Cybersecurity Statistics

Cybersecurity statistics / AI Security

53% of enterprises have had AI agents exceed their intended permissions, leaving them vulnerable to increased risk.

Publisher	Cloud Security Alliance (CSA) & Zenity
Report	Enterprise AI Security Starts with AI Agents
Published	16 April 2026
Topics	AI Security, AI Governance, Access Control, Enterprise

Published by Cloud Security Alliance (CSA) & Zenity in Enterprise AI Security Starts with AI Agents , 16 April 2026. The figure is taken from the report as published; the full methodology is in the source.

View the original report

Compare this across sources

12 sources answer what share of organisations have an ai governance policy in place. They report between 18% and 44%, with a median of 36.5%.

See all 12 sources

Related statistics

62% of enterprises are concerned about training data breaches in AI workflows.

Perforce Software, 25/07/2026

Only 7.6% of surveyed OT and ICS cybersecurity professionals are very confident in controls that protect AI tools and models because those controls have been tested.

Nozomi Networks, 25/07/2026

Only 7.9% of surveyed OT and ICS cybersecurity professionals have deployed AI for multiple OT cybersecurity functions.

Nozomi Networks, 25/07/2026

21% of surveyed OT and ICS cybersecurity professionals report that agentic AI has reached an active pilot, proof of concept, or production environment.

Nozomi Networks, 25/07/2026

US organizations report AI-enhanced attack effectiveness at 81%.

Proofpoint, 25/07/2026

63.9% of surveyed OT and ICS cybersecurity professionals consider attacks targeting AI systems or operational data either a top-tier operational risk (20.5%) or an emerging priority (43.4%).

Nozomi Networks, 25/07/2026

Get the newsletter

Weekly cybersecurity statistics by email.