Scam Prevention Tactics Using Organic Token Score | CASTANBUL 2025

this next speaker is a data nerd a data plumber a data fanatic he's going to talk about all the ways Jupiter will be protecting users using real metrics that provide signal in the noise welcome to the stage Yung everyone okay M my test can you hear me okay um wait hold on I seems like I need to test the clicker it's like there's always a problem that uh hold on okay so um um many of you will not know who I am uh I'm going to give a brief introduction about myself so my name is

Yong I used to be a software engineer in Facebook and now I'm doing data engineering on the biggest token distribution platform on the biggest uh blockchain yeah okay so um over the years there's been an increasing number of tokens uh being creat every day so just for example just yesterday there were more than 80,000 tokens created so I I know we I we say this number A lot of times but 880,000 that's a pretty huge number right yeah uh and um the reason because there's so many tokens being created is because scammers love to create

duplicates scamers love to mislead their users um so the big question is are you you able toid identify the scam tokens and are you able to identify the the right token within the duplicates so let's play a game the name of the game is called scam or save so the the rules are simple so I'm going to give you a set of Matrix and you tell me whether the token is likely to be a scam or is it likely to be save so uh the metrics that you can use will be holders volume and tvl

so token has around 9,000 holders has around 3.6 million volume and it has around half a million of tvl so is the token likely to be a scam or is it safe okay scam or safe okay I'm getting uh pretty pretty uh the response is around the same I guess so the correct answer is actually I don't know I actually don't know the reason is because this data is noisy gameable and inorganic so let let me tell you why so scammers can gain the hold account by creating many wallets and then they can send these

tokens to the wallets this creates an illusion that there's a huge number of pH count but when in fact it's just one user so this might be how uh typical bubble map will look like in this scenario and you might have seen this uh bubble map before additionally scammers can inflate the volume by repeatedly repeatedly buying and selling the same amount of tokens and the scammers can also game the TVR by manipulating the price as from this chart is I mean it's pretty obvious but I'm giving you an example okay so given this metrics you

will never win the the game scam or save we need to find more useful uh data for our users we need to do better so we we need a way to uncover the scams and also disc discover tokens safely so we are very very excited to share with you a new class of metrics that we have developed and we call them the organic index so the organic index is meant to find the signal in the noise so based on the metrix that you are familiar with uh we have integrated an organic component to them and

this include metrics like um organic holders organic volume and organic tvl so how it works so we use a set of heris sticks that we use to identify addresses that we deem as real wallet user real user wallets and then we'll track this wallets in real time to form the organic score yeah additionally all these metrics so holders organic holders organic tvl and or organic uh volume will be condensed to form a single organic index score so this score will be shown on our brand new token page you'll be you have more useful information differentiate

between the tokens with similar ticket as you can see here with this organic index you'll be able to identify the real token on the left and the fake token on the right this is not Financial advice by the way okay we will also use the organic index to power our very own trending list you will be able to see the organic score together with some of these other organic metrics uh which will be show on the UI it's currently n live on D.G so go go give it a try currently it's in better but we

look forward to your feedback thank you