FHE Rollups and Co-Processor in Action
summary
Fhenix claims Fully Homomorphic Encryption makes confidential smart contracts practical by letting sequencers execute encrypted inputs on an L2 rollup and publish an encrypted state; decryption uses a threshold network that splits keys across participants and supports post‑execution fraud proofs. The talk contrasts FHE with ZK (proofs not general computation), MPC (fast but chatty), and TEEs (trust and side‑channel risk), notes GPU/FPGA/ASIC optimizations arriving in early 2025, and points to a live Fhenix testnet and upcoming FHE co‑processor integrations.
AI-written summary · about
chapters (23)
- 0:00Introduction: Guy Itzhaki from Fhenix on FHE
- 0:52Understanding Blockchain & Lack of Privacy
- 1:26What Privacy Means: Anonymity vs. Confidentiality
- 2:01Confidential Smart Contracts Explained
- 3:12Use Case Example: Sealed-Bid Auctions
- 4:25Limitations of ZK Proofs for Confidentiality
- 4:59MPC: Benefits, Limitations, and Use Cases
- 5:34Issues with Trusted Execution Environments (TEEs)
- 6:07Introducing Fully Homomorphic Encryption (FHE)
- 6:42How FHE Enables Computation on Encrypted Data
- 7:21Overcoming Complexity: Easy Development with FHE
- 7:56Improving FHE Performance & Hardware Optimization
- 8:33What is an FHE Rollup & How Does It Work?
- 9:48Implementing FHE Co-Processors for L1, L2 & L3 Chains
- 11:07Step-by-Step: How FHE Co-Processors Function
- 11:42Real-World Blockchain Applications Using FHE
- 12:57Decentralized Identity, Credit Scoring & Lending
- 13:33Confidential AI: Intersection of AI & Blockchain
- 14:08How AI Can Improve Blockchain Applications
- 14:43Decentralized AI Models & Data Privacy with FHE
- 15:17Future of AI: Agent-to-Agent Transactions
- 15:47Ensuring Data Authenticity & Privacy with Blockchain
- 16:24Conclusion & Resources to Learn More about FHE
related talks
40mConfidential Computing and Privacy-Enhancing Technologies - The Landscape
56mBuilding a compiler and processor for fast processing of encrypted data
44mIntroduction to confidential computing – protecting sensitive workloads
44mIP Protection and Privacy in LLM: Leveraging Fully Homomorphic Encryption
26mBuilding Private And Secure Applications Easily Using The Sunscreen FHE Compiler&Rust
36mEverybody is talking about Confidential Computing - the minimum Devs should know
40mTrust No One
54mFully Homomorphic Encryption