Library · Crypto, word by word · Ideas and debates
What is trust minimization?
Trust minimization is the design goal of reducing how much a system depends on any single person or company behaving honestly. Instead of relying on a bank or middleman, the rules are checked by software, cryptography and many independent computers, so no one party has to be believed.
Where people disagree
Supporters say every intermediary is a point where money can be frozen, lost or mismanaged, and that replacing institutional promises with publicly checkable rules and cryptography removes costs and risks rather than relocating them. The Bitcoin whitepaper argues that relying on financial institutions brings mediation costs, reversible payments and accepted fraud.
Critics say trust does not disappear but moves to developers, large token holders and validators. BIS economists argue full decentralisation is an illusion because every platform keeps governance that someone controls, and many projects give a large share of their coins to insiders, so users end up relying on a small group anyway.
What it means for you
The more trust-minimized a setup is, the fewer parties can freeze, lose or misreport your funds, but more of the checking, and the mistakes, fall on you. A coin can run on a trust-minimized network and still be held through a company you depend on completely; those are separate questions.
A common mistake: “A trustless system means I don't have to trust anyone.”
In fact: You still rely on the software being correct, on most of the network's computing power being honest, and on whoever writes and maintains the code. Trust is reduced and spread out, not eliminated.
How it works
Nick Szabo's 2005 bit gold proposal set the goal plainly: create valuable bits online 'with minimal dependence on' third parties that must be relied on. The Bitcoin whitepaper applied the idea to payments, proposing 'cryptographic proof instead of trust': transactions are announced publicly and their order is fixed by proof-of-work, so the system holds as long as honest nodes control more computing power than any cooperating group of attackers. Trust is reduced, not removed; it shifts to that majority assumption and to the software.
Sources: Nick Szabo, Bit gold (2005), Bitcoin whitepaper (Satoshi Nakamoto), BIS Quarterly Review: DeFi risks and the decentralisation illusion (Aramonte, Huang, Schrimpf, 2021) · checked 4 October 2026
On Cryptominium
Related words
Educational content, not financial advice. Written by hand and checked against the source named above. Something wrong? Tell us and we reply within two business days.