Skip to content

Sybil resistance

The problem

If one person can cheaply appear as many, then “one member = one vote” (and quadratic mechanisms) collapse.

Sybil resistance is the craft of making membership/voice costly to fake at scale without demanding invasive identity.

The goal

A good system should be:

  • Inclusive: doesn’t require government ID by default
  • Private: proves enough (uniqueness / eligibility), not everything
  • Composable: different co-ops/DAOs can choose different gates
  • Economically robust: attacks become uneconomic, not “impossible”

Practical toolbox (mix-and-match)

  • Admission rules: sponsor/vouching, stake/bond, rate limits, probation periods
  • Proofs of participation: attestations from real activity (work, tenure, contributions)
  • Composite scores: multiple weak signals combined beat any single oracle
  • ZK / selective disclosure: prove claims without doxxing

Why it matters for Luminism

Luminism leans on co-ops and participatory governance.

Without Sybil resistance: - voting becomes pay-to-attack - subsidy programs get farmed - “fair pricing” turns into “many fake accounts”

With it: - quadratic voting/funding becomes usable - co-op membership stays credible - baseline access programs can be enforced without surveillance

Design note

Treat Sybil resistance like security engineering: layered defenses, clear threat models, and simple failure modes.

(See also: Co-op Primitives → Membership & Accountability for the “membership + dues/bond” framing.)