Gynopticon: Consensus-Based Cheating Detection System for Competitive Games

TL;DR

This work tackles cheating in competitive online games by proposing GYNOPTICON, a privacy-preserving framework that uses mutual user consensus to detect cheaters without kernel-level monitoring. It combines lightweight client-side detectors (voting initiators) with a server-side cheater discriminator that maintains a ground-truth through a consensus algorithm, using a Validity score $V_N$ and a Dubious score $D_{target}$ weighted by $W_v$ and $W_h$ and tracked via a history queue $Q_N$. The authors validate the approach through simulations and a real-world FPS-like dataset, showing reliable discrimination when the detection accuracy is sufficiently high (roughly $80\%$) and voting participation is adequate, while exploring policy mechanisms (adding vs counting) for long-term game management. The work highlights that a distributed, consensus-based, and privacy-conscious anti-cheat paradigm can be practical for large-scale competitive games and provides datasets and code to support further research. Future directions include extending to MOBA and Action genres and addressing scenarios where cheaters approximate or exceed half of the player population.

Abstract

Cheating in online games poses significant threats to the gaming industry, yet most prior research has concentrated on Massively Multiplayer Online Role-Playing Games (MMORPGs). Competitive genres-such as Multiplayer Online Battle Arena (MOBA), First Person Shooter (FPS), Real Time Strategy (RTS), and Action games-remain underexplored due to the difficulty of detecting cheating users and the demand for complex data and techniques. To address this gap, many game companies rely on kernel-level anti-cheat solutions, which, while effective, raise serious concerns regarding user privacy and system security. In this paper, we propose GYNOPTICON, a novel cheating detection framework that leverages user consensus to identify abnormal behavior. GYNOPTICON integrates a lightweight client-side detection mechanism with a server-side voting system: when suspicious activity is identified, clients cast votes to the server, which aggregates them to establish consensus and distinguish cheaters from legitimate players. This architecture enables transparency, reduces reliance on intrusive monitoring, and mitigates privacy risks. We evaluate GYNOPTICON in both a controlled simulation and a real-world FPS environment. Simulation results verify its feasibility and requirements, while real-world experiments confirm its effectiveness in reliably detecting cheating users. Furthermore, we demonstrate the system's applicability and sustainability for long-term game management using public datasets. GYNOPTICON represents a user-driven, consensus-based alternative to conventional anti-cheat systems, offering a practical and privacy-preserving solution for competitive online games.

Figures (7)

• Figure 1: Overall design of GYNOPTICON: the design and concept is described in \ref{['sec:methodoloy-overview']}---➊,➋,➌,➍, and the basic requirements is discussed in \ref{['subsec:simulation_overview']}---➀,➁,➂,➃
• Figure 2: Results of the first simulation. The proposed consensus algorithm was examined. (a) demonstrates the Dubious scores of normal users and cheating users at fixed 80% of model acc. (b) demonstrates the scores of cheating and normal users at different model acc from 50% to 100%
• Figure 3: Results of the second simulation. The minimum model acc was examined. (a) demonstrates the needed accuracy at a fixed cheater rate of 33%. (b) demonstrates the changing accuracy at different cheater rates from 0% to 50%
• Figure 4: Results of experiment. (a) demonstrates the standardized final Dubious scores of all validations and box plots of the final scores. To fix a threshold at 0 across all validations, for each validation, a threshold was subtracted from final scores. (b) is the confusion matrix of the results.
• Figure 5: Simulation setting for a real-like online game. A detailed explanation and the dataset used for each step can be found in \ref{['subsec:simulation_setting']}