An Ethereum Improvement Proposal (EIP) is a formal design document that describes a proposed change to Ethereum: a new feature, a protocol modification, an application-layer standard, or a process improvement. EIPs are Ethereum’s main mechanism for open technical governance. Anyone can draft one, the community debates it on GitHub and the Ethereum Magicians forum, core developers assess its feasibility, and accepted core EIPs ship in network upgrades (hard forks).
Each EIP follows a template covering motivation, specification, rationale, and, where relevant, a reference implementation. Proposals move through stages such as Draft, Review, Last Call, and Final. There are three main types:
- Standards Track: Changes to the core protocol, networking, or interfaces, plus ERCs
- Meta: Process changes, including the Meta EIPs that list the contents of each network upgrade
- Informational: Guidelines and general information that do not propose a change
ERCs (Ethereum Requests for Comments) are the application-layer subset of Standards Track EIPs, defining conventions such as ERC-20 for fungible tokens, ERC-721 for NFTs, and ERC-1155 for multi-tokens. Since 2023, ERCs are maintained in their own repository, separate from the core EIPs repository. ERCs are conventions that applications choose to follow, while core EIPs change the protocol and need agreement from the independent client teams that implement Ethereum.
Where Did EIPs Come From?
- 2015: Ethereum launches, and EIP-1 defines the proposal process, modeled on Bitcoin’s BIP system. EIP-20 (ERC-20) is proposed in November
- 2016: Homestead goes live in March, and the response to the DAO attack drives further protocol changes, including EIP-150 (gas repricing) and EIP-155 (replay protection)
- 2017 to 2018: ERC-721 is proposed and finalized in 2018, enabling NFTs such as CryptoKitties. ERC-1155 is proposed by Enjin in 2018 for efficient multi-token contracts
- August 2021: EIP-1559 activates in the London hard fork, introducing the base fee and the ETH burn
- September 2022: EIP-3675 delivers the Merge, moving Ethereum from proof of work to proof of stake and cutting energy use by roughly 99.95%
- March 2023: ERC-4337 account abstraction goes live on mainnet as a smart contract system, with no protocol change required
- March 2024: The Dencun upgrade activates EIP-4844 (proto-danksharding), adding blob transactions that cut Layer 2 data costs by roughly 90 to 99%
- May 2025: The Pectra upgrade ships EIP-7702, which lets ordinary accounts temporarily act like smart contract wallets, along with EIP-7251, which raises the maximum validator balance
- December 2025: The Fusaka upgrade activates EIP-7594 (PeerDAS), bringing data availability sampling to Ethereum blobs
- 2026: Glamsterdam, featuring EIP-7732 (enshrined proposer-builder separation) and EIP-7928 (block-level access lists), reaches public testing in August 2026. A mainnet date had not been finalized at the time of writing
How Does the EIP Process Work?
| EIP | Name | Impact |
|---|---|---|
| EIP-20 (ERC-20) | Token standard | Enabled the ICO era and nearly all fungible tokens |
| EIP-721 (ERC-721) | NFT standard | Underpins NFT marketplaces and collectibles |
| EIP-1559 | Fee market reform | Base fee plus priority tip, with the base fee burned |
| EIP-3675 | The Merge | Proof of work to proof of stake |
| EIP-4844 | Proto-danksharding | Blob transactions for cheap Layer 2 data |
| ERC-4337 | Account abstraction | Smart contract wallets without a protocol change |
| EIP-7702 | Account upgrade for EOAs | Lets ordinary accounts use smart wallet features such as batching and gas sponsorship |
| EIP-7594 | PeerDAS | Data availability sampling for blobs |
EIPs in Simple Terms
- An open upgrade process: A developer, researcher, or community member can propose an EIP. It is debated, refined, and adopted only if rough consensus forms
- Shaping the industry: ERC-20 created the token economy, ERC-721 created NFTs, and EIP-1559 changed how fees work for every user
- No single boss: Unlike a company roadmap, no one entity decides. Client teams, researchers, and the wider community all have a say, although core developers and Ethereum Foundation researchers carry substantial informal influence
- The burn (EIP-1559): The burn has destroyed roughly 4.6 million ETH since August 2021. Since Dencun made Layer 2 usage much cheaper on Layer 1, burn rates have fallen, and ETH supply has grown modestly rather than shrinking
- Smarter accounts (ERC-4337 and EIP-7702): These let wallets offer social recovery, gas sponsorship, and batched transactions. Accounts still rely on keys or passkeys, so the change is flexibility, not key-free wallets
What Are Real-World Examples of EIPs?
| Scenario | Implementation | Outcome |
|---|---|---|
| ERC-20 | USDC, LINK, UNI, and most tokens implement the standard | Wallets, exchanges, and DeFi protocols can handle any compliant token |
| EIP-1559 | Base fee is burned, tip goes to validators | Fees became more predictable, and about 4.6 million ETH has been burned |
| The Merge (EIP-3675) | Consensus switches to proof of stake | Energy use fell by about 99.95% and new issuance dropped by roughly 90% |
| EIP-4844 | Rollups post data as blobs | Layer 2 fees dropped to a fraction of a cent in many cases |
| ERC-4337 and EIP-7702 | Wallets such as Safe and Coinbase Smart Wallet use account abstraction | Features such as passkey login, social recovery, and gas sponsorship are available without changing the core protocol |
What Are the Advantages of the EIP Process?
| Advantage | Description |
|---|---|
| Open governance | Anyone can propose, and the process is visible on GitHub |
| Merit-based review | Technical quality and community consensus decide outcomes |
| Standardization | ERCs create interoperable application standards |
| Documented history | Every proposal, rationale, and debate is preserved |
| Continuous improvement | Ethereum can evolve through successive network upgrades |
What Are the Risks of the EIP Process?
| Risk | Description |
|---|---|
| Slow process | Contentious proposals can take years to finalize |
| Informal influence | Core developers and Ethereum Foundation researchers hold significant sway over direction, even within an open process |
| Chain split risk | Deep community disagreement over a change can, in extreme cases, lead to a contentious fork |
| Complexity | Meaningful participation requires deep technical expertise |
| Coordination | Several independent client teams must all implement a core change in step |
| Organizational change | The Ethereum Foundation, a major contributor to research and funding, lost its co-executive directors in 2026 and cut about 20% of its staff (54 roles) and roughly 40% of its budget in June 2026 |
| Quantum threat | Ethereum’s ECDSA signatures are vulnerable to a future large-scale quantum computer, so post-quantum migration is an active research priority |
Risk management tips:
- Follow core EIPs to anticipate network upgrades that affect your assets
- Watch the official EIPs and ERCs repositories on GitHub for upcoming changes
- Before a hard fork, confirm that the wallets and exchanges you use are upgraded
- Remember that ERCs are conventions, not rules enforced by the protocol
- Follow the AllCoreDevs calls for early signals on upcoming changes
Frequently Asked Questions About EIPs
Q: What is the difference between an EIP and an ERC? An ERC is a subset of EIPs that defines application-level standards. Every ERC is an EIP, but not every EIP is an ERC. Core EIPs change the protocol, while ERCs set conventions for smart contracts, such as ERC-20 and ERC-721.
Q: Who decides which EIPs get implemented? It combines several groups: authors who write the specification, client developers who implement it, EIP editors who curate the process, and the wider community that debates and signals support. No single entity holds a veto, but Ethereum Foundation researchers and core client developers have significant informal influence.
Q: Can an EIP be rejected? Yes. Many EIPs are never implemented, because of technical flaws, poor benefit-to-complexity ratios, lack of consensus, or better alternatives. Some contentious proposals are debated for years.
Q: What was EIP-1559, and why was it controversial? EIP-1559 replaced the first-price auction for fees with a burned base fee plus a priority tip. Miners opposed it because it reduced fee revenue, while users valued predictable fees and the burn. It activated in August 2021.
Q: How do EIPs affect wallet users? Most users need to do nothing, because wallets update automatically. Larger upgrades can require updating wallet or node software before the fork. EIPs also shape the user experience: EIP-4844 lowered Layer 2 fees, and ERC-4337 and EIP-7702 enable smarter wallet features.
Q: What upgrade is next for Ethereum? Glamsterdam, which includes EIP-7732 and EIP-7928, entered public testing in August 2026. The exact mainnet date had not been confirmed at the time of writing, so check the Ethereum Foundation blog or the AllCoreDevs notes for the latest timing.
Related Terms
- ERC (Ethereum Request for Comments): The application-layer category of EIPs, covering token and contract standards
- Hard Fork: A protocol change that is not backward compatible and requires nodes to upgrade
- Network Upgrade: A bundle of EIPs activated together, such as Dencun, Pectra, or Fusaka
- Account Abstraction: A set of changes, including ERC-4337 and EIP-7702, that make wallets programmable
- Proto-Danksharding (EIP-4844): The upgrade that introduced blob transactions for rollups
- AllCoreDevs: The regular call where Ethereum client developers coordinate upgrades
