AI agents pose security and governance risks to enterprises

Claims

What the coverage asserts, and how many sources carry each claim.

• [○ 1 SOURCE] There are 109 machine identities for every one human identity.
• [○ 1 SOURCE] 90% of IT and security leaders surveyed believe they can detect malfunctioning AI agents, but only 26% can trace downstream impacts within minutes. www.csoonline.com
• [○ 1 SOURCE] Non-human identities outnumber human users in 83% of organizations, but only 21% have governance in place. jumpcloud.com
• [○ 1 SOURCE] 40% of organizations reported inaccurate AI outputs in the past 12 months. techcoffeehouse.com
• [DISPUTED] 66% of organizations are piloting or scaling AI agents, while 53% lack clear governance frameworks. businessnewsweek.in
• [○ 1 SOURCE] The iNSiM technology uses a probe of less than 100 KB to monitor a device’s instruction set for unauthorized modifications. interestingengineering.com
• [○ 1 SOURCE] The average remediation latency for configuration flaws is 14 months. flagthis.com
• [○ 1 SOURCE] Operating AI models on fragmented or low-quality data can lead to hallucinations and flawed intelligence. www.enterprisesecuritytech.com
• [○ 1 SOURCE] Just under a third of healthcare executives are focusing on virtual-first, AI-orchestrated access. healthtechmagazine.net
• [○ 1 SOURCE] Half of organizations using AI-based security controls still experienced suspicious or confirmed AI-related incidents. www.itpro.com
• [○ 1 SOURCE] AI-driven attacks increased by 56 percent year-over-year in 2026. interestingengineering.com
• [DISPUTED] 87% of organizations have moved AI assistants beyond the pilot stage. www.itpro.com