# OpenAI Security Testing Exposes Monitoring Challenges for AI Agents
OpenAI’s security testing initiatives have revealed significant challenges in effectively monitoring autonomous AI agents during operation. The findings highlight gaps in current oversight mechanisms as AI systems become increasingly autonomous and complex.
