TL;DR Crypto bridging lets you move tokens from one blockchain to another using protocols that act as intermediaries between networks that can't communicate directly. Bridges are essential multi-chain infrastructure, but billions have been stolen from bridge exploits, making them one of the most targeted areas in crypto.
You've got ETH in your Ethereum wallet. You want to use a DeFi protocol on Arbitrum because the fees are a fraction of mainnet. Just send it over, right?
Blockchains don't work like that. Ethereum doesn't know Arbitrum exists. Solana can't read Polygon. Each network runs its own consensus, its own transaction history, its own rules. That isolation is what makes each chain secure, but it also means your assets are stuck wherever they started.
Crypto bridges solve that problem by creating a way to move tokens between chains that otherwise have zero connection. But bridges are also one of the most attacked pieces of infrastructure in crypto, and picking the wrong one can empty your wallet fast.
Below, we break down how bridges actually work and how to move assets between chains without taking on more risk than you need to.
→ Blockchains are isolated. You need a bridge protocol to move tokens between them.
→ Security varies wildly. Native L2 bridges inherit Ethereum's security. Third-party bridges carry more risk.
→ Over $2.5 billion has been stolen from bridge hacks since 2021. They're the highest-value targets in DeFi.
→ You can bridge safely with test transactions, verified URLs, and established protocols.
→ The future is zero-knowledge proofs, intent-based systems, and native chain interoperability are all working to reduce bridge risk.
Every blockchain is its own closed system. Ethereum has its own set of validators, its own way of reaching consensus, its own record of every transaction that's ever happened on it. Solana has a completely separate set. So does Arbitrum, Polygon, Avalanche, and every other chain.
This is deliberate. Isolation is what keeps each network secure. If Ethereum's validators had to trust information from Solana's validators, both chains would inherit each other's weaknesses. Keeping things separate means a bug on one chain can't cascade into another.
The trade-off is usability. You might hold ETH on Ethereum mainnet but want to trade on a DeFi protocol that only exists on Arbitrum. Or you've got stablecoins on Polygon but need them on Base.
Without some way to connect these networks, you're stuck going through a centralized exchange every time you want to move between chains, selling on one, withdrawing, depositing on another, and paying fees at every step.
That friction is the core problem bridges were built to solve.
A crypto bridge is a protocol that moves tokens from one blockchain to another. It sits between two networks that can't communicate on their own and handles the transfer so your assets end up where you need them.
People mix these up. When you use an exchange, you're selling one asset and buying another. You sell ETH, you get SOL. Two separate transactions, two separate assets.
A bridge is different. You're moving the same asset (or a direct representation of it) from one chain to another. You start with ETH on Ethereum, you end with ETH (or wrapped ETH) on Arbitrum. Nothing gets sold.
This depends on the bridge type.
Lock-and-mint bridges lock your original tokens in a smart contract on the source chain. Then they mint wrapped tokens on the destination chain. Want to bridge back? The wrapped tokens burn, and your originals unlock.
Liquidity pool bridges skip the wrapping entirely. The bridge holds reserves on both chains. You deposit into the source pool, withdraw from the destination pool. You get native assets, not wrapped versions.
Most bridging falls into a few categories:
Moving ETH to Layer 2s for lower fees,
Accessing DeFi protocols that only exist on specific chains
Transferring stablecoins between networks
Getting tokens that are only available on a particular network.
We touched on lock-and-mint and liquidity pools above. Now let's walk through exactly what happens when you use each one.
This is the most common type. Five steps:
Pick your route. Select the source chain, destination chain, and how much you're bridging.
Tokens get locked. Your tokens go into a smart contract on the source chain. They don't move. They sit there as collateral.
Validators confirm the lock. Bridge validators (or relayers) check that the deposit actually happened.
Wrapped tokens are minted. The bridge creates an equivalent amount of wrapped tokens on the destination chain.
Tokens land in your wallet. You now hold wrapped versions that represent your originals 1:1.
Want to come back? Reverse it. Wrapped tokens burn. Originals unlock. Done.
No wrapping here. The bridge holds reserves of tokens on both chains, funded by liquidity providers who earn fees for keeping the pools stocked.
You deposit tokens into the pool on the source chain.
The bridge releases an equal amount from its pool on the destination chain.
You receive native tokens. No minting, no burning, no wrapped anything.
Faster. Simpler. You're holding real tokens, not representations. But there's a limit: you can only bridge what the pool can cover. If a pool runs dry or you're moving a whale-sized amount, you'll hit slippage or get blocked entirely.
Someone has to confirm your tokens were actually locked before anything gets minted on the other side. Who that someone is matters a lot.
Trusted validators: A single entity or small group signs off on transactions. Fast. Cheap. But if they get compromised, the whole bridge is exposed. This is exactly how the Ronin Bridge lost $624 million.
Decentralized networks: Multiple independent validators have to agree before anything goes through. Slower, but there's no single point of failure.
Light client bridges: Use cryptographic proofs to verify transactions without trusting anyone. Most secure. Most complex. Most expensive to run.
Most bridges today still lean toward the trusted end. Few are genuinely trustless.
Type | Examples | Security | Speed | Best For |
Native L2 | Arbitrum Bridge, Optimism Bridge, Base Bridge | Highest (inherits Ethereum security) | Deposits fast, withdrawals up to 7 days | Large amounts, ETH to L2 transfers |
Third-Party | Across Protocol, Stargate, Wormhole | Varies (own validator sets) | Usually fast | Connecting multiple chains, uncommon pairs |
Trusted | Centrally operated bridges | Depends on operator | Fast | Convenience, smaller amounts |
Trustless | ZK-proof bridges, light client bridges | High (cryptographic verification) | Slower | Maximum security, no trust assumptions |
Trusted bridges rely on operators to verify transactions. Faster, simpler, but you're betting on those operators staying honest and secure.
Trustless bridges use cryptographic proofs instead. Less trust required, more complexity under the hood. In practice, most bridges sit somewhere on a spectrum between the two. Fully trustless is rare.
Native bridges are built by the blockchain teams themselves. Arbitrum Bridge, Optimism Bridge, Base Bridge. They're tightly integrated with the chain, have aligned incentives, and are generally the safest option. Downside: they usually only connect one chain pair.
Third-party bridges like Across Protocol and Stargate connect dozens of chains. More flexible, but they introduce their own validator sets and smart contracts, which means more risk.
Layer 2 bridges move assets between Ethereum and its rollups. They inherit Ethereum's security, making them the safest category. Trade-off: withdrawals back to mainnet can take up to seven days.
Cross-chain bridges connect independent blockchains that share no security model. Much higher risk profile.
Some bridges only handle one asset. They do one thing, and the narrower scope means fewer things can go wrong.
Generalized bridges support any token across multiple chains. More useful, but more code, more complexity, and a bigger attack surface for hackers to poke at.
Rule of thumb: use native bridges when available, especially for large amounts.
These are built by the blockchain teams themselves and inherit Ethereum's security model.
Arbitrum Bridge | ETH ↔ Arbitrum | Trustless. 7-day withdrawal window. |
Optimism Bridge | ETH ↔ Optimism | Same optimistic rollup model as Arbitrum. |
Base Bridge | ETH ↔ Base | Coinbase's L2. Built on the OP Stack. |
Polygon Bridge | ETH ↔ Polygon PoS | Plasma-based security. |
These connect multiple chains but rely on their own security models.
Across Protocol uses an optimistic design and has maintained a strong track record. Fast transfers, primarily focused on Ethereum and L2s.
Stargate (built on LayerZero) handles native asset bridging across major EVM chains and Solana. No wrapped tokens. Charges a flat 0.06% fee.
Wormhole connects 30+ chains and was the only bridge unconditionally approved by Uniswap's Bridge Assessment Committee, though it did suffer a $321 million exploit in 2022. It has since overhauled its security model.
Circle CCTP is the official way to move USDC between chains. If you're bridging USDC specifically, this is the lowest-risk option.
WBTC remains the dominant way to use Bitcoin on Ethereum, though it's centralized (custodied by BitGo). tBTC offers a decentralized alternative but with less liquidity.
Bridge security changes fast. Always check recent audits and incident history before trusting any protocol with your funds.
Bridges hold massive amounts of crypto locked in smart contracts. All those tokens sitting as collateral for wrapped assets create concentrated pools of value. For hackers, that's a single target worth hundreds of millions.
Unlike attacking a blockchain directly (which would require overwhelming its entire node network), attacking a bridge often means finding one flaw in one contract or compromising a handful of validators.
Year | Bridge | Amount Lost | What Went Wrong |
2022 | Ronin (Axie Infinity) | ~$624M | Attackers compromised 5 of 9 validators via social engineering. Took 6 days to notice. |
2022 | Wormhole | ~$321M | Smart contract bug let the attacker mint 120,000 wETH without depositing anything. |
2022 | Nomad | ~$190M | A contract flaw let anyone fake a valid withdrawal. Copycats piled in once word spread. |
2022 | BNB Bridge | ~$566M | Forged withdrawal proof exploited a bug in Merkle tree verification. |
2023 | Multichain | ~$126M | CEO's compromised private keys gave access to bridge funds. |
The pattern is consistent: validator compromises, contract bugs, and management failures.
Validator compromise is the big one. If an attacker controls enough validators to meet the signing threshold, they can authorize withdrawals for funds that were never deposited. Ronin and Orbit Chain both fell this way.
Smart contract bugs come from the sheer complexity of bridge code. Contracts handling cross-chain logic are harder to audit than standard ERC-20 token contracts, and a single overlooked flaw can drain everything.
Oracle manipulation exploits price feeds to trick bridges into releasing more value than was deposited. Flash loan attacks make this worse.
Economic attacks target liquidity pool imbalances, exploiting slippage in shallow pools or draining reserves faster than they can rebalance.
Every bridge is balancing three things: security, speed, and cost. You can optimize for two but not all three.
A fully trustless bridge using cryptographic proofs is very secure but slow and expensive to run. A trusted bridge with a small validator set is fast and cheap, but one compromise away from disaster. Liquidity pool bridges can be fast and reasonably secure, but cost more when pools are thin.
Where a bridge sits on that spectrum tells you a lot about its risk profile.
Use native bridges for large amounts. Check whether the bridge has been audited and by whom. Look up its hack history. If there's an insurance or security fund backing the protocol, that's a good sign. If you can't find any of this information, that tells you something too.
Verify the URL. Phishing sites that clone bridge interfaces are everywhere. Bookmark the official site. Don't trust links from social media or DMs.
Connect your wallet. For anything substantial, use a hardware wallet. Hot wallets are fine for small amounts.
Double-check everything. Source chain, destination chain, token, amount, receiving address. One wrong click, and your funds go somewhere you can't recover them from.
Send a test transaction first. Bridge $10-50 and wait for it to arrive before sending the rest.
Save your transaction hash. Track it on a block explorer so you can prove what happened if something goes wrong.
Know the wait times. Deposits to L2s are usually minutes. Withdrawals back to the Ethereum mainnet via native bridges take up to 7 days.
Walk away if a bridge asks for KYC or your email (legitimate bridges don't need either), requests your seed phrase or private keys (no bridge ever should), promises zero fees or instant transfers on routes that normally take time, has no audit history you can verify, or supports an unusual chain pair with no clear reason why.
Confirm your tokens arrived on the destination chain. Verify the token contract address matches what you expected. Save transaction records for tax purposes.
And if you no longer need assets on that chain, bridge them back rather than leaving funds scattered across networks you're not actively using.
Bridges aren't optional if you're doing anything beyond holding blue chip crypto on a single chain. The moment you want lower fees on an L2, access to a protocol on another network, or better yield farming rates somewhere else, you're going to need one.
The key is treating them with the respect they deserve. Use native bridges for large transfers. Stick to protocols with clean audit histories. Send test transactions. Verify URLs. Don't bridge more than you can afford to lose on an unproven protocol.
Bridges have improved since the wave of exploits in 2022, and ZK-proof technology is closing the trust gap further. But "better" doesn't mean "risk-free." Stay informed, stay cautious.
Now you know how to move assets between chains without getting burned. Ready to sharpen the rest of your toolkit?
LearningCrypto pairs live on-chain data with an AI assistant and education built by crypto educators who've been through multiple market cycles.
Join the community on Discord. Track smart money. Build knowledge that holds up when markets move.
It depends entirely on which bridge you use. Native L2 bridges inherit Ethereum's security and have strong track records. Third-party bridges carry more risk. Due diligence matters. Check audit history, look for past incidents, and start with small test transactions.
Swapping exchanges one token for another on the same chain (like trading ETH for USDC on Uniswap). Bridging moves the same token (or a wrapped version of it) from one blockchain to another. Different problems, different tools.
Usually, yes. You'll pay gas on the source chain to initiate the bridge transaction. On the destination chain, some bridges cover the gas for you, but many require you to already hold a small amount of the native token to interact with the received funds. Plan ahead.
Tax treatment varies by jurisdiction and isn't fully settled. Some regulators view the lock-and-mint process as disposing of one asset and acquiring another, which could trigger capital gains. Others don't treat it as taxable if there's no economic gain. Keep records of every bridge transaction either way. This isn't financial advice; check with a tax professional
Disclaimer: This article is for educational purposes only and does not constitute financial advice. Cryptocurrency investments carry risk; you should always do your own research before making any investment decisions.