Validators lock tokens and are selected, with probability weighted by the size of the stake, to propose and attest to blocks. Correct behaviour earns rewards; provable misbehaviour such as signing two conflicting blocks has part of the stake destroyed. The cost of attacking the chain is therefore borne in the network's own token rather than in electricity.
Compared with proof of work it consumes a tiny fraction of the energy and allows faster finality, since validators vote on blocks rather than racing to find them. The recurring criticism is concentration: influence is proportional to holdings, and large staking providers and exchanges accumulate a substantial share of the stake on most major networks.
Is proof of stake more secure than proof of work?
It is differently secured. Proof of work makes attacks cost external resources; proof of stake makes them cost the attacker's own capital. Proof of stake is far more energy-efficient and generally reaches finality faster.
Last updated October 17, 2026