How a bridge hack happens, step by step

Someone wants to move 1,000 ETH from Ethereum to a rollup. They lock it in a bridge contract, and a wrapped version appears on the other chain. That lock-and-mint step is exactly where the money disappears.

What is a bridge actually doing

A bridge doesn't teleport tokens. It locks or burns an asset on the source chain, then a separate process mints an equivalent asset on the destination chain.

• Lock-and-mint: assets sit in a contract on chain A, a wrapped IOU appears on chain B.
• Burn-and-mint: assets are destroyed on chain A, an equal amount is minted on chain B.
• Liquidity pools: no locking at all, just matched pools on both sides that swap against each other.

Every model needs something to verify that the source-chain event really happened before minting on the destination. That verifier is the attack surface.

Where do the funds actually leak out

Almost every major bridge hack breaks in one of four spots, not from users mismanaging their own funds.

• Validator or signer compromise. The Ronin Bridge lost $625M in March 2022 after attackers took control of 5 of 9 validator signing keys, enough to forge withdrawal approvals without ever touching the underlying contract. Chainalysis has the full writeup.
• Smart contract logic bugs. Wormhole lost $325M in February 2022 because its Solana contract failed to properly verify a guardian signature, letting an attacker mint 120,000 wrapped ETH from nothing.
• Fake deposit proofs. Nomad lost $190M in August 2022 after a routine upgrade set a default "trusted root" to zero, meaning any message looked valid. Once one person figured it out, hundreds of copycat wallets joined in within hours.
• Private key theft. Multichain lost over $125M in July 2023 after keys tied to its founder were reportedly compromised, and funds moved out with no exploit code at all, just signed transactions.

Why does this keep happening on bridges specifically

Bridges concentrate value in one place while spreading trust across systems that were never designed to work together.

• A lending protocol only has to secure its own logic. A bridge has to secure two chains, a messaging layer, and whatever off-chain validators or oracles sit in between.
• Locked collateral sits in a single contract as one large target instead of spread across many positions.
• Cross-chain messages are hard to audit end to end because the two chains have different security assumptions, block times, and finality guarantees.
• Total value lost to bridge exploits has topped $2.8 billion since 2021, according to Chainalysis's 2024 crypto crime report, making bridges one of the single largest categories of hack losses in the industry.

What should you check before bridging funds

You can't audit a bridge's code yourself in five minutes, but you can check who else already tried to.

• Look up how the bridge verifies messages: multisig validators, light clients, or optimistic fraud proofs each carry different risk profiles.
• Check if the validator set is small. Anything under 10-15 signers with public keys is a concentrated target, like Ronin's 9.
• See if the contracts have had a recent audit from a firm you recognize, and whether that audit covered the exact version deployed.
• Check TVL versus payout history. A bridge holding $500M with no bug bounty program is a bigger prize than a smaller, well-monitored one.
• For large transfers, consider splitting funds across multiple bridges or moving in smaller batches instead of one big transaction.

None of this guarantees safety. It just shifts the odds toward bridges that have survived scrutiny longer, which is the closest thing this part of crypto has to a track record.