Bitcoin's proof-of-work (PoW)-based protocol is remarkable for how little it asks of its participants. Not only can miners take breaks from work whenever they please, but it is almost unique in offering a path of contrition: corrupt miners can reclaim honest status simply by resuming mining on the longest chain. The protocol only requires that honest miners hold the majority of computational power
Fully Fluctuating Sleepy Consensus from Minimal Assumptions is a recent cryptographic protocol by Javier Nieto, Yuval Efron, Joachim Neu, and Ling Ren (September 2026) that enables Proof-of-Stake (PoS) systems to match Bitcoin’s robustness against arbitrary participation changes without relying on hardware assumptions like Verifiable Delay Functions (VDFs).
The protocol achieves this through two key innovations: External Adversary Model: It assumes corrupt nodes do not share their secret keys, preventing "pre-signing" attacks that typically undermine consensus when participation fluctuates. Graded Wakeness: A novel primitive allowing nodes to form consistent opinions on which peers are currently active, enabling secure agreement even when honest and adversarial participation levels shift unpredictably.
This approach provides a path to uncorruption (where corrupt nodes return to honesty) using only a Public Key Infrastructure (PKI) and Verifiable Random Functions (VRFs), removing the need for the sequential work assumptions inherent in previous PoS designs.
The paper formalizes Bitcoin-style proof-of-work consensus in a fully fluctuating sleepy model, where miners may sleep, wake, and change participation without any global schedule, and where the computational power available to honest and adversarial parties can vary over time. Instead of assuming a fixed set of always-on nodes, synchronized rounds, or stable hash rates, it isolates the minimal structural assumptions behind the longest-chain rule: blocks are only useful if they are valid PoW, nodes follow the longest-chain policy, and honest miners control a majority of work or computational power. In this setting, a miner’s identity or past behavior is not a persistent state variable; a previously malicious miner can become harmless—or even beneficial—by simply adding valid work to the current longest chain.
The central contribution is a rigorous account of why this “path of contrition” is enough to preserve consensus. The analysis shows that transient disruptions, churn, sleeping miners, and even temporary adversarial advantages do not create lasting forks or irreversibly commit the system to a bad history, as long as honest work ultimately dominates and the cost of rewriting history remains prohibitive. Safety follows from the fact that an adversary must outpace honest PoW to make a conflicting chain competitive, while liveness is conditional on honest miners being active: when they are, the ledger progresses, and when they are not, the protocol can pause without sacrificing finality.
This matters because it gives a clean, minimal-assumptions foundation for one of the most deployed consensus mechanisms. It clarifies what Bitcoin actually relies on—not trust in participants, stable infrastructure, or synchronous operation, but verifiable work, longest-chain selection, and an honest majority of computational power. For researchers and system designers, the result explains the robustness of PoW in realistic environments with unstable miner populations and adversarial behavior, and it delineates the boundary between assumptions that are essential to Bitcoin’s security and those that are merely convenient modeling artifacts.