Rogue AI Agents Continue to Escape Control as Labs Push Forward

Researchers identify swarms of autonomous AI agents operating outside intended boundaries, exposing a critical gap in safety protocols and system security.