Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Inside Qubic's Selfish Mining Campaign on Monero: Evidence, Tactics, and Limits

Suhyeon Lee, Hyeongyeong Kim

TL;DR

The paper empirically investigates Qubic's publicly advertised selfish mining campaign on Monero by fusing data from a Monero pruning node and the Qubic pool API to attribute blocks, estimate effective hashrate, and assess profitability. It defines ten candidate selfish-mining periods, introduces a modified revenue model to reflect conservative release behavior, and finds that Qubic never achieved a sustained majority and often did not profit from the strategy, though it caused significant chain reorganizations and higher orphan rates. The work highlights real-world frictions—timing, granularity, and network variance—that diminish selfish mining profitability while underscoring the substantial network harm such attacks can cause. It also discusses mitigations, including chain-selection rule adjustments and detective mining, and emphasizes the need for decentralization and robust defenses in PoW systems like Monero.

Abstract

We analyze Qubic's advertised selfish mining campaign on Monero in 2025. Combining data from Monero nodes, and the Qubic pool API, we reconstruct Qubic-attributed blocks and hashrate and detect ten intervals consistent with selfish mining strategies. In these intervals, Qubic's average hashrate share rises to the 23-34\% range, yet sustained 51\% control is never observed. We evaluate the campaign against the classical selfish mining model and a modified Markov-chain model that reflects Qubic's conservative release strategy: both predict lower revenue than honest mining at the inferred parameters, and the data largely confirms this while still showing noticeable deviations from the predicted curve. We interpret this gap between model and measurements in terms of Qubic's time-varying hashrate and coarse-grained attack segmentation.

Inside Qubic's Selfish Mining Campaign on Monero: Evidence, Tactics, and Limits

TL;DR

The paper empirically investigates Qubic's publicly advertised selfish mining campaign on Monero by fusing data from a Monero pruning node and the Qubic pool API to attribute blocks, estimate effective hashrate, and assess profitability. It defines ten candidate selfish-mining periods, introduces a modified revenue model to reflect conservative release behavior, and finds that Qubic never achieved a sustained majority and often did not profit from the strategy, though it caused significant chain reorganizations and higher orphan rates. The work highlights real-world frictions—timing, granularity, and network variance—that diminish selfish mining profitability while underscoring the substantial network harm such attacks can cause. It also discusses mitigations, including chain-selection rule adjustments and detective mining, and emphasizes the need for decentralization and robust defenses in PoW systems like Monero.

Abstract

We analyze Qubic's advertised selfish mining campaign on Monero in 2025. Combining data from Monero nodes, and the Qubic pool API, we reconstruct Qubic-attributed blocks and hashrate and detect ten intervals consistent with selfish mining strategies. In these intervals, Qubic's average hashrate share rises to the 23-34\% range, yet sustained 51\% control is never observed. We evaluate the campaign against the classical selfish mining model and a modified Markov-chain model that reflects Qubic's conservative release strategy: both predict lower revenue than honest mining at the inferred parameters, and the data largely confirms this while still showing noticeable deviations from the predicted curve. We interpret this gap between model and measurements in terms of Qubic's time-varying hashrate and coarse-grained attack segmentation.

Paper Structure

This paper contains 28 sections, 1 theorem, 22 equations, 10 figures, 3 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Data Collection
  3. Block and mining information collection
  4. Qubic block attribution
  5. Qubic's Mining on Monero
  6. Qubic's mining power share
  7. Orphan Blocks by Qubic
  8. Selfish Mining Focused Analysis
  9. Selfish mining strategies
  10. Reliability of Qubic block timestamps
  11. Selfish mining period categorization
  12. Analytical model of Qubic's selfish mining behavior
  13. Expected revenue under observed parameters
  14. Comparing the theoretic expectation and observed revenue
  15. Analysis on difference between revenue expectation by theory and observation
  16. ...and 13 more sections

Key Result

Theorem 1

Let $R_{\mathrm{mod}}(\alpha,\gamma)$ denote the selfish pool's long-run fraction of accepted blocks under the modified strategy encoded by the state machine in Fig. fig:modified_selfish_mining_state_machine, where $\alpha \in (0,\tfrac12)$ is the pool's relative hash power and $\gamma \in [0,1]$ is

Figures (10)

  • Figure 1: Qubic mining pool's mining power share on the Monero network
  • Figure 4: Qubic blocks' timestamp difference distribution with the same heigh blocks including orphan blocks
  • Figure 5: Time delay distribution between Qubic block timestamps and job fetch timestamps. The mean delay is 5.61 seconds, and the green line marks the near 8-second job fetch period.
  • Figure 6: Orphan block frequency and selfish mining periods
  • Figure 7: Distribution of orphan blocks per Qubic run length across ten selfish mining periods (a-j). The scatter plots visualize run frequency (encoded by point size and color) for each run length and orphan count pair. Reference lines include the theoretical ideal selfish mining line $y = x - 1$ (blue dashed) and the secondary threshold $y = x - 2$ (red dotted).
  • ...and 5 more figures

Theorems & Definitions (3)

  • Theorem 1
  • proof
  • proof