Biconomy: The Web3 UX Pioneer
Introduction: The Need for a UX Layer
Account Abstraction brought previously Externally Owned Accounts (EOAs) to an established environment of programmability and composable economic dynamics.
Efficiency has derived from this insertion so that DeFi and Web3 UX witnessed a new standard.
We’re finally seeing a framework that is future-proof and embraces innovations: Account Abstraction.
Account Abstraction was born with ERC4337, which completely avoids consensus-layer protocol changes, instead, it relies on higher-layer infrastructure.
This implementation, other than improving general efficiency, has brought opportunities to leverage the composability of blockchain ecosystems, customisation, and abstraction of technical and frictiony tasks.
In fact, almost every interaction on a dApp is a transaction, so several avoidable processes come up at a frequent pace.
Users should not stake their time for multiple transactions that directly communicate with a virtual machine, and therefore propose signs and approvals backed by technicalities, to complete a single task.
In addition, the whole gas paying processes should be optimized, hence the UX system is required to take charge of that.
Biconomy is positioned right at the beginning of the track, helping the runner (user) avoid the obstacles (bad UX), and delivering him the gun so he can be the starter and there is no need to wait for a “Get Set” before action.
Biconomy is laying the foundation for a UX platform, which features modular Smart Accounts, Paymasters and AI-optimized user operations.
Now, by taking into consideration that 30% of Smart Accounts derive from Biconomy (one deployed every 17s!!), if we look at data we can see how it’s playing a key role in shifting the industry’s standard from EOAs to AA.
UserOps are increasing, so more meta-transactions are in play, with bundlers generating more fees.
This increase could also be due to the current airdrop season that’s going on in the crypto space, and thanks to Smart Accounts users can optimize their farming process by abstracting tons of transactions and executing them in batches, experiencing gasless operations and paying gas in any token.
This is only one of the catalysts and use-cases that Smart Accounts powered by AA can offer.
These last ones also impact other applications: gaming, for example, can be a very good field where implementing Smart Accounts so that the user just needs an email to be ready to play and set up an account.
Also, customization in these cases is crucial. Users may want full abstraction when playing a game, while they need a different kind of UI/UX for a DeFi protocol.
Let’s take Perp DEXes as an example.
Perp DEXes’ volumes are still significantly lower compared to CEXes ones.
We can see through this dashboard how percentages are declining, and it’s partly related to the low-quality UX that DEXes provided until recent times.
Now, thanks to AA providers like Biconomy, some dots are going to be connected and a UX fundamental pillar is being developed.
Biconomy is working with several DEXes in order to improve the poor experience problem and onboard many users into DeFi.
Rage Trade is one of these, and it integrated Biconomy’s sessions module so that when a user creates a session, he won’t be asked to sign any action for 7 days.
The session is created on multiple chains thanks to the multi-chain validation module, therefore the user can navigate and operate seamlessly between chains.
However, this is just one of the benefits AA can bring to the table.
As we have seen before paymasters also help, but above all the specialization of accounts is what makes AA provide a tailored solution for every application.
Biconomy: Building a Decentralized Account Management Network, aka The UX Network
Fast and efficient specialization doesn’t come without a proper framework that supports it.
The importance of a thriving ecosystem is obvious, but it’s both the catalyst and something that has to be led and nurtured by an underlying architecture.
And so here’s where Biconomy has developed its products with consistent modularity, which enables composability across ecosystems, and allows for the customisation of modules according to needs.
“By permissionlessly augmenting their smart accounts with modules, users would have access to wallets that more closely align to their web3 needs (much like LEGOs)”
When developing applications for a similar target market, developers often face identical needs. However, they are forced to recreate the same functionalities from scratch, which can lead to duplicated effort, potential security vulnerabilities and inefficiencies.
Furthermore building and managing the infrastructure required for batched transactions, paymasters, and session keys is a complex and demanding task.
Handling session keys, for instance, necessitates a robust and secure storage system.
Modules require rigorous auditing.
To ensure that future accounts can offer features that meet high-quality standards without the overhead of maintaining such infrastructure, it becomes more efficient to integrate bundlers, paymasters, and session key solutions provided by specialized third-party infra providers rather than developing these capabilities internally.
Wallets may choose to integrate pre-audited and verified modules instead of developing everything from scratch.
Modularity and standards like ERC-7579 enable this level of composability, and push towards leveraging web3 values to attain that thriving ecosystem we talked about above, which will be responsible for offering the best UX.
Picture a scenario where numerous pre-built and pre-programmed functionalities are readily available for integration.
Developers could effortlessly select the features they want to incorporate into their smart accounts, enabling them in a safe way and with ease.
This is precisely where Biconomy Network wants to act.
Next Step: The Rising of the Intelligence Layer
AA paved the way, and still remains the power pack behind the technological infrastructure and the network architecture, but new synergies have been created and AA is going to expand and support even more powerful and effective solutions.
Intents are currently one of the most discussed ones.
These represent a layer of abstraction on top of account abstraction, and with ERC-4337 that enables users to let the third party execute the order flows, so Intents can be part of and add value to Smart Accounts.
AI Agents however are what, together with Intents, and powered by AA, will bring the enhanced user experience DeFi and Web3 are looking for.
In this sense, the platform becomes a resource that the user can direct and offload his efforts to, rather than something which loads him with infinite computer-coded interactions.
This way, the whole operational task workflow gets abstracted from final users, who just have to provide instructions on the goal they’re trying to reach with their transactions.
That’s why ERC4337 plays a key role: it enables devs to create a standard on how the users’ instructions get abstracted, and AI agents take this to the next level, unlocking efficient activities and composability across Web3.
These “order flows” instructions are then available to be executed by AI-powered Autonomous Agents, that help users achieve their initial goal without having to worry about all the hurdles that the fragmented web3 UX faces them with.
This kind of workflow would look like this:
Imagine for instance an interoperable dApp. It could support a network of AI Agents scouting for the cheapest solutions across chains and unifying the process into a single report to the user.
As the combined product of crypto and AI, Autonomous Agents behave like on-chain performers of users’ will.
AI Agents have the potential to disrupt the current Web3 landscape and reshape on-chain interactions as we know them, working as a catalyst to onboard new users to Web3 without having to subject them to the risky and cumbersome process of learning how on-chain transactions work.
If we combine AI Agents with AA-powered features such as Gasless and batched transactions, simplified log-in and the possibility to recover your account in case something goes wrong, we can dramatically boost web3 user retention and engagement rate.
Think about all the people who never entered, or worse, left Web3 due to the high technical know-how required to interact with dApps and chains: they could still be part of our ecosystem, if only we had provided them with the proper tools.
Luckily, the music is changing, and soon the times in which every on-chain step had to be performed manually will be just a bad memory.
Use cases for AI Agents are pretty much endless, but there are some trends worthy of being mentioned.
The most widely adopted use-case for AI-agent is Portfolio Management, and it’s already deeply rooted in TradFi and DeFi.
Thanks to their ability in identifying patterns and reliably executing fast and accurate data-driven decisions while complying with the given risk tolerance profile, AI Agents are already valuable resources for fund managers all over the world. Additionally, ZK-based features can be enabled to satisfy enhanced security needs.
A lesser-known but growing use case for AI Agents concerns the gaming industry.
Gaming agents are gradually becoming a reality in Web3 gaming: they act as pets or assistants that follow the player’s journey with the final goal of providing a richer and more immersive gaming experience.
These agents interact with players and items through on-chain smart contracts and wallets, and can be leveraged to perform trade on behalf of players.
Let’s say you need some assets to craft a certain item but you don’t have the time to manually go on-chain and perform the purchase, so you decide to just approve the transaction and let your game assistant handle that for you: UX-wise, that’s a very big improvement.
Finally, a very interesting use-case is given by Aperture Finance, who has developed a so-called “Intents-GPT” which basically provides an intuitive chatbox interface inspired by GPT, allowing users to express their goals in natural language.
Third-party participants, known as Solvers, handle requests by optimizing processes to ensure efficient and cost-effective execution.
Biconomy DAN (Delegated Authorisation Network)
We have seen how AA is able to be the fundamental pillar of the ultimate but also an opening standard for a UX solution.
However, AI Agents, given their great power of change, also hold a challenge.
The key challenge is to build an intent-centric network of solvers that is sufficiently decentralized.
Furthermore, agents lack true autonomy, as they can only suggest on-chain actions and prepare transactions for users to sign or approve. This limitation has led to the development of solutions like Biconomy’s Sessions, which we touched on earlier.
Not only, in the pursuit of automation, protocols often sacrifice security.
They resort to alternative measures, such as approvals, centralized vaults, and shared private key pairs, which ultimately put protocols in control of users’ assets, introducing significant risks.
To mitigate these risks, establishing a robust safeguard system for AI agents is needed, empowering users to define explicit permissions that dictate the actions AI agents can and cannot sign.
This requires a solution that enables the delegation of transaction authorization to AI agents, but only within the boundaries of predefined rules and permissions.
DAN is precisely the solution to these problems. In fact, to overcome these limitations, AI agents need to be granted secure on-chain task delegation capabilities.
One approach would be to give AI agents complete control over users’ keys, enabling them to sign multiple transactions on their behalf. However, this approach poses significant security risks.
So Biconomy Delegated Authorisation Network has been built to offer a secure solution for on-chain task delegation, enabling AI agents to operate autonomously while ensuring user-defined permissions and robust security.
To achieve this, Biconomy has partnered with EigenLayer and Silence Labs which helped build the architecture behind DAN.
Let’s see therefore the architecture’s core functions:
DAN Core Functions
User-Defined Programmable Authorization
Users can define the permissions granted to AI agents, ensuring that transactions occur within set rules. This includes:
- Restricting access to specific assets
- Setting maximum transaction amounts
- Limiting transaction times
Robust Security via EigenLayer AVS
Instead of giving AI agents full control over private keys, DAN uses EigenLayer AVS for decentralized key storage and permission management. This ensures that:
- User keys are securely stored
- Permissions are strictly enforced
- Cryptographic security prevents key leakage and permission abuse
Autonomous Operation
DAN empowers AI agents to execute transactions on behalf of users without requiring manual signing or approval. By providing agents with access to user-specific ‘Delegated Auth’ keys stored on AVS, they can securely sign transactions within their permissions, improving operational efficiency and user experience.
Now, Biconomy plans to introduce additional features to DAN, including human-readable permissions, permissions based on external data sources, and intent-based permissions. These developments will further enhance the flexibility and practicality of DAN and will further position Biconomy as a Web3 UX pioneer.
Conclusion
With that said, the intelligence layer has surely a disruptive potential, and Biconomy can really contribute to changing blockchain high-level infra dynamics, pushing so the Web3 adoption, and redefining dApps interactions.
Notable is then the decision of collaborating with EigenLayer for safeguarding and ensuring a proper system.
Leveraging Ethereum’s economic security to store user keys and permissions is in fact an efficient solution.
Finally, although DAN has already been applied in multiple projects, including Polywrap and zkAutomate, we remain curious to see it integrated by more projects. To see it being battle-tested, and collect feedback to enhance ever more the standard, and provide tailored experiences thanks to native modularity.
