• I lead Frontier AGI Security Research at Google DeepMind . I earned CS Ph.D. from Georgia Tech, advised by Dr. Wenke Lee.

• My research lies at the intersection of machine learning and systems security, with a focus on identifying and mitigating emerging threats in AI-driven environments through techniques such as safety alignment, adversarial defense. I share my findings with IEEE S&P, USENIX SEC, ACM CCS, BlackHat USA, etc.

Community Services

• ACM CCS Program Committee, 2026
• ICLR Program Committee, 2025
• ACM WWW Program Committee, 2024, 2025
• NeurIPS Program Committee, 2024
• SIG DIMVA Program Committee, 2023
• USENIX Security External Reviewer, 2021-2023

Recent News

• May 2025. My AI security research work at Palo Alto Networks directly enabled the launch of their “Agentic AI Red Teaming” product. Congratulations to the team!

• Aug 14, 2024. One US patent filed with Palo Alto Networks on Large Language Model (LLM) data leakage detection!

• Aug 1, 2024. Our work on context-aware LLM jailbreaking has uncovered 60+ security risks from OpenAI marketplace applications!

• Mar 1, 2024. Our work on defending adversarial learning attacks against bio-authentication systems is accepted to USENIX Security 2024!

Full list