Episode Summary
Executive Summary: This episode frames Flashbots’ Suave as the next evolution of MEV research: a decentralized, privacy-preserving platform for MEV-time applications such as encrypted mempools, decentralized block building, and intent matching. Phil Daian and Andrew Miller argue that without stronger decentralization and privacy, MEV will keep centralizing crypto; with Suave, crypto can preserve fairness, user value capture, and censorship resistance—and potentially apply the same coordination tech beyond crypto.
Main Topics: The MEV problem and Flashbots’ origin story (Priority: 5/5): Phil recounts how Flash Boys 2.0 surfaced MEV as a structural threat to blockchains: transaction ordering creates discretionary power, enabling reordering, insertion, censorship, and value extraction that can concentrate power among sophisticated actors. MEV-Boost and proposer-builder separation (Priority: 5/5): The discussion explains how MEV-Boost/PBS helped contain MEV by separating block proposal from block building, letting validators capture value competitively without needing to be sophisticated builders themselves, while acknowledging trusted centralized relays as a limitation. Suave as a platform for MEV-time apps (Priority: 5/5): Suave is described as a single unifying auction for value expression and a platform for building privacy-enabled, decentralized, credible computation for applications that run during block construction, including encrypted mempools and decentralized block builders. Privacy and trusted hardware as the enabling layer (Priority: 4/5): Andrew argues that privacy is necessary not only for user protection but to avoid trust-based centralization. Suave initially relies on SGX/TEE hardware to achieve credible execution and privacy for shared private state, especially for short-lived MEV applications. Stakeholder wins: Ethereum, L2s, users, and other chains (Priority: 4/5): The guests map Suave’s benefits across the ecosystem: Ethereum gets decentralization protection, L2s get options for sequencer/MEV design, alternative L1s can better handle MEV trade-offs, and users get better execution and a chance to internalize value from their data. Beyond crypto: coordination problems everywhere (Priority: 4/5): The episode closes by extending Suave’s relevance to real-world coordination failures like ticketing, TradFi auctions, ad markets, and potentially even future AI/LLM coordination systems, positioning Suave as a general-purpose coordination primitive.
Key Arguments: MEV is not a side issue; it is a structural force that can concentrate power and undermine blockchain decentralization if left unchecked. PBS/MEV-Boost improved fairness by making block-building competitive and permissionless, but it still depends on trusted relays and does not fully eliminate centralization pressure. Suave is meant to be a platform, not just a single app: it can support encrypted mempools, decentralized builders, relay logic, and other MEV-time mechanisms. Privacy is essential because without it, users cannot protect strategy, bids, or order flow; they must rely on trusted centralized intermediaries. SGX/TEEs are imperfect but practical for Suave because MEV applications are often short-lived and need shared private state, which zero-knowledge proofs alone cannot easily provide. The value of decentralization is high enough that many users/chains will accept some latency or complexity to avoid censorship, extraction, and jurisdictional capture. Users can only efficiently capture the value of their transaction data if privacy and competitive mechanisms are present; otherwise, value leaks to intermediaries as in TradFi. Suave’s broader goal is to preserve crypto’s difference from TradFi by giving the ecosystem tools to avoid becoming a centralized finance stack. MEV problems are likely to emerge more forcefully as chains and apps gain adoption; ignoring them early is a common but dangerous pattern. Suave could also act as a general coordination layer for intents, auctions, and other multi-party matching systems beyond Ethereum. Data Points: MEV-Boost validator adoption: 90% to 95% of validators - Phil says MEV-Boost has become the dominant validator workflow for Ethereum block construction. Flashbots thesis timeline: 4 to 5 years ago - Phil situates the origin of his MEV research and Flash Boys 2.0 around this period. Ethereum block time for MEV-time apps: ~15 seconds - Phil describes Suave’s initial target as apps that run in the time window while a block is being created on Ethereum. Celo active address growth: over 500% in the last 6 months - Mentioned in an ad read to illustrate Celo ecosystem growth. GMX incentives: $12 million in Arbitrum grants - Mentioned in an ad read for GMX V2 and ecosystem incentives. GMX trading volume: over $150 billion - Mentioned in an ad read to emphasize GMX’s on-chain volume.
Pivotal Quotes: "MEV is basically about what is the power distribution in your system, how is information propagated, and who profits off that." — Phil Daian: Phil defines the core conceptual issue behind MEV and why it matters for decentralization. "Suave is the single unifying auction for value expression." — Phil Daian: Phil’s compact definition of Suave as a platform and coordination layer. "If you want to be able to access shared private state, then there's not really a better alternative." — Andrew Miller: Andrew explains why SGX/TEEs are necessary for certain Suave applications that require private shared computation.
Implications: If Suave works, crypto can reduce MEV-driven centralization, improve user execution, and give chains/L2s better tools for decentralized coordination. More broadly, it could become a reusable primitive for auctions, intents, and other trust-heavy markets.