Fidelity Bond

A fidelity bond is a time-locked BTC deposit used as an anti-sybil mechanism. The participant commits to having capital locked up for a period; this makes spinning up many fake identities expensive, which makes attacking a privacy protocol economically unattractive.

The canonical use case is JoinMarket - a decentralized CoinJoin coordination protocol. JoinMarket's matching market lets participants act as either "makers" (provide liquidity, earn fees) or "takers" (pay for the mix). Without anti-sybil measures, an attacker could spin up many fake makers, all controlled by the same entity, and dominate the matching process to compromise the mix.

Fidelity bonds fix this. A maker can prove their commitment by locking BTC in a time-locked output - typically a CLTV-protected UTXO that can't be spent until some future block. Takers prefer to mix with makers who have larger bonds locked for longer, because:

• An attacker would need to lock real BTC to spin up convincing fake identities.
• Larger bonds + longer locks = higher attack cost for the same level of sybil capability.
• Bond holders have skin in the game - they can't recover the BTC for the lock period, so they're committed to behaving consistently.

The capital isn't lost; it's just unspendable for the lock period. The attacker bears the opportunity cost of capital tied up. At scale, that opportunity cost becomes prohibitive.

This concept generalizes. Similar economic-stake mechanisms appear in:

• Lightning Network's channel funding (channel parties have skin in the game).
• Proof-of-stake systems (validators stake to participate).
• Some federated systems where committee members post bonds.

Fidelity bonds are a clean example of how Bitcoin's primitives (time locks, public verifiability) can be composed to solve adjacent problems like sybil resistance, without changing the base protocol.