, Sapphire is probably the easiest privacy solution
on-chain to integrate with. So, yeah, I really like that we are building that ethos, you know, and
every project is basically doing what they do best, right? So, it's not that one project tries to do
everything, right? But Ocean's the best data, right? So, they're doing the data aspect of it. We focus on
privacy. We do the privacy thing, you know, and Seller does the bridging and this is the composability
that we like to see in crypto, you know, where you take the projects with what they do best and combine
together to get an awesome product in the end.
Thank you, Maten. Thank you, Wanli. We do have some questions from our community because we've been
sharing a feedback form for this event. So, yeah, I can start off with the first question and that is
if you can explain in more detail how Predictor, Oasis and Seller collaborate to create a secure and
private prediction platform in the Web3 ecosystem. And I guess this question is for everyone. So,
who'd like to take this?
I think Mathe summarized it well, but I'll paraphrase once again, too. You know, Ocean is building
the data layer in the focus of tokenizing data feeds and data assets and other cases as well.
But here it's really focusing on the data feeds, the prediction feeds for Predictor. So that's what
Ocean is bringing to the picture. It needed privacy inside an L1 and that's where
Sapphire came in very well. It's the only privacy preserving EVM and production out there.
So that's obviously a very straightforward, obvious thing. And then, of course,
Seller is the means to bridge the Ocean tokens from ETH mainnet to Sapphire mainnet in this case.
And I think all three projects have a very strong ethos around decentralization. And so
that really matters in the case of the bridge and the other aspects, too. So that's how I see them
fitting together. You know, Ocean with the DAP and the stack that, you know, the higher level stack,
the lower level stack, the L1 that's Oasis Sapphire, and then Seller for the interoperability with ETH mainnet
in the case of Ocean tokens and ROSE tokens, I should mention.
Yeah, I would echo here, Trent. So the security comes basically from, you know, all three solutions
being very big on decentralization of censorship resistance. We are very serious about the product
that there are products that we launched as well, you know, on the security side, us as well, you know,
keeping the private data really private, you know, separating, you know, the compute nodes from the
key manager nodes, the key manager forum, you know. So all these aspects tie in together.
You have like a really secure decentralized solution, you know, that now is used for price predictions,
but I believe that the vision of Trent, you know, basically the whole Ocean team to bring more
predictions, you know, from the real world in here, it's gonna be quite a game changer.
Thank you both. And I have two questions that kind of combine each other. One is for both Seller and
the Ocean team. Like, do you have other plans to use Sapphire? And the second one is how does Oasis
protocol benefit Ocean other than maintaining confidentiality of on-chain data? But yeah,
they kind of intertwine each other. I'll pass to Wanli from Seller first.
Yeah, definitely. So for our upcoming plans or strategies for Sapphire,
since we have already added support for both asset bridging with Seller Seabridge, as well as
our generic message passing layer, Seller Interchain Messaging, the next step would definitely be to
build use cases within the ecosystem and work with good projects like Ocean to create asset bridging
use cases, as well as cross-chain message passing use cases. And I think in terms of the privacy-focused
positioning of Sapphire, we would love to see more protocols trying to utilize Oasis Sapphire to
build some cross-chain privacy use cases like cross-chain features that uses interchain messaging,
but as targeting the privacy benefits that Oasis Sapphire can offer. And in terms of upcoming plans with
Ocean Protocol, I think we would be more looking at how we could facilitate the cross-chain messaging
part that Oasis Sapphire is offering for Ocean because we're basically right now like the largest
were only message passing layer connecting Oasis Sapphire to external blockchains. So how to basically help
Ocean build their crushing features and also how to promote their features to both communities as
well as the Sapphire community would be something we'd love to help with.
Great, yeah. And from my perspective, Ocean overall, its mission is to level the playing field around data
and AI, right? AI has some pretty centralizing tendencies, you know, to have large amounts of data to train
your model, et cetera. As we've seen even recently in the last six months, 12 months with these large
language models. So Ocean really has this overall mission to help, yeah, make it easier for the little
guy, for everyone to have more equal opportunity to get the benefits of AI and data. So within that,
right, Predictor is the most recent application with protocol goodies below, but it's part of this
broader stack that Ocean has, which is at the heart of it, we tokenize access control. So, and that's
access control to CSVs, to feeds, to APIs, even to web apps. So some other specific examples are things
like token-gated APIs, yeah, token-gated web apps and more. And then we also make it easy to build things
on top, such as decentralized data marketplaces. We have an example with Ocean Market, but there's other
set of people that have built for other ecosystems, such as Daimler built one for their customers in
a spin-off called Decentric and more. So that's what Ocean does. And then to the question of how would
then these solutions in Oasis Sapphire and Cellar extend? So first of all, basically it's all the other
functionality of Ocean can benefit from both things actually. So for data tokens to share access to
that and when not to extend the functionality of data tokens and data NFTs of Ocean, there are
affordances from the privacy that you get, you know, extra features that you get that people find useful
related to privacy, of course. So that's maybe an obvious thing, but it counts and it's pretty good.
Another one is, because we have, you know, these data assets, such as data feeds, they are tokenized,
they are ERC-20. Like the copyright is a data NFT. And then for each data NFT, you can have one or many
ERC-20 data tokens. So those things could readily be passed from network to network as well via seller.
So to me, that's, you know, an obvious place to extend this. And I think it'd be pretty neat to see,
you know, someone buy, say, a data feed for Predictor in Sapphire and then, you know, bridge that to
Ethereum and have it in a Uniswap DEX and a Uniswap pool on Ethereum, right, as an example. So there's a lot
of different possibilities. I think that's the summary. Basically, Sapphire to extend for the ocean stack,
the broader ocean stack to leverage the privacy features of Sapphire to get special benefits
around non-custodialness and so on. And then seller for bridging beyond the ocean token, bridging all
the other tokens, the hundreds, the thousands of data tokens that emerge from the ocean ecosystem.
Thanks.
Wow. Thank you for that. And sorry for this trend, but I do have another question for you. And thank
you, Arnal for the question. Will there be gas fee prediction functionality too, so that we can know
when it would be most cost efficient to make transactions?
That's actually a great idea. And in fact, we've had people suggest that in the past. So it is on
our list of possible things to add. And if anyone has more suggestions for things like this, absolutely,
right? It is, you know, we basically want to have feeds that people find valuable, right, that people
want, would buy and whatnot. So definitely, that's a good one. And I know we do practice that too. So
certainly, certainly, you know, other feeds too, things like volatility is very useful if you're
doing options trading, because then you see, you know, not just whether up or whether down, but
how much up or down, right, as an example.
I hope that answers your question or not. And thank you for the question again.
I got word that we do have somebody who wants to ask a live question, if we can get them on stage.
Hi, Biba. Hi, Biba. Do you have a question for us?
Oh, yeah, sorry. Yeah, I was just wondering, when will we be able to use predictor for anything
else than price feeds? I want to use it for sports predictions, in particular, but yeah,
for anything else besides price feeds?
Oh, yeah, so we are focusing on the vertical of DeFi to start with, you know, to get strong,
to prove it out much more. It's general enough, of course, to have much broader feeds. You know,
I mentioned, we started off with binary feeds every five minutes, there's two generalizations in terms of
the functionality there. One is continuous value feeds, and one is, say, every five minutes,
predict 60 minutes ahead, or every minute, predict five minutes ahead. So those are sort of two
enhancements to the core feed technology. So we do plan to support feeds beyond, but we don't have a
specific timeline yet, given that we just launched in mainnet, you know, this past week.
I do want to mention that predictor, there's prediction fees and prediction markets, and they
are different things. So it depends, and I'm putting this in the context of sports. So predictor
will never have a prediction of, you know, will the, you know, Edmonton Oilers win the hockey game
next Tuesday night, right? It's not about that. That's a prediction market, because that's a one-time
event. You know, will Joe Biden win president in 2024? That's not for predictor, that's for prediction
markets. What predictor is, is prediction feeds. So, you know, will ETH go up five minutes from now,
yes or no? Will Bitcoin go up five minutes from now, yes or no? And then every five minutes,
it updates, right? And then there's a payoff directly. Another key thing is, within prediction
markets, there's speculation on the predictions, the speculation in a market setting, back and forth
until the thing happens, right? So there's, you know, prediction markets right now for Joe Biden,
and some people are buying the token for yes, and some people are buying for the token for no,
and that keeps changing over time. And it will eventually converge to the true value,
and that was sort of one second before the announcement is made, right? In the case of
prediction feeds, it's set up where in the five minutes before the prediction comes true,
no one can see what the predictions are. So except for the predictor. So if I'm sorry,
I'll be precise. So I'm a predictor. I want to predict the price of Ethereum five minutes from now,
I actually have to predict, you know, five minutes, one second or five minutes, 10 seconds from now.
And then starting at five minutes from now, I have to, I can no longer submit my predictions
for that five minute thing. And in that five minute window, the traders can buy and see the value,
right? So there's no speculation back and forth within that five minute window. Whereas within
prediction markets, there is speculation. But this has the extremely good benefit of
the accuracy is fully dependent on what actually happens, right? And it's, you know, pretty often,
right, every five minutes, every 60 minutes, whatever, right? And so people can't speculate,
what will the other guy guess, etc, you get paid, if you're accurate, you don't get paid,
if you're inaccurate, as simple as that, rather than in prediction markets, I think prediction
markets are very cool, by the way. But you can get paid if you manage to trick everyone else into
thinking the other direction, right? Here, it's, you know, much more focused purely on accuracy,
rather than what is the other guy thinking. So to summarize, predictor to your question,
will predictor be applied to things like sports, it could apply to things like sports,
such as, you know, what is the number of cancellations that will happen in a given
in week X of hockey, you know, hockey season, and that would happen every week, for example,
right? Or what's the, you know, other things like that, that are happening on a regular
interval, a time series, but it's not for one time things, that's really what prediction markets are
for. Thank you.
Biba, any follow up questions? Or are we good?
No, no, it makes perfect sense. It's not possible for predictor. Thank you.
Thank you, Biba. And I have a question for one Lee. One Lee, could you provide more details on how
sellers technology helps in enhance the scalability and efficiency of predictor?
Yeah, I think sellers, like benefits where technology will mainly align its interoperability
features. So, for example, for crushing messaging, since like, we have, we have enabled that for right
now, Oasis and Ethereum mainnet. But for example, if predictor is trying to also attract users from
other external chains, that's also viable. Additionally, for the privacy
features or functionalities, seller is also helping another partner protocol called Redbus to
basically launch zero knowledge initiatives that could build privacy preserving crushing bridges,
or privacy preserving features for protocols. And we can also explore adding that functionality
for Oasis Sapphire. And I think that would also be like a viable solution for predictor when they are
trying to utilize the privacy attributes of Oasis Sapphire. So in general, like our help or our support for
predictor come in two folds. I think one is the technology side from interoperability, where we can help
predictor reach more users or easily reach users on Ethereum or on other chains. But aside from that, it's also
the potential zero knowledge capabilities that seller is building with Brevis that could potentially
help with Oasis Sapphire overall, but also we can explore whether that can be something to start with for
a predictor.
Thank you, Molly. I think that answers the question perfectly. And I do have one more question, and that is for Trent.
What skills would an individual need to become a predictor on your platform?
That's a great question. So overall, it is biased towards people who are data scientists, AI people.
And maybe I'll just walk through the sort of what the steps would be. And you'll kind of see what I mean.
So if you're interested in, you know, making money by being a predictor, the first thing you would do is
you would go to predictor.ai. So it's P-R-E-D-I-C-T, predict, O-O-R, so two O's, dot AI. So predictor.ai,
you go there and just like look around the page, see what's going on, get, you know, kind of help to
build your mental model of what it's about, right? You'll see 10 feeds for the five minute ahead
predictions for, you know, Bitcoin, USDT on Binance, ETH, USDT on Binance, etc. You'll see what
the value was, you know, 15 minutes ago, 10 minutes ago, what the prediction is, right now,
we have one free feed. And then you'll see that for each of the other coins, but the other coins,
you'll see a buy button. So because those ones you have to buy, it's three ocean for one day. So
that's about $1 worth for one feed for one day to give you a feel. So, you know, play with that,
get a feel. That's the first thing. And that's for anyone just to get a feel for what predictor is
about. But then for the next step, the next step is, you could jump into running a prediction bot,
a predictor bot directly, but we do have a nice intermediate step, which is challenge data farming.
And this is also linked from all of our docs, etc. And in it, what you do is, at Wednesdays on
midnight, you submit predictions of what is the price of ETH at 12.05am, 12.10am, 12.15am,
all up to 1am, and for the next day, right. And so just a small handful of predictions total,
but you don't have to run a bot or anything, you just have to submit them using a bit of ocean
infrastructure, data entities and stuff, it's pretty straightforward. And, but in doing that,
so you're not having to worry about running bots, but you have to think about,
how do I make accurate predictions? So where do you get historical data, for example,
for the price of ETH, etc. And we give off the shelf examples within the challenge DF repo
of how to grab data using the CCXT Python library, which has its own links to the Binance API and
the Kraken API, etc. So you grab your historical data, it shows you off the shelf ways to build
your own prediction models, etc. And once you get good at that, and that's very much an AI thing,
right, you can use off the shelf code. And so even if you're not, you know, deeply skilled in AI,
if you're, you know, pretty comfortable with Python, you can make your way through probably
fairly well, right? If you want to have a big edge, the AI ML stuff helps, but you can learn
that over time too. So that's the, so the first step, go to predictor.ai. Second step,
do challenge data farming, you know, once you're twice to submit, get a feel. Once you're comfortable
there, then the third and main step is to become a predictor. And for that, you go to the ocean repo called
PDR back end off of predictor.ai. It's a button on the top right called run bots, you click on that,
it takes you to this repo, you walk through a readme on how to be a predictor. And the heart of it is,
once again, it shows you, you know, how to get going running the well, how to get going running the bot
starts off with a random bot, you know, one that makes random predictions. But then it shows you how
to then start using model based bots as well, bots that are using AI models to make predictions.
So, you know, you can get going, doing that without any AI ML knowledge, you just, you know,
you can get going like that. And then, you know, as you build up your AI ML knowledge,
then you can gain an edge to earn more money. Yeah, right. So those are the main things. Step one,
two, three, you know, predictor AI, challenge data farming, and finally running a predictor bot
via the run bots button. And it's this ladder thing that, you know, that's the best earning
potential as a predictor. And like mentioned earlier in this AMA, your earnings are a function
of your accuracy, as well as the sales coming in to people buying these feeds. And, you know,
right now we just launched. So the sales are still small, but they're going to grow over time,
we're very confident. And also as a baseline to kickstart the sales, starting one month from now,
we will have one, uh, about $10,000 worth of ocean tokens every week, um, being paid to the
predictors. Um, and that acts as sales basically. So it's the same mechanics as, as if anyone else
was buying the feeds, but that means you have a baseline market, if you will, where if there's
say four predictors in any given week, you know, four different parties, um, running predictors,
then each one would earn on average with $2,500, right? So that gives you a sort of a baseline
incentive to know that you can earn decent money doing this. And of course, um, as time goes on,
you can earn more, there's going to be more and more sales as people discover that this,
that predictor is a really great source of alpha. Um, and then, you know, it's this positive feedback
cycle, the snowball effect, right? Um, people discover that predictor is a great source of alpha.
Um, so they started buying the feeds more and then there's more and more sales coming in and that
incentivizes more people to become predictors. It incentivizes the existing predictors to increase
their accuracy. And it's, you know, this positive, uh, snowball effect. So that's the summary,
the skills. So to summarize, uh, you, you have to know Python well to be, uh, you know,
to do well in predictor, uh, although I guess you could, if you want, you know,
write something in, uh, JavaScript or otherwise, and just, you know, submit to this more contract
directly, go ahead. We have all the examples of Python, but you could, so that's the baseline.
You have to be good at Python and you can get going that way. But if you won't really want
to have an edge, then the more AI ML, you know, the better off you'll be. Thanks.
Well, there you have it, folks, uh, a good incentive to brush up on your Python skills.
Just head on over to predictor.ai and, uh, get started. Uh, okay. Before we wrap up,
I do have one more question for one Lee. Uh, and that is what the current status of
brevis is and when it would be possible to integrate with Sapphire.
Yeah. So for brevis, um, it's mainnet ready for, um, Ethereum, BNB chain and Polygon. And since
Oasis Sapphire is also, um, EVM compatible, we could foresee that adding support later, um,
would be relatively easy compared to non-EVM chains. And, um, we have both, um, a solid SDK and
a set of APIs for, um, developers or protocols to integrate with. And for, in terms of the use
cases, basically brevis works for any kind of, um, data rich computation, that's like basically
impossible to do because of the, um, prohibitively, um, heavy, um, time because that need to be
consumed and also the cost on chain. And, um, if protocols want to still utilize the rich data,
including, um, um, um, to, to basically drive analytics for users who are for certain features,
or basically, um, run certain marketing programs or, um, user acquisition programs,
that's all viable if we are migrating that happy computation to off chain. And brevis does that
through, um, zero knowledge proofs, uh, where brevis runs arbitrary, um, data computation off chain,
and then submit a succinct proof on chain. For example, it can submit a succinct proof to Oasis Sapphire,
uh, once it's launching, uh, it's launched on Sapphire and any user or a DApp or a smart contract on
Sapphire would then be able to easily verify the validity of such a, um, proof easily with just
one transaction. Um, and that proof can basically, um, be mapped into 5,000 transactions where users pass
history throughout, um, a time, um, time range of five months or so. So, um, we're trying to, um,
basically see if there are solid use cases on Oasis Sapphire for us to prioritize the integration,
because, um, right now, like the tech stack is, um, there with like integrations, I mean,
SDK integrations readily available. It's just a matter of like what use cases we can, um, like build for
Oasis Sapphire so that projects where, uh, protocols within the ecosystem can benefit the most from
them. Um, cause otherwise, um, it's kind of hard for projects to start from scratch and utilize zero
knowledge proofs to build their own features, um, that's utilizing Brevis. So we would like to, um,
see if there are, uh, certain use cases that are like, um, from good projects, but also might see a lot
of adoption, um, on Oasis Sapphire and we would love to commit certain engineering resources to make
that happen. Um, in terms of the timeline, it's, um, ongoing conversation, I would say, with that ecosystem.
Thank you, Wanli. That was super helpful. Um, yeah, so, uh, that's kind of it for today. Uh, again,
thank you everyone for joining. Thank you to our guests. And before we, we end the call, uh, if, uh,
anyone has closing thoughts, uh, I would give over the floor to you.
I can just have a quick closing thought. Um, if you're interested to earn money by predicting or
trading, then please go check out predictor.ai and thanks to our collaborators in Oasis and in
Cellar for helping make this happen. Thanks. Thank you. Yeah. I would just echo, uh, Trent here,
uh, go try out predictor, you know, play around, uh, with it, you know, see, see how, how Sapphire also,
uh, integrates in the back, uh, and, you know, uh, bridge some tokens over, try it out, uh, and be,
I mean, I would say, uh, test it out now because in four weeks, you know, there will be some tokens
to be made, uh, in the data farming challenges. So yeah, it's going to be worthwhile.
Yeah. Grab it while it's hot. Uh, yeah, it's been a truly enlightening discussion. I hope everyone
had a good time. Uh, hope you enjoyed our updates. So yeah, stay tuned for more exciting developments
from us from our side and, uh, Ocean and Cellar. Uh, keep an eye out for future events. Yeah. Have a great day,
everyone. Thank you. Thank you. Thank you. Bye. Bye.
Bye.
