About

Security is the foundation of cooperation among AI agents, humans, institutions, and nations. Progress in areas such as LLMs/GenAI, multi-agent learning, and human-AI interaction offer great potential, but also highlight new security challenges. Yet, the cyber security, AI capabilities, safety and policy communities have limited platforms to share cross-cutting knowledge. The field of Multi-Agent Security (MASEC) aims to design a blueprint of the emerging and amplified security challenges of our present and future multi-agent world. 

At this in-person NeurIPS workshop, we will gather researchers from cyber security, and AI, including AI safety and AI governance, to connect the fields that complement each other. We will showcase on-going research via contributed talks and poster sessions, and provide networking opportunities for research and mentorship discussions.