J2 AG 5 Lama Mansour prediciton markets

okay hi i'm i'm lama from gnosis and i'll be talking about prediction markets which is the core product diagnosis is building so essentially prediction markets or the idea behind them is to make the potential outcomes of a future event tradable for the purpose of eliciting aggregating the general like collective wisdom of the crowd the collective opinion of the crowd about a certain topic in the future so traders with different beliefs can trade on different outcomes of a potential future of a future event and the market prices that these trades achieve represent the collective wisdom of the crowd or the crowd source source wisdom or the general belief of what the crowd thinks the future outcome is going to be and this is find like this is actually a great tool for forecasting future information and for understanding the information like how future events gonna evolve and actually from a research perspective it like it shows that prediction markets are often accurate if traders who trade on different outcomes have access to diversity of information and are making this is their decisions independently and in the form of decentralization and this is actually what agnosticism where these elements can be provided to provide as much as much transparency as possible for prediction markets to be accurate and incredible so how do prediction markets are on a high level work is that someone who's interested in a certain topic or understanding what the outcomes of a certain topic in the future will pose a question for example what will the medium gas price on April 1st 2018 be read the question cover all the potential outcomes for this question the potential outcomes can either be binary yes-or-no or a list of like I can't a list of different outcomes that one of them will be true in the future or a scale in like mostly in questions about prices it will be a scale and then with like an upper and lower bound and the resolve value will fall somewhere on the scale the market the creator of the market needs to also fund the market was with liquidity and then open the market to the public so they can trade on the different outcomes and then there will be an Oracle which captures real-world information that will resolve and set the outcome of the market out of which the potential the outcome will be known to everyone traders who traded on the correct outcome will turn a profit those who in corrected the the wrong outcome will turn a loss obviously the mechanism or the market mechanism that prediction markets on gnosis follows is an automated market maker so any person who creates a market needs to fund this market to provide liquidity and this liquidity will be used to determine the market prices of different outcomes depending on the collective traits of everyone who traded on this market thus far and this automated market way cart maker always also provides liquidity so it's a great thing for anyone who wants to trade it anytime there it's always offers the chance for for trading it also sets fair prices and determines the payouts at the end when the market gets resolved so it's what determines the the payout distribution and it's also a fair way of determining prices because it's hedges against a lot the huge financial losses due to the fair pricing mechanism so a person who funds the market needs a can or any market can be funded with any ERC 20 token and the same token needs to be used by traders who trade on this market and for each collateral token or any ERC 20 token that the market is funded with and then the traders trade trade with they get a complete set of outcome tokens which are which represent the different outcomes or potential outcomes of the market after which the user or the trader can use I can choose which outcome they want to keep and then the remaining ones will automatically be sold back to the automated market maker it's a very from a user perspective it's a very sleek in quick process of just trading directly of the collateral token of the market and getting back outcome tokens that represent what a trader believes is the correct outcome the collective or all the outcomes in a market need to add up or the prices of the outcomes in a market need to add up to one so at any point of the market the different outcomes always the price of different outcomes always falls between zero and one and then at the end when the market gets resolved if it's a categorical market or like we have a list of different outcomes the the winning outcome will a price will be one and the losing outcome will be zero and this is how the payout distribution is is calculated for foreign markets this is how like in in general this is how the market mechanism of prediction markets work if you have any questions about them you can feel free to come ask me later about how like the the mechanism works in more detail so prediction markets are like a great tool for so many business cases they're like endless some of one of the most important use cases are information discovery or using prediction markets so you can forecast information about the future that'll be incredibly significant for organizations companies stakeholders in a lot of topics so for example in construction projects if we incentivize people who are experts who are have insights into for example a construction project to come forward and talk about their beliefs on our expectations or on when a project will be finished it can save up to millions in research and cost and cost planning logistics planning and for structure planning and these are this information as I said can be really accurate if the experts are well incentivized and you can actually use this information to sell it to two stakeholders or people who are interested in this information another really great use case for a prediction markets is governance or a decision-making in a very decentralized democratic way by making prediction markets about the per for means of a certain entity or person in a certain position and actually this type of use case uses another format of prediction markets which is conditional markets where you can like predict the performance of a company or the performance of like the revenue of a company based on if for example we keep the CEO or we we fire the CEO so there is a condition on which the decision-making will be done and then conditionally there is a scale of performance that is met that's used to measure the performance or like a fair way to make the decision on in the organization and it can be used not just for like governance decision-making in in different and different topics another also use case can be insurance in hedging by predicting the probability of a certain like future event like a natural disaster or an accident we can actually hedge against the losses that can be can incur out of the side effects that this disaster might might incur these are some of the use cases that prediction markets have and there are like many other endless use cases and gnosis as a platform would like to see all the different use cases leveraged on its platform so we have we provide this infrastructure this market mechanism better I just explained that all the smart contracts and we'd love to see them other teams developers companies building endless awesome use cases on top of gnosis with these use cases or any others what like - one way for us to demonstrate the different use case of prediction markets how they work how traders can can leverage them how they can use them was Olympia which is a test run that we launched in December it's a prediction market tournament that runs on rinkeby source and test network it was great for like to get users to interact with prediction markets because there is no financial entry of barrier and no financial losses incurred users can basically register connect to wallet they get play money tokens and then they get access to a wide range of prediction markets on different topics they can trade on to get a sense of how prediction markets work how the mechanism works different use cases them we tried to offer markets on the cover different opportunities and use cases prediction markets and actually then users can get rewarded at the end with with G&L tokens for their correct predictions so basically the the interface that we had was was very similar to them the general prediction market platform that dosis offers with the exception that it's on a test network and its uses play money tokens and actually it has additional gamified features for users to interact with the platform's they have like a list of markets you can interact with and there is a scoreboard where users who performed well and the prediction markets can get GN or award payouts we launched this test run of Olympia in December 2017 it was a first release of Olympia our predict and old prediction markets in general and got like it was a really nice test run we got more than 8,000 transactions in the two weeks that the tournament trainin with more than 700 users and we actually got a lot of good engagement so we decided to make this a recurring event so we will starting from April will start running monthly Olympia's monthly tournaments or competitions where users can actually interact with prediction markets and test them all the time and it's a good opportunity also for users to when G&O tokens at the end there will be also a package provided for any company or team developers who want to roll out like tournaments and try to experiment with their own Olympia's in different use cases they can also roll out their own Olympians on another note we also for the entire prediction market platform will have the SDK package for the full dosis prediction markets available and live next week and we'll actually launch it through a competition that we called gnosis X where different teams developers companies can apply with different ideas about or different projects and use cases of prediction markets to a panel and notice that will evaluate the projects and will fund the winning projects at the end if you'd like to know more about it we have our team here the awesome team of gnosis are in and they're also have a booth with more information about gnosis X and how you can leverage the gnosis X to build awesome projects on prediction markets yeah this is basically it and yeah thank you so much have any questions yes I was wondering and you have a standard for Oracle's essentially ourselves were not an Oracle provider so you can any for others who built on top of gnosis they can connect any Oracle they want basically and for events where there is no data point is there like a community base or a co hole or something like that no with no data like there's no API Oracle available for example when you don't know with the construction of the Berlin Airport there's there's no website which says ok it's done or it's not done a person creates a market they need to define a source that's credible that the Oracle will be resolved from because users who trade on the market need to know that the provider of the resolution is going to be credible so it could also be a person it could be a person okay thank you you probably get this a lot but can you quickly explain the difference between hugger and grossest so auger uses an order book style for exchange so we're like the trading of different outcomes they is based on like supply and demand so if you want to buy an outcome on a event someone else needs to be selling the same outcome and and at a price that you would like to buy gnosis uses an automated market maker as I explained the mechanism where this market maker always sets surprise for everyone all traders - to purchase talk like trade at the same price and it shifts according to the different trades how do you govern the trades that are made when for example there's you you there's someone who makes a certain certain prediction market based on the price of some financial coin or financial distribution and they can put in more money that it will win on the trades then the the price of that financial occurrence your institution will be so is there a way to govern what trades are made based on persons who can make money laundering or money making schemes basically it's free access for everyone to trade and if you believe in like prediction markets and the mechanism of prediction markets work which is like it has a free access for everyone and anyone to trade it eventually like achieves fair prices and fair predictions for everyone because because the the power of the crowd will always be bigger than that one person who's trying to manipulate the market so every person their stake is based on just a single person and not on the amount of money that is put in for example the amount of tokens money that you cast in the prediction market affects the market prices yes it does so so in that case it couldn't be that a certain person who puts in a lot of money can wage their trade off when he also manipulates the outcome of the event this person like would would overcome all the market but then as far as the market is open for the crowd for anyone in the world there will always be another person who's incentivized to correct the market and and predict predict fairly with a fair amount of money - okay so yeah as I mentioned there are different types or different use case for prediction markets if you know that there's one person that can influence the result and can really benefit from this and the prediction market essentially is an incentive market yeah you can't set an incentive for someone to do it if it's not if there's no the intention to incentivize them then you should probably not create a prediction market on it because then obviously this person can just mean manipulated so I guess one thing how you can prevent that is you know registering your traders or your predictors are you doing any kyc or logging of trades I'm currently not last question well one of your goals was to cover all the possibilities yes and if we look at the possibilities of outcomes there are many outcomes and it is really difficult to make for an event when there are a lot of black swans and so on so if you have binary markets have you ever thought about making non binary markets if it's possible so basically if we have a limitation of up to having up to 5 outcomes so if the potential outcomes of an event are more than 5 we we recommend that having like 4 of them the 4 main ones and the fifth one to be other to cover all different outcomes but essentially yes they're like all the potential outcomes need to be covered otherwise it's it's not a fair prediction market as illegal I would like to ask you how do you avoid to fall into the baiting regulation which is actually quite strict around here how do you avoid - and how do you avoid the is a bit so baiting regulation like is a different gambling gambling regulation sorry yes I mean I mean how do you avoid because this kind of studies could be considered as a kind of gambling phone interaction yeah we are not a gambling platform and so basically it's the whole incentive prediction markets is information discovery which doesn't fall under gambling and under a a game of luck it's more of a game of like it's a skill so you need to have skills to be incentivized to participate in prediction markets and we are regulated under distributed DLC framework and in Gibraltar that for that purpose but for not for gambling now for other others who want to build on top of gnosis the obligation to be regulated for their specific use case falls upon them not on us as you said before anise it is with prediction markets there is a crowd needed for that so how are you taking care of that there is a crowd actually there's not like five people trading between each other and that the prediction makes sense because then he needs is kind of a hand and neck problem that you need people to make the to make the protection market and then you need people to bet on that ability and reach out we we have like it's it's a free access for everyone to trade on the markets and for everyone who is a plot like as a platform provider we want others who build applications for prediction markets on top of Moses to target their own audiences I know you want that but I actively inviting people to do that because wanting it is one thing and building a platform for that being be also wanted people to host other people but they started doing certain things so I'm asking whether I don't know like a hundred people that are going to launch or companies are going to lunch prediction markets or like what's what's the plan we're mostly focusing on being a like a b2b platform so we want others to build applications and and take care of this part are you able to go into the specifics of like what algorithm you're using for the automated market-maker and what are the properties for the market creator like I know a few of them have like bounded losses but presumably there's no profit for the market creator usually that yeah the market creator doesn't get like profit out of this the fun thing that that's casted in the market maker always gets lost basically because it's provided it's offered to provide liquidity for the market a market creator can have can offer it can levy a fee on the market though for each trade for the transactions that are made and this is how a market creator can choose to profit from the market oh sorry just really quickly and like those those fees you wouldn't see as like a distorting the prediction itself like wouldn't that introduce like a small amount of noise sorry I so wouldn't like a fee on like in each individual trade like who doesn't have two major disadvantages it doesn't it encouraged number one parasitic markets because then there's no necessary reason to trade like on that market itself which like defeats the purpose of the prediction market because then the pricing information isn't useful and then number two wouldn't that actually introduce like a little bit of noise to the actual number that even like assuming that the parasitic market isn't a problem when that introduced like a certain amount of noise to the actual prediction itself because the market the person creates the market is funding the market so they're actually investing in this in this market affair like a big amount and providing this opportunity of the platform for others to and for incentivizing others to make predictions the incentive for others to get like to get financially incentivized through making predictions to be bigger than whatever fee the market the creator of the market is levying on a market and it's like their rightful right also to get this fee since they're offering this market so what prevents multiple contracts from being listed so if two events have the same contract yeah that's my first question and the second question is what's the Oracle process like if I miss that initially sorry I'm just answering this because I wrote the smart contracts so basically there's nothing preventing others to create to copy the smart contracts and try to create the same but obviously the goal that we have is to create a global liquidity pool so there should be only one market and we have to make sure that our basically those that are creating markets on top of us they have to make sure that there's visibility for their markets and that there's a incentive for everyone to participate in their market so there's no guarantee obviously and everyone can do this but the goal is if you want to have the best predictions and of course we would like to have a global liquidity pool also coming back to the market maker questions so the automated market maker is based on the constraints that the theorem currently has like unchain other books they don't really work at all they are often order books but they suffer from front-running and nothing is really scalable so we decided to go for an approach where we can show that prediction markets actually work and automate market makers are a great tool for information discovery it's right that there's one person or can a crowdfunded and what they're actually paying for is the information that they would like to receive right so they're interested in a specific piece of information so they have to get this funding for the automated market maker to set an incentive for everyone who has information to trade on this market if they have invested a better estimate then then what the market currently predicts that doesn't mean that we don't want to offer other market mechanisms and we're also working on a decentralized exchange and of course we would like to leverage this exchange also for the trading of the outcome talk into that create aa prediction market but for now we start with this automated market maker which is a great tool to just show that prediction markets actually work I just interested was there all of the gnosis token in the system like what's the trait of it having difficulty hearing the questions actually what's the role of the gnosis platform token in the system what's the we also as a platform provider we have a platform fee for create like for trading on markets the the fee is levied when the trader of the market creates a set of outcome tokens using collateral token according to the mechanism I explained when the and the fee starts with 0.5% of the amount that traders are trading this fee can be levied in any in the collateral token of the market or it can be paid in owl tokens which are the utility tokens like that can be created through locking down Gao tokens so nobody is forcing people to make great market with your token right so yeah and what would be your reason to use token to create to market reason to create the markets using tokens any user can use any basically unique utility advantage would be that the if the more fees are paid with Gao tokens the less the fee becomes so there's a fee reduction mechanism in place and the the fees can only be reduced when ya paid this way but apart from that we open the door for all market creators to create to use NERC 20 tokens to create markets okay thank you