51% Attack

The primary and most economically damaging action in a 51% attack. An attacker with majority control can:

• Reverse transactions they sent, allowing them to spend the same coins twice.
• Create a private chain where they send coins to an exchange, then reorganize the main chain to erase that transaction after withdrawing fiat.
• This undermines the fundamental immutability and finality guarantees of the blockchain for their own transactions.

The attacker can selectively exclude or censor transactions from being included in new blocks.

• They can prevent specific addresses from sending or receiving funds.
• This can be used for targeted denial-of-service against individuals, businesses, or smart contracts.
• However, they cannot alter or forge transactions from other users, as this requires the sender's private keys.

A critical boundary of a 51% attack. The attacker cannot:

• Spend coins from addresses they do not control, as this requires forging digital signatures.
• Alter the content of existing, confirmed transactions from other users.
• Change the blockchain's protocol rules, such as the block reward or difficulty adjustment algorithm.

The attacker is constrained by the network's native monetary policy.

• They cannot arbitrarily inflate the supply by creating coins out of thin air.
• They are limited to the block rewards and transaction fees available to the miner or validator of each new block.
• This makes sustained attacks economically costly, as they must cover hardware and energy expenses.

Executing a 51% attack is not just a technical challenge but an economic one.

• Cost: Renting or acquiring majority hashrate (Proof of Work) or stake (Proof of Stake) is extremely expensive.
• Detection: Network monitoring services and nodes can detect unusual chain reorganizations, alerting the community.
• Defense: The community can execute a hard fork to reject the attacker's chain, rendering the attack futile if consensus is reached.

Most successful 51% attacks are facilitated by the rental of hash power from services like NiceHash. Attackers can temporarily redirect massive computational power to a target chain at a known cost. The economic calculation is simple: if the cost of renting hash power is less than the potential profit from double-spending, an attack is viable. This makes low-hash-rate chains perpetually vulnerable to this Sybil attack variant.

In response to these attacks, the industry developed several countermeasures:

• Checkpointing: Networks like Ethereum Classic implemented MESS to make deep reorgs computationally prohibitive.
• Exchange Safeguards: Major exchanges increased confirmation requirements for deposits from vulnerable chains and implemented chain monitoring.
• Consensus Shifts: Some projects migrated from Proof-of-Work (PoW) to Proof-of-Stake (PoS) or hybrid models to eliminate hash power-based attacks entirely, as seen with Ethereum's Merge.

A 51% attack occurs when an attacker controls more than 50% of the network's mining power (hashrate) in Proof of Work, or staked assets in Proof of Stake. This majority control allows them to:

• Orphan legitimate blocks by creating a longer, competing chain.
• Double-spend coins by reversing confirmed transactions.
• Censor transactions by excluding them from newly mined blocks.

Successful execution undermines the fundamental security guarantees of a blockchain.

• Loss of Finality: Previously confirmed transactions can be reversed, destroying trust.
• Network Disruption: The attack can halt or significantly slow down legitimate block production.
• Economic Damage: The native token's value typically plummets due to the breach of trust, and exchanges may halt deposits/withdrawals.

The primary defense is making an attack prohibitively expensive. The cost is tied to acquiring the majority of the network's security budget.

• In Proof of Work: Attack cost ≈ cost of acquiring and operating >50% of the global hashrate.
• In Proof of Stake: Attack cost ≈ cost of acquiring >50% of the total staked supply, which the attacker risks having slashed (destroyed) if caught. Larger, more decentralized networks have a higher attack cost.

These attacks are most feasible against smaller, less secure chains.

• Bitcoin Gold (2018 & 2020): Suffered multiple 51% attacks leading to double-spends exceeding $70k.
• Ethereum Classic (2020): Attacked multiple times, with one reorganization depth of over 7,000 blocks.
• Verge (2018): Exploited via a flaw in its algorithm, not pure hashrate, but demonstrated the vulnerability of low-hashrate chains.

Networks and services employ monitoring and response strategies.

• Chain Analysis: Services like Coinbase monitor for deep chain reorganizations.
• Checkpointing: Some chains (e.g., Ethereum Classic post-attack) implement modified consensus with trusted checkpoints.
• Increased Confirmations: Exchanges dramatically increase the required confirmation count for deposits from vulnerable chains.

Understanding a 51% attack requires knowledge of adjacent security models.

• Nothing at Stake (PoS): A theoretical problem where validators have no cost to validate on multiple chains, though mitigated by slashing.
• Long-Range Attack: A PoS-specific attack where an attacker rewrites history from a distant past block.
• Sybil Attack: Creating many fake identities to gain influence, which Proof of Work and Proof of Stake are designed to resist.

The consensus mechanism where a 51% attack is most classically defined. Miners compete to solve cryptographic puzzles to add blocks. Controlling a majority of the network's hashrate allows an attacker to:

• Double-spend coins by reorganizing the blockchain.
• Censor transactions by excluding them from new blocks.
• Halt block production for other miners. Networks like Bitcoin and Ethereum (pre-merge) are secured by the immense cost of acquiring this majority.

A consensus mechanism where validators are chosen based on the amount of cryptocurrency they stake as collateral. The analogous attack is a staking attack, where an entity controls >33% or >66% of the staked tokens to disrupt consensus. Key differences from a PoW 51% attack:

• Attack requires capital ownership, not physical hardware.
• Slashing mechanisms can destroy the attacker's staked funds.
• Defense often involves social consensus and client diversity to reject malicious chains.

A theoretical attack on Proof of Stake networks where an attacker creates an alternative chain starting from a point far back in blockchain history. Unlike a 51% attack, it doesn't require current majority control but exploits weak subjectivity or the ability to acquire old validator keys. Defenses include:

• Checkpoints: Hard-coding recent block hashes in client software.
• Weak Subjectivity Periods: Requiring nodes to sync within a defined recent timeframe.
• Validator key rotation to limit historical key exposure.

An attack where a single adversary creates many fake identities (Sybil nodes) to gain disproportionate influence in a peer-to-peer network. While distinct, it relates to a 51% attack as a foundational threat:

• In peer-to-peer networking, Sybil nodes can eclipse honest nodes.
• In decentralized governance, they can manipulate voting outcomes.
• Consensus mechanisms like PoW and PoS are designed to be Sybil-resistant by tying influence to a costly resource (hash power or stake).

A mining strategy where a miner or pool discovers a new block but keeps it secret, creating a private fork. By withholding blocks, they can gain a revenue advantage over honest miners. This is a subset of a 51% attack strategy:

• It becomes exponentially more profitable with higher hashrate shares.
• At over 33% hashrate, it can be used to double-spend.
• It demonstrates how sub-majority control can still destabilize network incentives and security assumptions.

The total combined computational power dedicated to mining and securing a Proof of Work blockchain. It is the primary security metric against a 51% attack.

• Higher hashrate increases the cost and practical difficulty of an attack.
• Hashrate distribution is critical; concentration in a few large pools increases risk.
• A sudden drop in total hashrate (e.g., after a price crash or China mining ban) can temporarily lower the attack cost, making the network more vulnerable.