Question
What is a zero-knowledge proof (ZKP)?
Answer
A method by which one party (the prover) convinces another (the verifier) that a statement is true — without revealing anything beyond the fact that it is true.
* The classic "cave" illustration of a zero-knowledge protocol. — RokerHRO, Public domain, via Wikimedia Commons. *
The magic is in what is not shared. A ZKP lets you prove you know a secret, or that a fact holds, while leaking zero additional information — not the secret itself, not how you know it.
Classic framing — two things happen at once:
- You prove a statement is true (the verifier ends up convinced).
- The verifier learns only that it's true — nothing about the secret behind it.
(The formal version of these guarantees — completeness, soundness, and zero-knowledge — is covered in the next card.)
Everyday analogies:
- Proving you're over 18 without showing your birth date.
- Proving you know a password without sending the password.
- Proving you can solve a maze without revealing the path.
Tip: Remember it as "I can prove I know it, without telling you what it is."
Go deeper:
Zero-knowledge proof (Wikipedia) — the prover/verifier framing and worked protocol examples.
Note saved — thanks!
Question
How do the "two pens" and "red/green card" intuitions illustrate zero-knowledge proofs?
Answer
Both let a prover convince a verifier they can tell two things apart (or know a secret) — through repeated challenges (the pens) or by elimination (the cards) — while the verifier learns nothing about the distinction itself.
* One commit-challenge-response round; repetition convinces the verifier without revealing the secret. *
In the two-pens intuition, a colour-blind verifier holds two pens that look identical to them but are different colours. The prover (who can see colour) is repeatedly asked "did I swap them?" and always answers correctly. After many rounds the verifier is convinced the pens really differ — but never learns which is which.
The red card intuition makes the same point by elimination: the prover holds one secret card from a 52-card deck and, to prove it is red without revealing its rank or suit, reveals all 26 black cards from the rest of the deck. If every black card is accounted for, the hidden card must be red — yet the verifier still learns nothing about which red card it is.
Why this matters: A single round could be a lucky guess. The interaction repeated many times is what drives the cheating probability toward zero — this is the heart of soundness.
Note saved — thanks!