# Are Frontier LLMs Ready for Cybersecurity? Evidence for Vertical Foundation Models from Dual-Mode Vulnerability Benchmarks

arXiv: https://arxiv.org/abs/2605.23243 · HTML: https://mlsanity.com/paper/2605.23243/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 29 references that are in the map.
- First posted on arXiv 22 May 2026.

## Builds on (most cited first, 18 of 18)

- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [BloombergGPT: A Large Language Model for Finance](https://mlsanity.com/paper/2303.17564/) (2023)
- [Large Language Models for Software Engineering: Survey and Open Problems](https://mlsanity.com/paper/2310.03533/) (2023)
- [Vulnerability Detection with Code Language Models: How Far Are We?](https://mlsanity.com/paper/2403.18624/) (2024)
- [PentestGPT: An LLM-empowered Automatic Penetration Testing Tool](https://mlsanity.com/paper/2308.06782/) (2023)
- [CVE-Bench: A Benchmark for AI Agents' Ability to Exploit Real-World Web Application Vulnerabilities](https://mlsanity.com/paper/2503.17332/) (2025)
- [NYU CTF Bench: A Scalable Open-Source Benchmark Dataset for Evaluating LLMs in Offensive Security](https://mlsanity.com/paper/2406.05590/) (2024)
- [CyberSecEval 2: A Wide-Ranging Cybersecurity Evaluation Suite for Large Language Models](https://mlsanity.com/paper/2404.13161/) (2024)
- [HackSynth: LLM Agent and Evaluation Framework for Autonomous Penetration Testing](https://mlsanity.com/paper/2412.01778/) (2024)
- [Comparing AI Agents to Cybersecurity Professionals in Real-World Penetration Testing](https://mlsanity.com/paper/2512.09882/) (2025)
- [Chasing Shadows: Pitfalls in LLM Security Research](https://mlsanity.com/paper/2512.09549/) (2025)
- [Cybersecurity AI Benchmark (CAIBench): A Meta-Benchmark for Evaluating Cybersecurity AI Agents](https://mlsanity.com/paper/2510.24317/) (2025)
- [VulnLLM-R: Specialized Reasoning LLM with Agent Scaffold for Vulnerability Detection](https://mlsanity.com/paper/2512.07533/) (2025)
- [Sifting the Noise: A Comparative Study of LLM Agents in Vulnerability False Positive Filtering](https://mlsanity.com/paper/2601.22952/) (2026)
- [LLM-based Vulnerability Discovery through the Lens of Code Metrics](https://mlsanity.com/paper/2509.19117/) (2025)
- [LLM Agents for Automated Web Vulnerability Reproduction: Are We There Yet?](https://mlsanity.com/paper/2510.14700/) (2025)
- [ZeroDayBench: Evaluating LLM Agents on Unseen Zero-Day Vulnerabilities for Cyberdefense](https://mlsanity.com/paper/2603.02297/) (2026)
- [Measuring and Exploiting Contextual Bias in LLM-Assisted Security Code Review](https://mlsanity.com/paper/2603.18740/) (2026)

## Abstract (arXiv)

We evaluate whether frontier LLMs are ready for cybersecurity through a dual-mode benchmark: white-box function-level vulnerability detection (VulnLLM-R, across C/Java/Python) and black-box web application security testing (five production-style applications with 118 ground-truth vulnerabilities across 20+ CWE families, which we will open-source). We test six frontier models (GPT-5.4, Codex~5.3, Claude Opus~4.7, Sonnet~4.6, Gemini~3.1~Pro and Gemini~3~Flash) and two domain-specialized models across four testing paradigms. Our findings are sobering: (1)~every frontier model produces 10-50% false positive rates in white-box detection, systematically over-predicting vulnerabilities; (2)~in black-box testing, frontier models achieve only 4-8% ground-truth coverage, improving to just 10-19% even with external security tools (Playwright MCP, Burp Suite MCP); (3)~structured penetration-testing methodology encoded in domain-specialized agents raises per-family detection above 50%, demonstrating that methodology, not scale, is the primary lever; and (4)~a domain-specialized defense model achieves the highest precision (0.904) and lowest false positive rate (9.7%) among all models, on a single GPU. We identify the absence of structured security testing traces end-to-end request/response sequences, failure-heavy data, and multi-step attack chains as the fundamental training data bottleneck, and propose self-play security testing as a data generation strategy. Our results make the case for vertical foundation models purpose-built for cybersecurity.