Page last updated: September 21, 2023
Users interact with Ethereum using externally owned accounts (EOAs). This is the only way to start a transaction or execute a smart contract. This limits how users can interact with Ethereum. For example, it makes it difficult to do batches of transactions and requires users to always keep an ETH balance to cover gas.
Account abstraction is a way to solve these problems by allowing users to flexibly program more security and better user experiences into their accounts. This can happen by upgrading EOAs(opens in a new tab) so they can be controlled by smart contracts, or by upgrading smart contracts(opens in a new tab) so they can initiate transactions. These options both require changes to the Ethereum protocol. There is also a third path involving adding a second, separate transaction system(opens in a new tab) to run in parallel to the existing protocol. Regardless of the route, the outcome is access to Ethereum via smart contract wallets, either natively supported as part of the existing protocol or via an add-on transaction network.
Smart contract wallets unlock many benefits for the user, including:
- define your own flexible security rules
- recover your account if you lose the keys
- share your account security across trusted devices or individuals
- pay someone else's gas, or have someone else pay yours
- batch transactions together (e.g. approve and execute a swap in one go)
- more opportunities for dapps and wallet developers to innovate on user experiences
These benefits are not natively supported today because only externally-owned accounts (EOAs) can start transactions. EOAs are simply public-private key pairs. They work like this:
- if you have the private key you can do anything within the rules of the Ethereum Virtual Machine (EVM)
- if you do not have the private key you can do nothing.
If you lose your keys they can't be recovered, and stolen keys give thieves instant access to all the funds in an account.
Smart contract wallets are the solution to these problems, but today they are difficult to program because in the end any logic they implement has to be translated into a set of EOA transactions before they can be processed by Ethereum. Account abstraction enables smart contracts to initiate transactions themselves, so that any logic that the user wishes to implement can be coded into the smart contract wallet itself and executed on Ethereum.
Ultimately, account abstraction improves support for smart contract wallets, making them easier to build and safer to use. In the end, with account abstraction, users can enjoy all the benefits of Ethereum without knowing or caring about the underlying technology.
Today's accounts are secured using private keys that are calculated from seed phrases. Any person who has access to a seed phrase can easily discover the private key protecting an account and gain access to all the assets it protects. If a private key and seed phrase are lost, they can never be recovered and the assets they control are frozen forever. Securing these seed phrases is awkward, even for expert users and seed phrase phishing is one of the most common ways users get scammed.
Account abstraction will solve this problem by using a smart contract to hold assets and authorize transactions. These smart contracts can then be decorated with custom logic to make them as secure and tailored to the user as possible. Ultimately, you still use private keys to control access to your account, but with safety nets that make them easier and safer to manage.
For example, backup keys can be added to a wallet so that if you lose or accidentally expose your main key, it can be replaced with a new, secure one with permission from the backup keys. You might secure each of these keys in a different way, or split them across trusted guardians. This makes it much harder for a thief to gain full control over your funds. Similarly, you can add rules to the wallet to reduce the impact if your main key gets compromised, for example you might allow low value transactions to be verified by a single signature, whereas higher value transactions require approval from multiple authenticated signers. There are other ways smart contract wallets can help you to thwart thieves too, for example a whitelist can be used to block every transaction unless it is to a trusted address or verified by several of your pre-approved keys.
- Multisig authorization: You can share authorization credentials across multiple trusted people or devices. Then the contract can be configured so that transactions of more than some preset value require authorization from a certain proportion (e.g. 3/5) of the trusted parties. For example, high value transactions might require approval from both a mobile device and a hardware wallet, or signatures from accounts distributed to trusted family members.
- Account freezing: If a device is lost or compromised the account can be locked from another authorized device, protecting the user's assets.
- Account recovery: Lost a device or forgotten a password? In the current paradigm this means your assets could be frozen forever. With a smart contract wallet, you can set some pre-approved accounts that can authorize new devices and reset access.
- Set transaction limits: Specify daily thresholds for how much value can be transferred from the account in a day/week/month. This means if an attacker does gain access to your account they can't drain everything at once and you have opportunities to freeze and reset access.
- Create whitelists: Only allow transactions to certain addresses that you know to be safe. This means that even if your private key was stolen, the attacker could not send funds to non-whitelisted destination accounts. These whitelists would require multiple signatures to change them so that an attacker couldn't add their own address to the list unless they had access to several of your backup keys.
Account abstraction allows for a better overall user experience as well as improved security because it adds support for smart contract wallets at the protocol level. The most important reason for this is that it will provide developers of smart contracts, wallets and applications with much more freedom to innovate on the user experience in ways that we may not yet be able to anticipate. Some obvious improvements that will come along with account abstraction include bundling of transactions for speed and efficiency. For example, a simple swap should be a one-click operation, but today it requires signing multiple transactions for approving spending of individual tokens before the swap is executed. Account abstraction removes that friction by allowing transaction bundling. Furthermore, the bundled transaction could approve precisely the right value of tokens required for each transaction and then revoke the approvals after the transaction completes, providing additional security.
Gas management is also much improved with account abstraction. Not only can applications offer to pay their users' gas fees, but gas fees can be paid in tokens other than ETH, freeing users from having to maintain an ETH balance for funding transactions. This would work by swapping the user's tokens for ETH inside the contract and then using the ETH to pay for gas.
Trusted sessions are also potentially transformative for user experiences, especially for applications like gaming, where large numbers of small transactions might need approval in a short time. Individually approving each transaction would break the gaming experience, but permanent approval is unsafe. A smart contract wallet could approve certain transactions for a fixed time, up to a specific value or only to certain addresses.
It is also interesting to consider how purchases could change with account abstraction. Today, each transaction has to be approved and executed from a wallet pre-funded with a sufficient amount of the correct token. With account abstraction, the experience could be more like familiar online shopping where a user could fill a "basket" with items and click once to purchase all at once, with all the logic required handled by the contract, not the user.
These are just a few examples of how user experiences could be levelled up by account abstraction, but there will be many more that we haven't imagined yet. Account abstraction frees developers from the constraints of present-day EOAs, allows them to bring the good aspects of web2 into web3 without sacrificing self-custody and to hack creatively on inventive new user experiences.
Smart contract wallets exist today but are challenging to implement because the EVM does not support them. Instead, they rely on wrapping relatively complex code around standard Ethereum transactions. Ethereum can change this by allowing smart contracts to initiate transactions, handling the necessary logic in Ethereum smart contracts instead of off-chain. Putting logic into smart contracts also increases Ethereum's decentralization since it removes the need for "relayers" run by wallet developers to translate messages signed by the user to regular Ethereum transactions.
Smart contract wallets are already available, but more upgrades are required to make them as decentralized and permissionless as possible. EIP-4337 is a mature proposal that does not require any changes to Ethereum's protocol, so it is possible that this could be implemented quickly. However, upgrades that alter Ethereum's protocol are currently not in active development, so those changes may take much longer to ship. It is also possible that account abstraction is achieved well enough by EIP-4337 that no protocol changes are ever required.
- erc4337.io(opens in a new tab)
- Account abstraction panel discussion from Devcon Bogota(opens in a new tab)
- "Why account abstraction is a game changer for dapps" from Devcon Bogota(opens in a new tab)
- "Account abstraction ELI5" from Devcon Bogota(opens in a new tab)
- Vitalik's "Road to Account Abstraction" notes(opens in a new tab)
- Vitalik's blog post on social recovery wallets(opens in a new tab)
- EIP-2938 notes(opens in a new tab)
- EIP-2938 documentation(opens in a new tab)
- EIP-4337 notes(opens in a new tab)
- EIP-4337 documentation(opens in a new tab)
- EIP-2771 documentation(opens in a new tab)
- "Basics of Account Abstraction" -- What is Account Abstraction Part I(opens in a new tab)