PETs Series - Zero-Knowledge Proof Explained

In the last episode, Chris explored the opportunities and challenges associated with Secure Multi-Party Computation. This time, Chris delves into the world of Zero-Knowledge Proof.

Episode 4 - Zero-Knowledge Proof Explained

Zero-Knowledge Proof (ZKP), a wider concept of which SMPC is an application thereof, is
difficult to describe simply, evidenced by the popular “How to explain Zero-Knowledge Proof
to your children” Ali Baba example being six pages long and incredibly complicated11.

A one-line description would be: “a Prover wants to convince a Verifier that a statement is
true without revealing any further information.”

SMPC: Explained

I have found that the most effective way to describe Zero-Knowledge Proof in non-mathematical terms is the “two balls and a colour-blind friend description”, which I have summarised below:

• My friend (let’s call him Russell) is red-green colour blind. I have a red ball and a green ball, which aside from the colours, are otherwise identical. These balls look identical to Russell.
  
  
• I wish to convince Russell that the balls are different colours.
  
  
• I give the balls to Russell who puts them behind his back. He pulls one ball from behind his back and holds it in front of him, so I can see it. He puts it behind his back. He then does this again and asks me whether he has switched the ball.
  
  
• This procedure is repeated as often as required.
  
  
• By looking at the colours, I can say with certainty whether Russell has switched the ball, and after enough times, the probability of having guessed the correct colour each time becomes inconceivably small.
  
  
• Russell now knows the balls are different colours (proof) but does not know which ball is which colour (zero knowledge).12

Zero-Knowledge Proof can be seen as an extension to Secure Multi-Party Computation, as SMPC needs to be in place in order to ask ZKP questions.

A practical use case is to ask a number of financial institutions as to whether a new customer to a particular institution has been listed as a PEP at any other institution, without disclosing the customer’s personal details or which other institutions the customer banks with.

Next episode:

Episode 5 - Homomorphic Encryption Explained
Tuesday 26th July 2022

Questions?

If you have any questions about how traditional analytical and data matching techniques can help you prevent fraud and understand your customers, click here to contact us.

In the meantime, please feel free to read more articles and thought pieces in the 'Our Thinking' section of our website.

More of 'Our Thinking'

10 – https://research.aimultiple.com/secure-multi-party-computation/