CyberGym: Evaluating AI Agents' Real-World Cybersecurity Capabilities at Scale

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Zhun, Shi, Tianneng, He, Jingxuan, Cai, Matthew, Zhang, Jialin, Song, Dawn
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912979488866304
author Wang, Zhun
Shi, Tianneng
He, Jingxuan
Cai, Matthew
Zhang, Jialin
Song, Dawn
author_facet Wang, Zhun
Shi, Tianneng
He, Jingxuan
Cai, Matthew
Zhang, Jialin
Song, Dawn
contents AI agents have significant potential to reshape cybersecurity, making a thorough assessment of their capabilities critical. However, existing evaluations fall short, because they are based on small-scale benchmarks and only measure static outcomes, failing to capture the full, dynamic range of real-world security challenges. To address these limitations, we introduce CyberGym, a large-scale benchmark featuring 1,507 real-world vulnerabilities across 188 software projects. Adjustable to different vulnerability analysis settings, CyberGym primarily tasks agents with generating a proof-of-concept test that reproduces a vulnerability, given only its text description and the corresponding codebase. Our extensive evaluation highlights that CyberGym effectively differentiates agents' and models' cybersecurity capabilities. Even the top-performing combinations only achieve a ~20% success rate, demonstrating the overall difficulty of CyberGym. Beyond static benchmarking, we show that CyberGym leads to the discovery of 34 zero-day vulnerabilities and 18 historically incomplete patches. These results underscore that CyberGym is not only a robust benchmark for measuring AI's progress in cybersecurity but also a platform for creating direct, real-world security impact.
format Preprint
id arxiv_https___arxiv_org_abs_2506_02548
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CyberGym: Evaluating AI Agents' Real-World Cybersecurity Capabilities at Scale
Wang, Zhun
Shi, Tianneng
He, Jingxuan
Cai, Matthew
Zhang, Jialin
Song, Dawn
Cryptography and Security
Artificial Intelligence
Machine Learning
AI agents have significant potential to reshape cybersecurity, making a thorough assessment of their capabilities critical. However, existing evaluations fall short, because they are based on small-scale benchmarks and only measure static outcomes, failing to capture the full, dynamic range of real-world security challenges. To address these limitations, we introduce CyberGym, a large-scale benchmark featuring 1,507 real-world vulnerabilities across 188 software projects. Adjustable to different vulnerability analysis settings, CyberGym primarily tasks agents with generating a proof-of-concept test that reproduces a vulnerability, given only its text description and the corresponding codebase. Our extensive evaluation highlights that CyberGym effectively differentiates agents' and models' cybersecurity capabilities. Even the top-performing combinations only achieve a ~20% success rate, demonstrating the overall difficulty of CyberGym. Beyond static benchmarking, we show that CyberGym leads to the discovery of 34 zero-day vulnerabilities and 18 historically incomplete patches. These results underscore that CyberGym is not only a robust benchmark for measuring AI's progress in cybersecurity but also a platform for creating direct, real-world security impact.
title CyberGym: Evaluating AI Agents' Real-World Cybersecurity Capabilities at Scale
topic Cryptography and Security
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2506.02548