Why would the Ethereum Foundation suddenly slam the brakes on the “move fast and break things” approach that has driven crypto innovation for years? The answer lies in their ambitious three-stage roadmap targeting 128-bit provable security for zkEVMs by the end of 2026.
The Ethereum Foundation’s shift from “move fast and break things” to 128-bit security represents crypto’s maturation from reckless innovation to institutional-grade reliability.
This security standard represents a dramatic shift from performance to cryptographic resilience. Think of it like moving from building fast race cars to constructing bulletproof tanks. While current Layer-2 solutions often operate below 100-bit security, this new minimum standard matches what mainstream cryptography considers safe for long-lived systems.
The roadmap unfolds in three careful phases. Phase 1 introduces soundcalc, a new tool that measures concrete security levels for proof systems. By February 2026, zkEVM teams must integrate this tool to get unified security assessments. It’s like giving everyone the same ruler to measure their progress.
Phase 2 raises the bar to at least 100-bit provable security while keeping proof sizes below 600 kilobytes by May 2026. Teams also need to describe their recursion architectures clearly.
Phase 3 delivers the full 128-bit security goal with proofs capped at 300 kilobytes by December 2026, complete with formal soundness arguments. Formal verification will serve as a cornerstone technology throughout this implementation process.
This shift comes at an interesting time. Proving times have already dropped from 16 minutes to 16 seconds for Ethereum blocks, while costs fell 45-fold. Now 99% of mainnet blocks get proved in under 10 seconds. The infrastructure is finally ready for security-first thinking.
The implications stretch far beyond technical specifications. Institutional confidence could skyrocket with genuine security guarantees, giving Ethereum a competitive advantage. The foundation envisions scaling to 1 tera-gas per second with real-time proving while maintaining rock-solid security. Recent advances in cryptographic research have made the ambitious 128-bit security target both technically feasible and economically viable.
However, zkEVM teams face a clear choice: comply with these requirements or risk becoming irrelevant for mainnet deployment. The foundation warns against high-speed proving without verified soundness. Unlike dividend cuts during financial troubles, this forced evolution strengthens the entire ecosystem’s foundation.
This marks the end of the “move fast and break things” era and the beginning of what they call the foundational robustness era.








