Sybil Attack

Why It Matters

“One node, one vote” sounds democratic until you realize a node is just a process — and one attacker can run a million of them. The Sybil attack (named after a famous case study in multiple personalities) is why naive voting can’t secure an open network, and it’s the problem that stalled digital cash for decades.

It’s also the cleanest lens for understanding consensus mechanisms: every one of them is, at bottom, an answer to “what do we weight votes by, if not identities?” Get that, and the differences between PoW, PoS, and their variants snap into focus.

How It Works

Beginner

Imagine a town vote where registering to vote is free, unlimited, and anonymous. One bad actor prints ten thousand voter cards and outvotes the whole town. That’s a Sybil attack. The fix isn’t checking IDs — there are no IDs on the internet — it’s making each vote cost something real, so a flood of fake voters becomes a flood of real expenses.

Intermediate

The casebook frames it as an old, known problem that Bitcoin had to solve to do consensus without known participants. Classic BFT protocols assume a fixed membership list precisely to dodge this; an open network can’t.

Satoshi’s answer: make identity costly. In Proof of Work, your “vote” on the next block is hashing power — spinning up a thousand fake nodes gains you nothing, because nodes don’t vote, work does. Stake-based systems take the other route: influence is tied to money held (Proof of Stake), so fake identities just split your stake without multiplying it. Either way, the resource — joules or capital — is what’s scarce, not names.

Builder

The general principle: Sybil resistance requires binding influence to an unforgeable, costly resource — computation (PoW), bonded capital (PoS), storage (proof-of-space), or even verified personhood (proof-of-identity experiments). Note what Sybil resistance does not buy: it caps influence-per-dollar, but says nothing about how many dollars one party has. Concentration of the underlying resource re-opens the door — that’s the 51% attack, the Sybil attack’s better-funded sibling. Peer-to-peer layers face their own variant: eclipse attacks Sybil a single node’s connections rather than the global vote.

Examples

Bitcoin — PoW as costly identity; the canonical solution.
PoS networks — Stake-weighting as the alternative answer.
Outside consensus — Fake reviews, bot followers, airdrop farming: the same attack against any free-identity voting system.

Tradeoffs

What costly-identity defenses buy

Open membership — anyone can join, yet votes can’t be forged out of nowhere.
Quantifiable security — attack cost is measurable in hashrate or stake, not in unverifiable headcounts.

What they don’t

Plutocracy by design — weighting by resources means the wealthy weigh more; “one person, one vote” is abandoned, not achieved.
No protection against concentration — a majority owner of the scarce resource defeats the scheme (see 51% Attack).
Cost is borne constantly — the network pays for Sybil resistance every block (energy or locked capital), not just under attack.

Proof of Work — Costly identity via computation
Proof of Stake — Influence weighted by capital
Distributed Consensus — The process being defended
51% Attack — What Sybil resistance can’t prevent

Sources & Last Updated

MIT BLC Module 2: Maintaining Blockchain Integrity (primary source)
Douceur, J. “The Sybil Attack” (2002)
Vault note: Sybil Attack (M2 cluster)

Last updated: June 10, 2026