A cybersecurity report indicates that 54% of firms lack the capability to restrict access for AI agents as of 2027. The finding highlights a significant security gap in organizational AI governance as adoption of autonomous AI systems continues to expand across enterprises.