The expectation of shared acceptance
We accept a banknote mainly because we expect to spend it again. That expectation creates a network effect: the more people use the same unit, the more useful it becomes. Law, taxes and public payments can reinforce acceptance, while day-to-day usefulness also depends on prices, merchants and payment infrastructure.
Trust is not blind belief in a fixed price. People may use a currency with modest inflation if it is practical and sufficiently predictable in the short run. During hyperinflation or withdrawal restrictions, coordination may shift to a foreign currency, goods or another record of value.
Institutional layers
With cash, we rely on the issuer, anti-counterfeiting features and the legal environment. A bank deposit also relies on the bank's solvency, payment systems, supervision and deposit-protection rules. A card adds the issuer, acquirer, network and technical processor. A simple-looking payment rests on several linked promises.
These layers deliver services users value: reversal of some payments, access recovery, consumer protection and credit. They also create control and failure points. An account can be frozen by mistake, an institution can fail, and rules can change.
What Bitcoin changes
Bitcoin lets a user verify rules and transactions with their own node rather than relying on one central ledger. A private-key holder can authorise a transfer without a bank's permission. Consensus and proof of work make confirmed history costly to alter.
Trustless is therefore better understood as minimising trust in a particular intermediary, not living without trust. Users rely on their understanding or competent review of software, uncompromised hardware and a safe backup. When using an exchange, they again trust an operator.
- verifying rules is not the same as understanding every line of code
- self-custody removes custodian risk but adds the risk of personal error
- decentralisation reduces some control points, not every human and technical risk
Trust, control and recourse
A centralised system can reverse some fraud and restore forgotten access because an authority can amend the record. Bitcoin deliberately favours payment finality and key control. That helps resist arbitrary censorship, but a transfer to the wrong address generally has no central complaints desk.
The useful question is not whether a system requires trust, but whom or what we trust for which operation, what we can verify, what failure would mean and whether recourse exists. Different users can reasonably choose different balances.