What Is Ethereum and Smart Contracts?

Bitcoin vs Ethereum — the key difference

Bitcoin was designed with deliberate constraints. Its scripting language is intentionally limited — it can verify conditions for spending coins (multisig requirements, timelocks, hash preimages) but cannot run general-purpose programs. This simplicity is a feature: a more complex language would introduce more attack surface in a system securing hundreds of billions of dollars. Bitcoin's design philosophy is conservative — change slowly, break nothing.

Vitalik Buterin, a 19-year-old Canadian-Russian programmer, proposed a different vision in his 2013 Ethereum whitepaper. His core argument was that a blockchain did not need to be only a ledger — it could be a global computer. Rather than building individual blockchains for each specific use case (a prediction market, a decentralised exchange, a token issuance platform), you could build one programmable blockchain on which any application could be deployed. Ethereum launched in July 2015 with exactly this capability: the Ethereum Virtual Machine (EVM), a Turing-complete computing environment embedded in the blockchain consensus mechanism.

The distinction matters practically. Bitcoin can tell you who owns which BTC and enforce rules for transferring it. Ethereum can do this and additionally run any program whose logic can be expressed in code — automatically, on every node in the network simultaneously, without any company or person controlling the execution. This distinction is why the entire DeFi ecosystem, the NFT market, and most of what people mean when they say “Web3” is built on Ethereum or compatible chains — not on Bitcoin.

What smart contracts actually are

A smart contract is a program stored on the Ethereum blockchain. Like all programs, it is a set of instructions: “if condition X is true, execute action Y.” What makes it “smart” is not artificial intelligence — it is that the contract self-executes automatically and deterministically when its conditions are met, with no human operator required and no possibility of the outcome differing based on who is calling it.

Consider a simple example: an escrow smart contract. Alice wants to hire Bob to build a website. Normally, Alice either pays upfront (risking Bob disappearing) or pays on completion (risking Alice refusing to pay). A smart contract encodes the agreement: Alice deposits funds into the contract; Bob submits a deliverable; a predefined condition (perhaps a third-party arbiter's cryptographic sign-off) triggers automatic payment to Bob. Neither party can manipulate the outcome once the contract is deployed — the code is the law. This eliminates the trust problem without requiring a lawyer, an escrow company, or any intermediary.

More complex smart contracts manage billions of dollars. Uniswap, a decentralised exchange running on Ethereum, holds over $3 billion in liquidity across thousands of token pairs. Every trade is executed automatically by the smart contract code, with prices set algorithmically based on token ratios in liquidity pools. No company holds custody of the funds, no order book is maintained by any server, no human executes any trade. The entire system operates as code running on the Ethereum network — available 24/7 to anyone with an internet connection and an Ethereum wallet.

Gas fees — paying for computation

The Ethereum Virtual Machine is a shared global computer. Every node in the network must execute every smart contract call — which means computational resources are scarce and must be rationed. Ethereum's mechanism for this rationing is gas.

Gas is a unit measuring the computational work required to execute a specific operation in the EVM. A simple ETH transfer costs 21,000 gas. A complex DeFi transaction involving multiple contract calls might cost 200,000–500,000 gas. Users set a gas price (in gwei — billionths of an ETH) they are willing to pay per unit of gas; validators prioritise higher-paying transactions. The total fee paid is (gas used) × (gas price). During periods of high network demand — the 2021 DeFi summer, NFT minting frenzies — gas prices spiked to hundreds of gwei, making simple transfers cost $50–200 and complex DeFi operations cost $500+.

EIP-1559 (August 2021) reformed the gas fee mechanism. It introduced a base fee determined algorithmically by network congestion that is burned (destroyed) rather than paid to validators. Users add a priority tip for validators. This mechanism made fees more predictable, reduced variance, and introduced a deflationary mechanism: when network activity is high enough, more ETH is burned per block than is issued to validators — making ETH deflationary. Between the Merge and early 2024, Ethereum burned over $10 billion worth of ETH through this mechanism.

Decentralised Finance — financial services without banks

Decentralised Finance (DeFi) is the umbrella term for financial applications built on smart contracts. The core proposition is simple: every financial service that banks and brokerages provide — lending, borrowing, trading, earning yield, insurance — can be replicated through smart contracts that execute automatically without human intermediaries. Access requires only an internet connection and a wallet; there is no credit check, no KYC requirement, no business hours, and no geography restriction.

The numbers from the 2021 DeFi peak illustrate the scale of adoption the thesis attracted. Total Value Locked (TVL) — the combined value of assets deposited in DeFi protocols — reached $180 billion across all chains. Uniswap processed $1+ billion in daily trading volume without a single employee executing a trade. Aave and Compound enabled billions in crypto-collateralised loans where interest rates adjusted algorithmically second by second. MakerDAO maintained DAI, an algorithmically stabilised stablecoin worth $1, backed by crypto collateral managed entirely by smart contract logic.

DeFi demonstrated that the smart contract model could achieve institutional financial scale. It also demonstrated its vulnerabilities: over $3 billion was stolen from DeFi protocols in 2022 alone, primarily through smart contract exploits, flash loan attacks (borrowing enormous sums within a single transaction to manipulate prices), and oracle manipulation (corrupting the price feeds that contracts rely on for accurate market data). DeFi remains a high-risk environment where protocol exploits can result in total loss of deposited funds with no recourse.

Lending, borrowing, and liquidity provision

Aave is the largest decentralised lending protocol. Users deposit assets (ETH, USDC, WBTC) into smart contract pools and earn yield from borrowers who pay interest. Borrowing requires over-collateralisation — you must deposit more value than you borrow, preventing default. If your collateral value falls below the liquidation threshold, an automated process sells your collateral to repay the loan. No human decision is involved in any step: deposit, borrow, interest accrual, and liquidation are all automated by contract logic.

NFTs — digital ownership on the blockchain

Non-Fungible Tokens (NFTs) use Ethereum's ERC-721 token standard to represent ownership of unique digital items. Unlike ERC-20 tokens (where each unit is identical — one USDC equals any other USDC), each NFT has a unique token ID that distinguishes it from all others in the same collection. Ownership is recorded on the blockchain: transferring an NFT requires a transaction signed by the current owner's private key.

The 2021 NFT boom brought the technology mainstream. Beeple sold an NFT for $69 million at Christie's. CryptoPunks — 10,000 unique pixel art characters minted in 2017 — traded for millions each. OpenSea, a marketplace for NFT trading built on Ethereum smart contracts, processed $3.4 billion in volume in a single month in January 2022. The crash that followed was equally dramatic: 95%+ of NFT collections lost almost all their value, exposing how much of the market was speculation rather than genuine utility.

The legitimate use cases for NFTs — digital art with royalties encoded in the smart contract, event tickets that cannot be counterfeited, gaming items with verifiable scarcity, membership credentials — remain valid regardless of the speculative cycle. The technology works; the valuation during the boom did not.

The Merge — September 2022

Ethereum launched in 2015 using Proof of Work, the same consensus mechanism as Bitcoin. Critics argued — correctly — that Ethereum's energy consumption was inconsistent with its stated mission of building open, equitable financial infrastructure. Planning for a transition to Proof of Stake began in 2016; execution took six years of careful engineering. “The Merge” occurred on 15 September 2022, when Ethereum successfully transitioned its consensus mechanism from Proof of Work to Proof of Stake without a service interruption — one of the most technically challenging upgrades in blockchain history.

Under Proof of Stake, validators replace miners. To become a validator, you stake a minimum of 32 ETH (~$100,000 at 2024 prices) as collateral. The protocol randomly selects validators to propose new blocks and attest to existing ones, weighted by stake size. Validators who behave dishonestly — proposing conflicting blocks or signing fraudulent transactions — have a portion of their stake slashed. The economic incentive structure is equivalent to Proof of Work: honest behaviour earns rewards; dishonest behaviour costs money. But instead of expending electricity, validators lock capital.

The result of the Merge was immediate and dramatic. Ethereum's energy consumption fell from approximately 80 TWh/year (comparable to Chile's national grid) to under 0.01 TWh/year — a reduction of 99.95%. ETH issuance to validators also fell sharply, and combined with EIP-1559's fee burning, Ethereum's net ETH supply has been deflationary during periods of high network activity — meaning the total supply of ETH decreases, rather than increases, when the network is busy.

Layer 2 — solving the scalability problem

Ethereum's base layer processes approximately 15 transactions per second — a severe constraint for any globally adopted financial system. The Ethereum community's solution is a “rollup-centric roadmap”: Layer 2 networks that process transactions off-chain in high volumes and post compressed proofs to Ethereum mainnet for final settlement. This allows Ethereum to serve as a trust anchor and settlement layer while delegating high-volume execution to specialised networks.

Two types of rollups dominate the landscape. Optimistic rollups (Arbitrum One, Optimism, Base) assume all transactions are valid and post them to Ethereum with a challenge period: during this window, anyone can submit a fraud proof if they detect invalid transactions. ZK rollups (zkSync Era, Starknet, Polygon zkEVM) post cryptographic validity proofs with each batch, mathematically proving the correctness of all included transactions without requiring a challenge period. Arbitrum processes roughly 10–40 TPS; ZK rollups can theoretically handle thousands.

By 2024, Layer 2 networks collectively processed more transactions than Ethereum mainnet. The user experience — sub-cent fees, near-instant confirmations — brought DeFi applications within reach of everyday users who had been priced out by mainnet gas fees. Coinbase's Base (an Optimism-based L2) processed over 1 million transactions per day within months of launch. The scaling thesis is playing out — the question is whether it arrives fast enough to prevent alternative Layer 1 blockchains (Solana, Avalanche, Sui) from capturing the high-throughput application market.