What is a multisig wallet, and what does it protect?

A single private key is a single point of failure. Lose it and the funds are gone. Leak it and someone else can move them. The most common answer that exchanges, protocol teams, and treasuries give to that problem is a multi-signature wallet (multisig). It works as designed. It also has limits that two of the largest thefts in crypto history made very public.

What is a multisig wallet?

A multisig wallet is a wallet that needs approval from several separate keys before it can move funds. The rule is written as a threshold: a two-of-three wallet has three keys and needs any two of them to sign, while a five-of-nine wallet needs five of its nine. Each key is usually held by a different person or device, so no single holder can move anything alone.

How does a multisig wallet work?

On Bitcoin, multisig is built into the scripting rules. An address can require several signatures before its coins can be spent. On Ethereum and similar chains, a multisig is usually a smart contract: a small program that lives on the blockchain and holds the funds itself. The program keeps a list of the people allowed to sign and the number of approvals it needs. It checks each signature, counts the approvals, and releases the money only when there are enough.

In both designs, the rule is enforced on-chain rather than by a company's internal policy. A transaction carrying too few valid signatures is simply rejected by the network or the contract.

What does a multisig protect against?

A multisig protects against the loss or theft of any one key. If a signer's laptop is compromised, the attacker holds one key out of several, and still cannot move funds. If a signer loses a key, the others can still meet the threshold. On most smart-contract multisigs, they can also vote to replace the lost key.

A multisig also guards against a single insider acting alone. One person cannot send the money on their own. Several people have to agree first. This is why multisigs commonly sit behind exchange cold wallets, project treasuries, and the admin keys that can pause or upgrade a protocol's contracts. Our note on what a governance token actually controls (opens in a new tab) covers that last case.

What does a multisig not protect against?

A multisig does not protect against signers who are deceived into approving the wrong transaction. The threshold counts signatures. It cannot tell whether the people signing understood what they signed for.

In February 2025, exactly this happened to the exchange, Bybit, which lost about US$1.5bn in ether from a multisig cold wallet. According to post-incident analyses, attackers had tampered with the web interface the signers used to review transactions. The signers saw what looked like a routine internal transfer. What they approved handed control of the wallet's contract to the attackers. No key was stolen, and every signature was genuine.

Hardware wallets did not catch it either. Many show only a string of characters for a complex contract call, rather than a readable description, a practice known as blind signing. When a signer cannot read what the device is approving, they end up trusting whatever the screen beside it says.

What happens when the keys are not really separate?

A multisig is only as independent as the people, and systems holding its keys. If several keys sit with the same organisation, on the same network or behind the same login, one breach can reach all of them at once.

The Ronin bridge hack of March 2022 shows how. Ronin's bridge needed five of nine validator signatures to approve a withdrawal. According to Ronin's own post-mortem, four of those keys belonged to one company, Sky Mavis. Sky Mavis also still held a temporary permission to sign for a fifth validator. That permission had been granted in November 2021, and never revoked.

Breaching one company therefore gave the attacker five of nine keys. Roughly 173,600 ether, and 25.5 million USDC were drained through forged withdrawals on 23 March 2022. Nobody noticed until 29 March, when a user found they could not withdraw.

The threshold itself carries a trade-off. Set it too low, and a few compromised keys are enough to move funds. Set it too high, and losing a handful of keys can lock the funds for good.

How is a multisig different from an MPC wallet?

A multi-party computation (MPC) wallet splits one private key into shares held by several parties. Those parties jointly produce a single ordinary signature without the full key ever being assembled in one place. A multisig instead collects several separate signatures that the chain counts against a published rule.

The practical difference is visibility. A smart-contract multisig's signers, and threshold can usually be read on a block explorer (opens in a new tab), while an MPC wallet looks like an ordinary single-key wallet from outside. Both share the same blind spot: neither can stop a signer from approving a transaction they were misled about.

Where Northtape fits

Northtape holds no private keys, and sits on no signing threshold. It is a news desk. Stories about hacked treasuries, and drained bridges are detected by the Risk Radar protocol lens. For how bridge withdrawals get approved, see our note on what a bridge hack is (opens in a new tab).

FAQs

Is a multisig wallet safer than a single-key wallet? It removes the single point of failure, so losing or leaking one key is no longer enough to lose funds. It adds no protection against signers who are tricked into approving a malicious transaction.

What happens if a multisig signer loses their key? If the remaining signers still meet the threshold, they can usually move the funds or replace the lost key. If too many keys are lost to reach the threshold, the funds cannot be moved.

Do individuals use multisig wallets? Some do, often spreading keys across several hardware devices kept in different places. The trade-off is complexity: more keys to store, back up and keep track of.

Can you tell from outside that a wallet is a multisig? On Ethereum a smart-contract multisig's signers and threshold can usually be read on a block explorer. On Bitcoin the spending conditions become visible once coins are spent from the address.

None of this is investment advice. It describes how multisig wallets work, not whether any exchange, protocol, or wallet that uses one is a safe place to hold funds.