Does AI Help Cyber Attackers or Defenders? Evidence from Nonpublic Vulnerabilities and Subsequent Attacks
cs.CR, cs.AI
Submitted: 2026-10-05
Updated: 2026-10-05
Comments: 8 pages, 1 figure
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Cybench: A Framework for Evaluating Cybersecurity Capabilities and Risks of Language Models
- CTFusion: A CTF-based Benchmark for LLM Agent Evaluation
- AutoBaxBuilder: Bootstrapping Code Security Benchmarking
- Benchmarking Practices in LLM-driven Offensive Security: Testbeds, Metrics, and Experiment Design
- How Do LLM Agents Actually Get the Flag? Trace-Level Provenance for Agentic Offensive Security Evaluation
- The Oracle's Gambit: A Game-Theoretic Framework for Responsible AI Release
- CyberGym: Evaluating AI Agents' Real-World Cybersecurity Capabilities at Scale
- ARVO: Atlas of Reproducible Vulnerabilities for Open-Source Software
- SWE-bench: Can Language Models Resolve Real-World GitHub Issues?
- SEC-bench: Automated Benchmarking of LLM Agents on Real-World Software Security Tasks
- BountyBench: Dollar Impact of AI Agent Attackers and Defenders on Real-World Cybersecurity Systems
Related papers
- SoK: AI-Augmented Binary Reversing
- Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
- Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
- Efficient Fuzzy PSI under One-Sided Assumptions
- Sealing the Audit-Runtime Gap for LLM Skills
- Token Composition: A Graph Based on EVM Logs