What Is Proof of Stake?

Proof of stake lets validators propose and verify blocks by staking coins. Learn how Ethereum validators, slots, attestations, and slashing penalties work.

By Jonathan West · Updated October 2026

Proof of stake is a blockchain consensus mechanism where validators lock up native coins as a security deposit to earn the right to propose and verify new blocks.

It determines how transactions gain finality, replacing proof of work with direct financial collateral that can be slashed if participants cheat.

About proof of stake

Proof of stake is a way for a blockchain to agree on new blocks where validators lock up their own coins as a deposit, are chosen to propose and vote on blocks, and lose part of the deposit if they cheat. Instead of miners expending CPU power to find a proof of work, proof of stake relies on bonded participants called validators. Stakers put capital on the line to confirm honest records. While Bitcoin continues to use computational work, Ethereum moved to proof of stake during the Merge in September 2022 to change how consensus is secured.

How it actually works

In a proof of stake network like Ethereum, consensus starts when participants deposit funds into an official deposit contract. To activate a single validator instance, a node operator must deposit 32 ETH. Once active, the protocol schedules time into regular units called slots and epochs. Each slot lasts 12 seconds, and a collection of 32 slots forms one epoch. For every 12-second slot, the network uses a pseudo-random mechanism named RANDAO to select exactly one block proposer from the global validator set. Simultaneously, the system assigns a committee of random validators to review the proposed transactions, evaluate their state transitions, and submit official votes called attestations. Every active validator across the network attests to a block once per epoch.

After a block proposer packages transactions and broadcasts the payload, the committee members verify the contents and cast their attestations. If the proposed data follows the consensus rules, these votes accumulate across the slots. Finality requires a two-thirds supermajority of all staked ETH across the network. Reverting a finalized block would require an attacker to be willing to lose at least one-third of all staked ETH. Under ordinary operating conditions, validators receive variable rewards for proposing valid blocks and submitting timely attestations. These rewards are not fixed or guaranteed, and they vary over time depending on network participation and state.

A validator that breaks the rules can be slashed. Slashing enforces an immediate penalty of 1/4096 of the validator's effective balance, capped at up to 0.5 ETH, followed by a mandatory exit sequence. On day 18 after the infraction, a correlation penalty follows. Finally, the node is completely ejected from the validator set on day 36. While a single validator slashed alone loses less than 0.1% of its stake, a coordinated mass slashing can destroy up to 100% of the deposit. Also, if the chain fails to finalize for more than four epochs, the inactivity leak activates.

The security assumptions of proof of stake differ fundamentally from proof of work. In the original 2008 Bitcoin whitepaper, nodes expend CPU power to create proof of work, the longest chain with the most accumulated work wins, and subverting the chain requires controlling a majority of the network CPU power. By contrast, a 51% attack on proof of stake would require controlling 51% of all staked ETH. The primary documentation on Ethereum.org notes that proof of stake is more energy-efficient than proof of work and requires less new ETH issuance to secure the network. However, staking carries risks including slashing, software mistakes, and hardware mistakes. Stakers who lack 32 ETH or technical infrastructure often use centralized exchanges or pooled staking services, though delegating custody introduces counterparty risk and leaves users reliant on third-party security.

Start here

1Review the mechanics of consensus rules, slot timing, and validator bonding on [Ethereum.org](https://ethereum.org/).
2Determine whether a given blockchain relies on staked collateral or proof of work by reviewing its official primary documentation.
3Decide between operating an independent validator node requiring 32 ETH or participating through pooled staking options.
4Follow the [How to stake Ethereum](/staking/ethereum-staking-guide) walkthrough before locking capital.

Strengths

Consensus runs without physical mining rigs, making the network more energy-efficient according to primary documentation on Ethereum.org.
Less native coin issuance is required to pay for ongoing network security compared to proof of work systems.
Dishonest actors face programmatic slashing that can destroy collateral, scaling from under 0.1% for isolated errors up to 100% for mass attacks.

Common misunderstandings

Independent validation requires 32 ETH and carries risks of software mistakes, hardware mistakes, and slashing.
Staking pooled assets or holding positions through exchange intermediaries introduces counterparty risk and third-party custody dependence.
Validating rewards vary over time and are never guaranteed by the protocol.

Common questions

What crypto uses proof of stake?

Ethereum is the most widely recognized blockchain that uses proof of stake, having transitioned away from proof of work during the Merge in September 2022. Many networks use proof of stake or related variants to achieve consensus among validators. Readers should consult the official documentation of any specific network to confirm its exact consensus model.

How does proof of stake work?

Proof of stake coordinates network consensus by selecting validators based on locked capital rather than hardware computation. On Ethereum, nodes stake 32 ETH and take turns proposing blocks during 12-second slots using a pseudo-random selection process called RANDAO. Other validators form random committees to attest to proposed blocks, and finality is reached once a two-thirds supermajority approves the state transitions.

Is proof of stake more secure than proof of work?

Proof of stake and proof of work rely on different economic and operational trade-offs rather than a single security standard. In proof of work, as outlined in the Bitcoin whitepaper, miners expend CPU power, and compromising the history requires controlling a majority of hardware computing power. In proof of stake, attacking the chain requires acquiring and risking 51% of all staked ETH, where slashing can, in a mass slashing, destroy up to 100% of the stake.

Can I lose my crypto if I stake it?

Staking involves direct operational and financial risks that can lead to a loss of funds. A validator that breaks the rules faces slashing penalties that range from under 0.1% to 100% of their balance. Additionally, software or hardware mistakes, or trusting third-party custodial pools and centralized exchanges can result in lost assets or counterparty risk.

What is slashing?

Slashing is an automated protocol penalty designed to punish validators that violate core consensus rules. When a validator is slashed on Ethereum, it suffers an immediate balance penalty of 1/4096 of its effective balance up to 0.5 ETH, followed by a correlation penalty on day 18 and complete ejection on day 36. While an isolated mistake costs less than 0.1% of the validator's stake, an incident occurring alongside many other nodes can burn up to 100% of the deposit.

Does proof of stake require mining?

Proof of stake does not require mining or expending CPU power. Instead of miners expending CPU power to find a proof of work, proof of stake validators are chosen pseudo-randomly to propose and vote on blocks based on their deposited coins.

Sources

Ready to put this into practice?

Exchange sign-up bonuses pay both you and a referrer after a qualifying trade.

See bonuses →