By - A bridge does not move a coin between chains. It locks the asset on one chain and issues a claim on another.
- That lock-and-mint design concentrates value in one contract, which is why bridges account for several of the largest losses in crypto history.
- The three recurring failure modes are validator compromise, signature verification bugs, and price oracle manipulation.
A crypto bridge is the plumbing that lets an asset on one blockchain be used on another. It is also, consistently, where the largest single losses in the industry happen. Understanding why requires understanding what a bridge actually does, which is not what most people assume.
What is a crypto bridge, in plain English?
Nothing crosses. That is the first thing to fix.
If the underlying ledgers are unfamiliar, our guide to how blockchain technology works covers the foundation this sits on.
Bitcoin cannot exist on Ethereum. Ethereum cannot exist on Solana. Each chain is a closed ledger that only knows about its own state. When you “bridge” an asset, the original never leaves its home chain.
What happens instead is a two-step. The bridge takes custody of your asset on the source chain and locks it in a contract. It then issues, or mints, a representative token on the destination chain. That new token is a claim: a promise that the locked original can be redeemed.
Wrapped Bitcoin on Ethereum works exactly this way, a pattern documented in Ethereum’s own developer materials. Every WBTC is backed by a bitcoin held somewhere else. The token is not bitcoin. It is a receipt for bitcoin.
How does the bridge know the lock happened?
This is the hard part, and it is where bridges differ from one another.
The destination chain cannot read the source chain. Something has to observe the lock and attest to it. That something is the bridge’s trust model, and it comes in roughly three shapes.
External validators. A set of named parties watch the source chain and sign off that a deposit occurred. Fast, cheap, and only as trustworthy as the signers. If enough of their keys are compromised, the attacker can authorise mints that were never backed.
Light clients. The destination chain runs a minimal verification of the source chain’s consensus in a contract. Much stronger, because the proof is cryptographic rather than social. Also far more expensive and harder to build, which is why fewer bridges do it.
Liquidity networks. No minting at all. Pools of the real asset sit on both chains, and the bridge pays out of the destination pool while taking your deposit into the source pool. Limited by available liquidity, but nothing synthetic is created.
Why do crypto bridges keep getting hacked?
Three reasons, and they compound.
The value is concentrated and unattended. A bridge contract holding a billion dollars is a single target with a permanently public address. Unlike an exchange, there is no operations team watching a dashboard at 3am, and no ability to freeze withdrawals by hand.
Verification is subtle code. Most bridge exploits are not clever economics. They are bugs in the function that checks whether a deposit proof is valid. Accept a malformed proof and the attacker mints unbacked tokens directly. The contract does what it was told; it was told wrong.
The trust model is often smaller than advertised. A bridge described as decentralised may depend on a validator set of nine, or a multisig of five. Compromise the keys and the cryptography is irrelevant, because the attacker is producing valid signatures.
What does a bridge failure actually look like?
Usually not a dramatic break-in. Usually a mint that should not have been possible.
The attacker submits a withdrawal or mint request with a proof the contract accepts but should not. New tokens appear on the destination chain with nothing locked behind them. The attacker sells them into whatever liquidity exists, and the remaining holders of the wrapped asset are left holding claims against a reserve that is now short.
The tell is a rapid, one-directional outflow from a contract that normally sees balanced two-way traffic. By the time it is visible on a dashboard, the funds are usually several hops away.
How should you think about bridge risk?
Ask three questions before using one.
Who attests, and how many of them are there? If the answer is a multisig, the number of signers is your real security assumption. Treat “decentralised” as marketing until you see the count.
How long has the contract held value without incident? Time under adversarial conditions is the only meaningful audit. A contract that has held nine figures for two years has been attacked continuously and survived.
What are you actually holding afterwards? A wrapped asset is a credit position against the bridge, not the underlying. If the bridge fails, the wrapped token does not become the original. It becomes worthless.
Chain-level interventions are not hypothetical either. When a lending exploit hit Cronos in August 2026, validators a small validator set can act in hours to reverse it, which tells you how much of this infrastructure still depends on a small group of operators agreeing to act.
Industry loss tallies are published by firms such as Chainalysis, and bridges appear disproportionately in them.
The practical advice is unglamorous: bridge only what you need, only when you need it, and do not treat a wrapped asset as a place to sit. The design that makes bridges useful is the same design that makes them the most attacked infrastructure in crypto.
What's your reaction?
