Run Wang · Wuhan University
We study the security boundaries of generative AI, build safeguards for responsible intelligent systems, and use AI to make security analysis more autonomous.
Our research
Our work follows threats across the full AI lifecycle: uncovering failure modes, designing practical defenses, and building AI systems that strengthen security.
Identify and mitigate emerging risks in agentic systems across memory, tools, integrations, and runtime environments.
Improve the authenticity, alignment, and safety of AI-generated content through detection, evaluation, and model safeguards.
Use the reasoning, coding, and tool-use capabilities of AI to make vulnerability research deeper and more autonomous.
Latest news
Our team won the Grand Prize in the Finals of the 11th Network Technology Challenge.
Our team won the First Prize in the 2nd University Student Artificial Intelligence Security Competition.
Run Wang was invited to serve on the Technical Program Committee for USENIX Security 2027.
One paper was accepted by ICML 2026. Congratulations to Yanhao Wei.
One paper was accepted by ACM CCS 2026. Congratulations to Yuyang Zhang.
Join the lab
We welcome motivated students interested in AI Security, AI Safety, and AI for Security.
Undergraduate·Master·PhD
Start a conversation
