Bridges & Their Risks
Chains can’t natively talk to each other, so bridges move value between them — by locking a pile of assets that becomes the juiciest target in crypto.
You’ve got funds on one chain (say an L1) and want to use an app on another (say a rollup from lessons 31–32). Problem: chains can’t natively see each other. A blockchain only knows about its own ledger — it has no built-in way to know what happened on a different chain. So how does value cross between them? The answer is a bridge — and bridges have been, by a wide margin, the single most catastrophic source of hacks in all of crypto, with billions lost. Understanding why is a masterclass in where risk concentrates. Hold the question: if two chains can’t see each other, how could you possibly move a coin from one to the other — and what does that mechanism put at risk?
You don’t move a coin — you lock and mint
The key realization: a coin never actually leaves its home chain. A bridge uses a lock-and-mint trick. To move 10 coins from Chain A to Chain B: (1) you send the 10 coins into the bridge’s contract on Chain A, which locks them (they’re frozen, not spent); (2) the bridge then mints 10 equivalent “wrapped” tokens on Chain B — new tokens that are claims on the locked originals. To come back, you burn the 10 wrapped tokens on B, and the bridge unlocks your originals on A. The originals sit locked the whole time; the wrapped version is just an IOU that’s redeemable 1-for-1. Nothing teleported — value was represented on the other side while the real thing stayed frozen.
Bridging 10 coins A → B and back: • Lock: send 10 A-coins to the bridge contract on Chain A → they’re frozen there. • Mint: bridge creates 10 wrapped-A tokens on Chain B → you spend/use those. • Return: burn the 10 wrapped-A on B → bridge unlocks your 10 originals on A. • At all times, the 10 real coins are locked on A; the wrapped tokens are just redeemable claims.
Why bridges are a honeypot
Notice what lock-and-mint requires: a big pile of real assets locked in one place — the bridge contract — backing all the wrapped tokens out there. That pile is a honeypot: concentrate hundreds of millions of dollars behind a single contract and every attacker on earth focuses on that one target (recall “where the value concentrates, attackers follow” from security, lesson 17). Worse, the wrapped tokens are only worth anything if the lock holds — so if an attacker drains the locked assets, the wrapped tokens on the other chain become backed by nothing, instantly worthless, hitting everyone holding them. Bridges also tend to be complex (two chains, off-chain relayers watching for events) and often rely on a small set of validators attesting “yes, the lock happened” — and if you can forge or compromise those attestations, you can mint wrapped tokens without locking anything. Big value + big complexity + concentrated trust = the perfect storm that’s produced crypto’s largest hacks.
Reading a bridge’s trust model
So how do you reason about a specific bridge? Ask who is trusted to confirm the lock, because that’s the weak point. On the riskier end, a bridge relies on a small external committee of validators — compromise them and the whole thing falls (this is essentially the custody question from lesson 18, wearing a bridge costume). On the safer end, a bridge is trust-minimized: it verifies the other chain’s events using cryptographic proofs the destination chain can check itself, or (for a rollup bridging to its own L1) it inherits the L1’s security directly. Rule of thumb: the fewer extra people you must trust beyond the two chains themselves, the safer the bridge. And a blunt honest takeaway that seasoned users internalize — don’t leave more on a bridge, or in wrapped form, than you’d be willing to lose, and prefer bridges whose trust model you can actually name. If you can’t say who’s trusted, you are the exit liquidity.
Think of a coat check that works across two buildings. You hand your coat (the real coin) to the attendant in Building A, who locks it in a closet and gives you a numbered ticket (the wrapped token) you can carry into Building B. The ticket is only valuable because a real coat backs it. Now imagine that closet holds ten thousand expensive coats behind one lock — every thief in the city is eyeing that closet, not individual pockets. If they crack it and steal the coats, everyone’s tickets in Building B become worthless paper. And if the attendant can be bribed to hand out tickets for coats that were never checked in, the whole system inflates into fraud. That closet is a bridge: concentrated value, single point of failure.
Comparing two bridges’ risk: 1. Bridge X: a 5-member committee signs off that deposits happened; 5 keys control hundreds of millions. → Compromise 3–5 keys and mint unbacked wrapped tokens. High, concentrated trust. 2. Bridge Y: the destination chain cryptographically verifies the source chain’s events itself, no external signers. → Far fewer people to trust; much stronger model. 3. A rollup’s “canonical bridge” to its L1 inherits L1 security directly — among the safest. 4. Same job (move value across chains), very different risk — set by who confirms the lock.
This is the reading. The interactive version — active-recall quiz, a hands-on experiment you run in your own AI, and an earned mastery check — is free in the app.
Start this lesson free →