A bitcoin and a claim on a bitcoin

The whitepaper does not hide the trust model. It states that custodians are trusted in the wrapped framework, because assets could be stolen or the custodian might not honour the one-to-one backing, and then lists the ways the design tries to minimise that trust. Those mitigations are worth reading carefully, because they describe the shape of the risk rather than removing it.

The first mitigation is that the custodian cannot mint on its own: creation of new tokens requires a merchant to initiate it, so two separate institutions are involved in every mint. The second is that the user is insulated from the custodian by the merchant layer, so an individual merchant need not be trusted but the set of merchants together must be. The third is external audit and proof of reserves, with the whitepaper describing quarterly third-party audits and reserve proofs published as signatures from the addresses holding the bitcoin. The fourth is reputational: the whitepaper notes that the existing credibility of the institutions involved is at stake.

Each of these is a real control and none of them is a guarantee. A proof of reserves shows that the custodian controls addresses holding at least the stated amount at a point in time; it does not prove that the bitcoin is unencumbered, that it will still be there tomorrow, or that redemption will be honoured on demand. The whitepaper's own framing is that the framework aims to minimise trust, not to eliminate it, and a holder should read the difference literally.

Custodial risk is the first and the largest. The bitcoin backing the token is held by an institution, and the holder's claim depends on that institution's solvency, security and willingness to redeem. This is a different risk from holding bitcoin directly, where the holder controls the keys and no third party stands between them and the asset. It is also a risk that cannot be diversified away by holding more of the token, because every unit of the token carries the same claim on the same custodian.

Bridge and contract risk is the second. The token exists as a contract on the host chain, and the mint and burn functions are governed by a multisignature arrangement in which the custodian and merchant institutions hold keys. The whitepaper describes the contract as governed by a multisignature contract requiring signatures from its members to add or remove members. A flaw in that contract, or a compromise of the keys that control it, is a risk to the token that has nothing to do with the security of the Bitcoin chain underneath it.

Liquidity and redemption risk is the third. The token trades on the host chain at whatever price the market sets, which can diverge from the price of bitcoin, and redemption is not instant: it runs through a merchant and a custodian and is subject to their processes and their fees. A holder who needs the underlying bitcoin quickly may find that the token is easier to sell than to redeem, and that the two routes do not produce the same outcome.

None of this makes wrapped bitcoin illegitimate. It makes it a different instrument. A reader who wants exposure to bitcoin's price inside an application that cannot hold bitcoin natively is using the token for exactly what it was designed for. A reader who wants to hold bitcoin should understand that the token is a claim, and that the claim is only as good as the institution behind it. The layer-two and sidechain page covers the other family of approaches to moving bitcoin between chains, which make different trade-offs again.