Crypto isn't one network. Bitcoin, Ethereum, Solana, Arbitrum, Base, Polygon — each is a separate blockchain with its own version of any given asset. The "USDC on Ethereum" in your wallet is a different on-chain entity than the "USDC on Solana" that might be on the exchange you want to use.

If you want to move value between chains, you need a bridge. Bridges are the connecting tissue of the multi-chain crypto world. They're also where most of the largest crypto hacks have happened.

Here's what you need to know.

What a bridge actually does

Bridges between blockchains: how they work, what to watch for (education)

The simplest version: when you bridge tokens from Chain A to Chain B, the bridge:

  1. Locks your tokens on Chain A — they get sent to a smart contract that holds them.
  2. Mints an equivalent amount on Chain B — the bridge's contract on B issues a new "wrapped" version of the token.
  3. The wrapped tokens are redeemable — if you bridge back, the wrapped tokens on B get burned and the originals on A are unlocked.

This means: when you have "USDC" on Arbitrum, what you actually have is the bridge's IOU. It's redeemable for real USDC on Ethereum mainnet, but it requires the bridge to be working and not exploited.

Why bridges are dangerous

The bridge contract is holding a lot of money — sometimes hundreds of millions of dollars. If someone finds a bug in the contract, they can drain the entire pool.

The list of bridge exploits is long and expensive:

  • Ronin Bridge (March 2022): $625M lost. Validators compromised.
  • Wormhole (February 2022): $325M lost. Smart contract bug.
  • Nomad (August 2022): $190M lost. Initialization bug.
  • Multichain (July 2023): $130M+ lost. Founder went missing with private keys.
  • Poly Network (August 2021): $610M lost (returned voluntarily after).
  • Orbit Bridge (January 2024): $82M lost.

This isn't a complete list. Bridge security is genuinely an unsolved problem in crypto.

How to use bridges safely

The realistic approach for non-technical users:

1. Prefer "canonical" bridges

For every L2, there's a "canonical" bridge run by the protocol team itself. Arbitrum has its own bridge. Optimism has its own. Base has its own.

These are usually safer than third-party bridges because:

  • The team has the most incentive to keep it secure
  • Audits are more thorough
  • Recovery procedures exist if something goes wrong

Use them: bridge.arbitrum.io, app.optimism.io/bridge, bridge.base.org, etc.

2. Avoid bridges to obscure chains

If you're bridging from Ethereum to Arbitrum, the canonical bridge is heavily used and well-tested. If you're bridging from a small alt-L1 to an even smaller L2, the bridge might be a small team's hobby project. Risk goes up sharply.

3. Don't keep large amounts in bridged form

If you bridged $50K of USDC to a chain and you're not actively using it there, bridge it back. The bridge holding the assets is a continuous exposure. The exposure is finite when you bridge in and out as needed.

4. Watch for "wrapped" tokens with no exit

Some chains have wrapped versions of tokens where the bridge has been compromised or abandoned. The wrapped token may still trade on the chain, but you can't redeem it for the underlying. Common red flag: a "USDC" or "BTC" on a smaller chain that trades at a discount to the real version.

If something trades at a discount, the discount usually reflects bridge risk.

5. Use legitimate aggregators if you're bridging often

Services like Across, Stargate, and Squid aggregate across multiple bridges, sometimes using their own clever mechanics to reduce risk. They're not magic, but they're usually safer than picking an unknown bridge from a random Discord recommendation.

What bridging actually costs

For a typical bridge from Ethereum to Arbitrum:

  • Gas fee on Ethereum to initiate: $3-15, depending on network congestion
  • Time to complete: 10-30 minutes for the canonical bridge
  • Reverse: Arbitrum to Ethereum: 7-day fraud-proof delay for the canonical bridge (or pay extra for a fast-exit service)

The 7-day delay is the part most people don't expect. The canonical bridge for an "optimistic rollup" (Arbitrum, Optimism, Base) has a 7-day challenge period before withdrawals finalize. Fast exits skip this by paying liquidity providers a fee.

For Solana ↔ Ethereum bridges, times are typically 1-5 minutes and the security model is different.

A reasonable user pattern

For most people:

  • Bridge in batches, not for every transaction
  • Use canonical bridges when possible
  • Don't bridge "for fun" or to small chains you won't actively use
  • Don't keep more value bridged than you're actively using

If you find yourself bridging often and complaining about fees, that's a sign to consolidate your activity on fewer chains.

The honest summary

Bridges are useful but they're the riskiest piece of crypto infrastructure. The largest exploits have been bridge exploits. The pattern doesn't seem to be improving.

The defensive playbook: prefer canonical bridges, prefer fewer chains, don't keep large amounts in wrapped form for long. Most users don't need to bridge as much as they think they do.

Crypto is volatile. Bridges add a layer of smart-contract risk on top of price risk. None of this is financial advice.