Types of Bridges: Trusted vs Trustless

Not all bridges are created equal. The most important distinction in bridge design is the spectrum between trusted and trustless (or trust-minimized) architectures. Where a bridge falls on this spectrum determines its security model, decentralization, and risk profile.

Trusted (Centralized) Bridges

A trusted bridge relies on a centralized entity or a small federation of known parties to custody locked assets and verify cross-chain transactions.

• How it works: Users deposit tokens with a custodian (or multi-sig group), who is responsible for minting wrapped tokens on the other side. Users must trust that the custodian will not steal funds, go offline, or act maliciously.
• Example — Wrapped Bitcoin (WBTC): The original WBTC system relied on BitGo as the custodian holding the underlying BTC. Users trusted BitGo to maintain a 1:1 backing of WBTC to BTC.
• Advantages: Simpler to build, faster transaction finality, easier to maintain, and generally cheaper to use.
• Disadvantages: Single point of failure, custodial risk, censorship risk, and regulatory vulnerability. If the custodian is compromised or goes rogue, all bridged funds are at risk.

Trustless (Trust-Minimized) Bridges

A trustless bridge aims to remove or minimize reliance on any single trusted party. Instead, it relies on cryptographic proofs, economic incentives, or decentralized validator sets.

• Light client bridges: Run an on-chain light client that verifies block headers from the source chain. The destination chain can cryptographically verify that a transaction occurred without trusting anyone. Examples include IBC (Inter-Blockchain Communication) used in the Cosmos ecosystem.
• Optimistic bridges: Assume transactions are valid by default but allow a challenge period during which any watcher can submit a fraud proof. If no challenge is raised, the transaction is finalized. Requires at least one honest participant monitoring the bridge.
• ZK-proof bridges: Use zero-knowledge proofs to verify source chain state on the destination chain. This is the gold standard for trust minimization — the math itself provides the guarantee. Projects like zkBridge and Succinct Labs are pioneering this approach.

The Trust Spectrum

Type	Trust Assumption	Security	Speed
Centralized Custodian	Trust one entity	Lowest	Fastest
Multi-sig Federation	Trust majority of signers	Medium	Fast
Optimistic	At least 1 honest watcher	High	Slow (challenge period)
ZK-Proof	Math / cryptography	Highest	Moderate

Practical Implications for Users

For Sri Lankan crypto users evaluating which bridge to use, the trust model should be a primary consideration. A trusted bridge may be convenient for small amounts, but for significant holdings, a trust-minimized bridge provides far better security guarantees. Always research who controls the bridge's keys, what the validator set looks like, and whether there have been any security incidents.

Remember the crypto adage: "Don't trust, verify." This applies just as strongly to bridges as it does to running your own Bitcoin node.