---
marp: true
title: "Taipei ZK Workshop: A Friendly Introduction to Zero Knowledge"
paginate: true
_paginate: false
---

## Taipei ZK Workshop

- Intro by ZKET team
- First session: Intro to ZK (1030-1115)

Note:

---

## A Friendly Introduction to Zero Knowledge


Note:
- Adapted from zkintro.com
- Booklets available
- General intro, then code/math

---

## Introduction (1)

- ZKPs sometimes called “magic” tech
- Let us do things that previously seemed impossible

Note: Emphasize the “magic” aspect—technology offering new trust models.

---

## Introduction (1 cont)

> “Any sufficiently advanced technology is indistinguishable from magic.” - Arthur C. Clarke

---

## Introduction (1 cont)

> “Civilization advances by extending the number of operations we can perform without thinking about them.” - Alfred North Whitehead

---

## Introduction (1 cont)

> “Privacy is the power to selectively reveal oneself to the world.” - A Cypherpunk's Manifesto

Note: Quotes highlight key ZKP themes: magic, privacy, succinctness, and a future that’s unfolding.

---

## Introduction (1 cont)
- Internet/computers touch every part of life
- We feed data (consciously or not) into countless services
- Complexity of systems → few fully understand them
- Trust becomes a big question: Who do we trust? Why?

Note:
- Frames the problem: data usage, large-scale complexity, trust dilemmas.
- MAMAA, B2B, government services, etc

---

## Introduction (1 cont)

- ZKPs reduce blind trust requirements
- Not a universal fix, but widely applicable
- We’ll discuss:
- What are ZKPs?
- Why care?
- How they work
- Applications in the real world

Note: Overview of the talk’s outline.

---

## What is a ZKP? (2)

- A way to prove a statement is true without revealing extra data
- Example: Confirm “I’m over 18” with no personal details
- Zero knowledge = share only the claim’s truth

Note:
- The core principle is minimal disclosure of hidden info.
- Bar example; prove paid correct taxes w/o details

---

## Where’s Waldo (2 cont)


Note:
- Analogy
- You know Waldo’s location but keep it hidden
- Large paper with a small hole → only Waldo visible

---

## What is a ZKP? (2 cont)

- ZKPs are general-purpose:
- Prove knowledge of a solution (e.g., puzzle, ID, financial detail)
- Preserve privacy while guaranteeing correctness

Note: Sets the stage for advanced usage. Arbitrary statements.

---

## Why Care About ZKPs? (3)

- **Privacy** (zero-knowledge):
- Only reveal relevant facts
- “Selective reveal” for digital data
- **Compression** (succinctness):
- Short proofs for big computations
- Crucial in constrained settings (e.g., blockchains)

Note: Two major benefits that power new possibilities.

---

## Privacy (3.1)

- Privacy = data belonging to a single person/group
- Physical world: private homes, keys, rooms
- Digital world: also needs privacy (often neglected)
- ZKPs = “programmable privacy,” revealing only what’s needed

Note:
- Reflect on the normalcy of privacy in physical contexts vs. the digital gap.
- Passwords; convos/cc/social graphs...

---

## Silent Reading (3.1)


Note:
- Historically, “silent reading” was novel—privacy norms can evolve similarly now.
- Used to be surprising 2000 years ago, bishop

---

## Compression (3.2)

> “I have made this longer… only because I haven’t had time to make it shorter.” – Blaise Pascal
- Compress: press into small space
- Succinctness: express clearly in few words
- Example: Blockchains have limited space

Note: Think of it as a “compressed detective conclusion.”
Size stays constant; useful outside too

---

## Sherlock Holmes (3.2)


Note:
- A metaphor for revealing “whodunit” succinctly.
- A lot of back and forth, some argument of knowledge, ~proof

---

## What Do You Know? (3.3)

> Civilization advances by extending the number of operations we can perform without thinking about them. (Alfred North Whitehead)

---

## What Do You Know? (3.3)

- We trust banks, institutions, big tech daily
- Mentally taxing - what if they fail or misuse data?
- Audits/fines aim to reduce risks, but add complexity
- ZKPs can streamline trust → do more automatically
- Social scalability

Note: Less mental overhead if proofs handle trust checks.
Esp as social complexity increases; composing ZKPs

---

## General-Purpose Computation (3.4)


Note:
- Hash function to proof; private input and public input, proof
- Like calculator to computer

---

## General-Purpose Computation (3.4)

- ZKPs shift from single-purpose (e.g., hashes) to universal programs
- Like going from a **calculator** → **computer**
- Example: prove valid ID, over-18, and a valid ticket without revealing personal info

Note: “Program any condition” and get a privacy-preserving proof.
Waldo, or Sudoku example - diff but can e express both
Group signature, Publius

---

## Why Now? (3.5)

- ZKP concept known since 1980s, only recently practical
- Blockchains (Ethereum, Zcash) accelerated R&D
- Rapid improvements: faster proof generation, robust tools
- Comparable to “Moore’s law” leaps in performance

Note: Perfect storm of research + real-world demand fueling the ZKP boom.

---

## How ZKPs Work (4.1)

- Involves a **Prover** and **Verifier**
- Prover has private input (“witness”) + public input
- Often “non-interactive”: single proof, no back-and-forth
- Protocol ensures correctness

Note: High-level structure: one proof, easy to verify.

---

## How ZKPs Work (4.1 cont)

- Protocol: rules for correct conduct
- Proof: argument establishing truth of stm
- Prover: someone who proves
- Verifier: someone who verifies correctness
- Private input: visible to prover (witness)
- Public input: visible to prover & verfier (instance)

Note:
- Some terminoloy
Protocol chess, Holmes, reasonable doubt
Protocol prover verfier, interactive, then not

---

## How ZKPs Work (4.1 cont)

- Called “zk-SNARKs”: Zero-Knowledge Succinct Non-Interactive ARgument of Knowledge
- ZKPs as a Zoo: Different systems (varying math assumptions/setups)
- Common theme: short proofs, data privacy

Note: Summarizes acronym and broad categories.
Different ones, SNARKs, some ZK system not even ZK

---

## Protocol (4.2)

- Special program: *circuit* made up of *constraints*
- (Setup: Prover and verification key)

Note:
- Special program; e.g. preimg hash; verify t/f

---

## Constraints (4.3)

- Program is broken into **constraints**
- Example: Sudoku rules (each row/column has unique digits)
- Satisfying all constraints → valid solution

Note: ZKPs prove constraints hold without revealing internal secrets.
Circuit because constraints related, electric circuit
Any problem can be expressed; but diff paradigm!

---

## Sudoku example (4.4)


Note: Another example: prove you solved Sudoku without showing your final grid.
Need all constraints valid!

---

## Proof (4.4)

<table>
<tr>
<td><img src="https://hackmd.io/_uploads/B1ndB2kayl.png" width="80%"></td>
<td><img src="https://hackmd.io/_uploads/rkuOrny6kx.png" width="80%"></td>
</tr>
</table>

Note:
- Magic spell left
- Right is proof that works; we'll go into what this is more

---

## Some properties (4.5)

- Completeness: if statement is true, verifier will be convinced
- Soundness: if statement false, ~verifier won't be convinced by proof
- Zero knowledge: if stm true, won't reveal anything else

Note: Key takeaway: the solver’s secret remains hidden, but correctness is certain.
Soundness is statistical

---

## Some properties (4.5)

- Considerations: assumptions, security
- Trusted setup
- Performance for generating/verifying proofs
- Software support, DevEx
- Etc

Note:
- Many diff types of ZK Zoo, both in theory and practice
- We'll see some here; focus on gist

---

## Real-World Applications (5)

- **Electronic Cash (Zcash, Tornado Cash, Railgun)**
- Private transactions, preserving fungibility
- Anonymous Signaling (Semaphore)
- Group membership/voting
- **ZK Rollups/zkEVM**
- Bundle txs off-chain, post a small proof
- E.g., Polygon zk-EVM, zkSync
- **Data Provenance & ZK-ID**
- Prove access to e.g. passport or email
- E.g. Rarimo, ZK-Email; also TLS Notary

Note: Widespread usage across blockchain privacy and scaling.
Last recently

---

## Real-World Applications (5 cont)

- **ZK-VM (Mina, Aleo)** - new blockchain architecture
- **Dark Forest** - “fog of war” in real-time strategy
- **ZK Bridges** - cross-chain asset transfers
- **ZK-ML** - private model/data
- **Future**: - photo authenticity, nuclear disarmament, etc.

Note: Rapidly expanding domain—beyond just blockchains.

---

## Real-World Applications (5 cont)

> "By 2040, someone will have won a Nobel Peace Prize for using Zero Knowledge Proofs."
> Barry Whitehat

Note:
- outlandish but who knows

---

## Conclusion & References (6)

- ZKPs = privacy + succinctness → new trust paradigm
- Deployed in real apps (finance, identity, gaming)
- Ecosystem evolving quickly, more breakthroughs ahead
- Original article & background:
[zkintro.com/articles/friendly-introduction-to-zero-knowledge](https://zkintro.com/articles/friendly-introduction-to-zero-knowledge)

Note: End of deck—encourage deeper study.

---

## Thanks!

- Questions?
- Logistics:
- Short break
- Then coding session