TL;DR
With the ever evolving landscape of Web 3.0, the personalized user experience (UX) has not really been utilized when compared to Web 2.0. Within Web 2.0, user data plays a HUGE part customizing the UX which increases overall acquisition, engagement, and retention. Even though Web 3.0 has even more data, thanks to the nature of blockchains, UX personalization has not really been a possibility due to the lack of cross-chain and cross-app profiling as well as trustless data access and computation within smart contracts.
We are collaborating with Trusta to provide a solution to this and are excited to introduce to you an innovative ZKML-powered UX personalization framework for dApps, the ZK MEDIA Score! Brevis takes Trusta’s MEDIA Score, an AI metric that analyzes user activity across chains, and enhances it with ZK Coprocessing technology—allowing smart contracts to process historical on-chain data completely trustlessly. This opens up the doors for developers to create custom data-driven personalized UXs while maintaining Web 3.0’s privacy and decentralization principles.
As an initial first step, we’ve created and launched an alpha release of the ZK MEDIA Score together and have many new features planned for the future.
The Case of the Missing UX Personalization in Web 3.0
Let’s define exactly what we mean when we are talking about “UX personalization.”
UX personalization refers to the practice of tailoring features and the presentation of applications based on user data, such as user preferences, historical in-app behaviors, lifetime value, and much much more. This is arguably the most important UX design principle in the Web 2.0 world because users experience it every day:
- Twitter (X)/Facebook shows each user a different set of recommended content based on their social networks, likes, and search behavior.
- People receive special promotions in their online banking app because they have reached their investment goals.
- People will get a push notification to receive 50 special “magic beans” if they open up a game that they haven’t played in a while.
There are good reasons that this UX personalization is so prevalent in Web 2.0 applications. It helps with making the onboarding journey easier, it strengthens overall user engagement, improves retention, and by helping with all of these, boosts revenue!
However, in the Web 3.0 world, there is almost NO UX personalization happening. So even some of the low-hanging fruit mentioned above go unharvested.
Just pick any major DeFi application as an example: it doesn’t matter if you are a whale casually doing multi-millions in volume on a daily basis or a completely new user buying your first $10 in crypto, you are presented with the exact same user interface (UI) and UX journey.
This is really VERY UNFORTUNATE, because Web 3.0 is able to provide a much more powerful and compelling personalized UX compared to Web 2.0.
Within Web 2.0, the UX personalization is limited to each application individually. This is due to the fact that the developers only have access to the data from that single application. They can only provide a personalized UX after a user has spent a significant amount of time in the application to build a user profile.
However, with Web 3.0 things are a bit different. Since a user’s every interaction with every dApp is faithfully recorded in the blockchain’s transaction and state history, creating a cross-application user profile using real historical data is a real possibility.
So, in theory, before a user even starts to use a new dApp, the dApp should be able to know whether this user is an OG NFT holder, a hard-core gamer, a newbie, a bot hunting for airdrops, or a sophisticated trader by analyzing the user’s historical interactions and behavior across any other applications they have used.
To name a few examples of how this could work—a new DEX could offer lower trading fees for whale traders that use other DEXes to help entice them over to their platform. For new NFT sales, they could open the gates early for users who already hold any NFT that is at least 6 months old, providing incentive to purchase and hold onto NFTs. Just think about it, literally every dApp can benefit from features like these.
The Need for Trusta and Brevis
So—why isn’t there a UX Personalization feature based on all of this powerful cross-application user profiling data today?
Two reasons:
First, there is no readily available data-driven model today to handle this type of cross-application and cross-chain (domain) user profiling. This kind of profiling is simply impossible to do in the old Web 2.0 era.
So, can each application just build their own cross-app user profiling model?
Well, no. Not really.
In fact, it is not really feasible for an application’s development team to understand the numerous factors that would be required to build a user profile that suits their needs. It would require them to understand how other applications work and which interactions within those applications would prove valuable in building a user profile.
What we do need is a shared infrastructure to continuously track and analyze user behavior across numerous chains and applications. This infrastructure should also use advanced Machine Learning (ML) and AI algorithms to extract a simple-to-use user profiling abstraction that can be easily ingested for use in dApps.
This is where Trusta comes into play. Trusta Labs created the MEDIA Score—an AI-powered metric designed to accurately assess a user’s profile across chains and across applications. There are over 1.5 million users who have already queried their MEDIA Scores to gain a deeper understanding of their on-chain footprint on zkSync / OP / Starknet / Linea / Base / Scroll / Manta / Mantle via the TrustGo Dashboard. The core philosophy behind the MEDIA Score is to provide an objective, fair, and quantifiable evaluation of an account’s on-chain engagement level and value based on these factors: Monetary, Engagement, Diversity, Identity and Age. Different dApps are able to rely on this shared AI model to build UX Personalization features to boost acquisition, retention, and revenue for the protocol.
The second reason is that the autonomous smart contracts deployed on the blockchain cannot simply rely on data access and computation results generated based on centralized off-chain databases and indexing solutions. On the security side of things, relying on a centralized point of trust for arguably the most important user life cycle features really defeats the purpose of the trustless nature of blockchain applications. On the regulatory side, centralized intervention of daily user operations of many DeFi applications may make them a centralized operator of financial applications.
We need a trustless solution that can prove to the smart contracts that both the AI/ML computation results and the original historical blockchain data that these results rely on are correct. However, smart contracts have no built-in functions to access historical blockchain data.
Brevis, as a ZK coprocessor, solves this problem by enabling the smart contracts to read from and utilize the full historical on-chain data from any chain and run customizable computations in a completely trust-free way.
By integrating Brevis into Trusta, users can generate a ZK proof of their user profiling MEDIA Score and can prove to any dApp that this score is generated based on true historical on-chain interaction records and a correct computing result using Trusta’s AI model.
Without trusting either Brevis or Trusta, dApps can now easily identify users who are at different engagement and value levels using the ZK MEDIA Scores to directly personalize their user experiences at the smart contract level in a fully decentralized manner!
Get Your ZK MEDIA Score and Share in an $11,000 Reward Pool
Brevis and Trusta have worked together to launch the first alpha release of the ZK MEDIA Score! This is the world’s first Web 3.0 trustless UX Personalization infrastructure powered by ZKML.
You can try it out by generating your ZK MEDIA Score and sharing in an $11,000 reward pool based on your ZK MEDIA score. Verified all in a completely trustless way!
The details of this campaign will be shared soon.
This ZK MEDIA Score is a subset of the in-product MEDIA Score and covers the following aspects:
- A sub proof for each of the sub scores is generated validating that the data for the calculations exist on-chain for the user’s Uniswap transactions, Aave staking history, on-chain engagement, contract interactions count, airdrop eligibility history, and account age.
- A proof of the aggregated total MEDIA Score combining all of the sub scores.
- A proof that the MEDIA Score was calculated using Trusta’s AI model.
What’s Next
Brevis and Trusta are committed to form a long-term partnership to see ZK MEDIA Score all the way to production, including some exciting new features like:
- Expanded AI models with privacy-preserving features
- Cross-user proof aggregation to reduce costs of on-chain verification
- Integration of the ZK MEDIA Score into modularized DeFi dApps
We also expect to expand our partnership with amazing teams such as Nebra and Celer Network.
Stay tuned for more news soon!
About Brevis
Brevis is a smart ZK coprocessor that empowers smart contracts to read the full historical on-chain data from any chain and run customizable computations in a completely trust-free way.
Brevis can trustlessly read the states, transactions, and receipts of any time frame. It efficiently runs computations on data of any size: as the data to prove increases in size, Brevis significantly reduces the per unit cost and time for proof generation. Additionally, Brevis seamlessly verifies Zero-Knowledge (ZK) proofs on any blockchain, providing a unified and trustless user experience for decentralized applications (dApps) operating across multiple chains.
Brevis empowers data-rich use cases including data-driven DeFi, user-segment optimized live-ops features, zkBridges, zkDID, and smart-contract native risk control, all executed in a trustless manner.
Dive Deeper into Brevis:
🌐Website | 🗨️Telegram | 🐦X | 🎮Demo | 📘SDK Docs
About Trusta Labs
Trusta Labs aims to continuously develop MEDIA Score as the infrastructure for on-chain user value assessment. In the KYA (Know Your Account) product called TrustGo from Trusta Labs, every user can look up their own unique MEDIA Score. It helps to fairly identify accounts that contribute to the L2 ecosystem. Through MEDIA Score, users can gain a deeper understanding of their on-chain activities and value, while project teams can accurately allocate resources and incentives to users who truly contribute.