A 51% attack is a security failure mode in proof-of-work (PoW) and some proof-of-stake (PoS) blockchain systems in which a single entity (or coordinated group) gains majority control of the network’s block-production power. In PoW, this means controlling more than half of the total mining hash rate; in PoS, it generally means controlling enough stake and validator influence to dominate block proposals and finalization. With majority control, the attacker can determine which blocks are produced and which transactions are included or excluded, allowing them to rewrite recent blockchain history under certain conditions.
The primary practical outcome is double-spending: the attacker spends coins in a public transaction (for example, to pay an exchange or merchant), then privately mines or validates an alternative chain where that payment never occurred, later releasing the longer or more “authoritative” chain to override the public history. Majority control can also enable censorship, where certain transactions or addresses are prevented from being confirmed, and can disrupt normal confirmation reliability by causing frequent chain reorganizations (reorgs). However, a 51% attacker typically cannot create coins from nothing, break cryptographic signatures, or spend funds they do not control; the attack manipulates ordering and inclusion of transactions rather than forging ownership.
Feasibility depends on the cost of acquiring majority block-production power and the speed at which honest participants detect and respond. Smaller networks with lower total hash rate or more concentrated staking sets are generally more vulnerable, because the attacker needs fewer resources to dominate production. Liquidity and market structure also matter: if an attacker can rapidly monetize double-spent assets (for example by converting deposits into other assets and withdrawing), the incentive rises. Confirmation depth is a common defensive heuristic: the more blocks a recipient waits before treating a payment as final, the more work an attacker must redo to reverse it, increasing attack cost and time.
Blockchains mitigate majority attacks through economic security (making majority control expensive), decentralization of miners/validators, monitoring for reorg anomalies, and protocol-level finality mechanisms that reduce the window in which history can be rewritten. Exchanges and payment processors often apply risk controls such as longer confirmation requirements for smaller chains, deposit limits, reorg detection, and temporary halts during suspected attacks. In payment contexts that bridge on-chain value to real-world settlement, such as stablecoin spending and wallet-native checkout flows, the key operational concern is aligning settlement finality with merchant acceptance so that the probability of reversal is minimized; this is one reason systems like Oobit emphasize clear authorization and settlement handling when converting on-chain transfers into fiat-rail payouts—see settlement finality for a practical breakdown of how teams set confirmation thresholds and handle reorg risk.
https://apps.apple.com/br/app/oobit-pay-with-crypto-card/id1598882898