Bridges

Nothing crosses the gap

When an asset is bridged, it is usually locked or burned in one system and represented by a newly issued claim in another. The destination chain cannot directly verify all events on the source. It relies on a mechanism—multisignature signers, light-client proofs, validators, optimistic challenges, or some hybrid—to accept a statement.

Several security models share one word

“Bridge” covers designs with radically different assumptions. A small signer set can act quickly but concentrates key risk. A light client can verify more of the source consensus but costs more. An optimistic bridge creates a challenge window and needs vigilant watchers. Counting bridges or validators without naming the trust model compares unlike things.

Attacks occur between valid ledgers

In major bridge failures, each chain could continue producing internally valid blocks. The break happened in the assertion joining them: compromised keys, faulty verification, a bad initialization, or an accepted message that should never have been accepted. Two correct local histories can be connected by a false sentence.

Interoperability is an institution

Bridges are often presented as plumbing, but they decide whose observation crosses a boundary and who can halt, upgrade, reimburse, or abandon the system. Their true product is not frictionless movement; it is governed belief between ledgers. Security begins when that governance is named rather than concealed behind an arrow in a wallet.