Can Private Proof-of-Stake Blockchains Obtain All of Safety, Liveness, and Privacy?
summary
Shows that private PoS cannot simultaneously guarantee safety, liveness, and privacy when an adversary can delay messages: network-level delays let an attacker manipulate mempools and perform reverse-tagging attack and other stake inference attacks (improved to near-true stake via binary-search style sampling), with a formal lower bound for deterministic protocols. Implements a defense using differential-privacy noisy stake mechanisms, evaluated on Ethereum parameters (trade-off curves for Alpha/Epsilon and adversary stakes 20/25/30%), which reduce leakage but lower the tolerated adversary stake.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 1:10Proof of Work
- 4:07Zero Knowledge Proof
- 7:59Key Attack Idea
- 10:41Can it just ignore transactions
- 12:23Reverse tagging attack
- 15:57Reverse tagging drawbacks
- 16:45New attacks
- 18:21A potential approach
- 19:46A noisy stake
- 21:46A distorted stake
- 23:26Tradeoffs
- 26:40Conclusion
related talks
30mBringing Privacy to Public Blockchain
58mBlockchains and Publicly-Verifiable Lotteries
32mAnalysis of Nakamoto Consensus
01h 24mPseudonymity, Automation, and Responsibility on Public, Permissionless Blockchains under the GDPR
13mSurvey of Privacy Solutions for Minimizing MEV
56mCollaborative zk-SNARKs: Proving as One over Distributed Secrets
12mThe Limits of Technology in Achieving Privacy
21m"SECRETS ARE LIES, SHARING IS CARING, PRIVACY IS THEFT." - A Dive into Privacy Preserving