AI vs AI: Agent Runtime Security That Gets Stronger with Every Use
Tencent security expert Duan Haotong will share a full-chain runtime security solution for AI agents at AICon 2026 Shenzhen, using AI to combat AI. The approach covers pre-, during-, and post-incident defenses, including prompt injection detection, intent deviation detection, sandboxing, and data leak prevention, enabling continuous self-improvement based on real production data.
Tencent
