Smart Contracts

The name is misleading

A smart contract is neither smart nor a contract.

It is not smart: there is no judgement or intelligence, only code that runs the same way every time. It is not a contract: no court recognises it as one, and it cannot be renegotiated when circumstances change.

What it is: a program stored on a blockchain, which anyone can call, whose result every participant can independently verify.

What Ethereum added

Bitcoin's scripting is deliberately limited. Ethereum, launched in 2015, made the ledger programmable in a general way, so instead of only recording "A paid B", a blockchain could record "when X happens, move funds to Y".

That opened token issuance, lending protocols, exchanges without an operator, and much of what followed.

Where it works

The strong case is agreements where the terms are unambiguous, the inputs are already on-chain, and neither side should be able to change their mind: escrowing funds until a condition is met, distributing payments among recipients by fixed shares, governing which addresses may act.

Where it breaks

Code is the agreement. A bug is not a mistake to be corrected; it is the terms, and it executes. The 2016 DAO incident drained a large fund by using the contract exactly as written, and Ethereum's community split over whether to reverse it.

Outside facts need an oracle. A contract cannot know a flight was delayed or a price moved. Something must tell it, and that something is a trusted party reintroduced at the edge.

Immutability cuts both ways. A deployed contract usually cannot be patched. Upgrade patterns exist, but they hand someone the power to change the rules, which is the thing the design was avoiding.

The honest summary

Smart contracts remove the need to trust a counterparty to follow the terms. They do not remove the need to trust whoever wrote the code, whoever audited it, and whoever supplies its outside information.