Taiga: A dark forest of zero-knowledge programmability
summary
Taiga is an implemented architecture for private counterparty discovery that claims a fundamental privacy boundary: the who (identities) can be hidden but the what (assets, amounts, constraints) must be revealed to enable matching. It uses validity predicates (user and token circuits), a peer-to-peer intent gossip network, off‑chain partial composition for atomic multi‑party swaps, and a customized Plonk‑esque proof system with a Halo2 backend. A GitHub spec and early prototype exist (proof generation is still slow but functional).
AI-written summary · about
chapters (12)
- 0:00Introduction to Taiga
- 1:14Counterparty discovery
- 3:56Current discovery methods
- 6:21The intent-based model
- 9:46Privacy trade-offs
- 13:41Partial settlement mechanics
- 17:09Architectural design
- 20:12Validity predicates
- 24:29Use cases and examples
- 29:44Research directions
- 33:30Compiler stacks
- 39:28Conclusion and Q&A
related talks
26mHow zero-knowledge proofs can be used to balance privacy and compliance
48mSoK: Programmable Privacy
30mBringing Privacy to Public Blockchain
01h 01mKey transparency: Blockchain meets NoiseSocket
33mAn Introduction to Zero Knowledge Proofs
17mDecentralization & Privacy: Amir Taaki on Monero's Technological Advancements
19mBeyond the Blacklist, The Compliant Way To Do Private Transactions
01h 00mThe Path to Shielded Programmability