Gelei Deng

☕️

Gelei Deng

AI Engineer, OpenAI Red Team Member

About Me

I am a Ph.D. in Computer Science from Nanyang Technological University, where I was advised by Prof. Tianwei Zhang and Prof. Yang Liu. Currently, I lead AI research at Quantstamp as an AI Engineer, while contributing to AI safety through my role in the OpenAI Red Team and participation in the Anthropic Red Teaming Challenge. I have contributed to several high-impact projects, including GPT-4o and Operator.

My research interests span a broad range of cybersecurity domains, encompassing system security, AI security, blockchain security, and security testing. I am particularly passionate about AI for Security: leveraging artificial intelligence for security-related tasks, and advancing AI Safety and Security research.

Download CV

Interests

Artificial Intelligence Security
Large Language Models
System Security

Education

PhD in Computer Science
Nanyang Technological University

📚 My Research

I’m working toward a future where AI systems are both powerful and inherently secure—where artificial intelligence can defend itself and protect the digital world around us. By advancing AI for Security and building robust defenses against emerging threats, we’re laying the foundation for an era where intelligent systems can autonomously safeguard our increasingly connected world. Let’s collaborate to make this vision a reality!

Featured Publications

Large Language Models

PentestGPT: Evaluating and Harnessing Large Language Models for Automated Penetration Testing

An LLM-empowered automated penetration testing framework that leverages domain knowledge inherent in LLMs, achieving 228.6% task completion improvement over baseline GPT models.

Aug 14, 2024

Large Language Models

MASTERKEY: Automated Jailbreaking of Large Language Model Chatbots

A comprehensive framework for automated jailbreaking of Large Language Model chatbots, featuring novel attack methodologies and systematic analysis of defense mechanisms.

Feb 26, 2024

Recent Publications

Gelei Deng, Yi Liu, Víctor Mayoral-Vilches, Peng Liu, Yuekang Li, Yuan Xu, Tianwei Zhang, Yang Liu, Martin Pinzger, Stefan Rass (2024). PentestGPT: Evaluating and Harnessing Large Language Models for Automated Penetration Testing. USENIX Security 2024.

PDF Code USENIX arXiv

Gelei Deng, Yi Liu, Yuekang Li, Kailong Wang, Ying Zhang, Zefeng Li, Haoyu Wang, Tianwei Zhang, Yang Liu (2024). MASTERKEY: Automated Jailbreaking of Large Language Model Chatbots. NDSS 2024.

PDF Cite Code DOI NDSS

Gelei Deng, Robert Ford (2015). An example journal article. Journal of Source Themes, 1(1).

PDF Cite Code

Recent & Upcoming Talks