The following is a guest post and analysis from Nanfeng Jie, Lead Product Manager at Trust Wallet.
Now, let’s take a step back and examine the broader market context and the real-world pain points that led us to rethink how gas fees should work.
But scale alone doesn’t tell the whole story, because the user experience hasn’t kept pace. People still juggle across chains, manually manage gas balances, and abandon transactions when approvals don’t make sense.
One of the most common pieces of feedback wallet teams hear from users is simple: “I don’t want to think about gas — I just want the transaction to go through.” And it’s a fair point. Gas fees aren’t just a cost; they’re a cognitive burden. Every failed or delayed transaction chips away at trust in the system.
The issue’s root lies in the architecture we’ve relied on for over a decade: EOA, the default wallet type for most users. It’s lightweight and secure, but wasn’t designed for the programmable, dynamic interactions that define today’s decentralized applications (DApps).
That’s why the Ethereum EIP-7702 proposal represents such a meaningful shift.
With EIP-7702, users maintain full custody of their assets while gaining access to more flexible transaction logic. That means bundling approvals and actions into a single tap, enabling recurring payments, or supporting delegated session keys without separate smart contracts.
Simply put, EIP-7702 means fewer steps, less confusion, and a smoother user experience. Transactions are faster and more predictable, allowing gas fees to be paid using tokens already held without the need for native assets in advance.
Technically speaking, EIP-7702 acts as a modular extension to the EOA model. The user signs an intent, which may contain custom logic, and the wallet executes that intent through a temporary contract. Once the transaction is complete, the account returns to its standard EOA state, unlocking a more intelligent transaction layer for developers and infrastructure providers.
In turn, Web3 starts to behave more like something built for real people, not protocols.
Supporting EIP-7702 at scale requires more than integrating a new transaction type or making user interface (UI) updates. It demands a robust, modular backend infrastructure capable of interpreting user intent, dynamically routing gas, and reliably executing complex actions across chains.
At Trust Wallet, we chose not to rely on third-party abstractions or SDKs. Instead, we developed our own account abstraction engine entirely in-house, built to be secure, scalable, and chain-agnostic. This modular system includes:
This internal architecture gives Trust Wallet a lasting edge in performance and reliability, while setting a new standard for EOA-based smart wallet design. It also paved the way for FlexGas, the first prominent feature we built on this foundation. FlexGas allows users to pay gas fees with tokens such as USDT or TWT on Ethereum and BNB Chain.
Crucially, all of these enhancements preserve the essence of what makes self-custody appealing: users retain full control of their private keys, their seed phrases remain unchanged, and there’s no need to upgrade to a contract-based wallet. This balance (between power and autonomy) is what makes EIP-7702 such a significant upgrade rather than a disruptive replacement.
If widely adopted, EIP-7702 could become a defining layer in the next generation of Web3 infrastructure. It enables a future where wallets are responsive, intelligent agents — automating complex strategies, onboarding users, and unlocking frictionless interactions at scale.
The first real-world applications are already in motion. Features allowing users to pay for gas with tokens they already hold are nearly here. Gasless onboarding, automated execution strategies, wallet-as-a-service models, and smart transaction policies for institutional-grade use are all within reach.
The biggest breakthroughs in crypto often come not from radical overhauls, but from quiet upgrades that remove invisible frictions. EIP-7702 may be exactly that. It doesn’t change the way we think about Ethereum’s security model — it changes what that model can do for users.
Because at the end of the day, progress in Web3 doesn’t depend on how smart our contracts are. It depends on how natural they feel to use.
