Research
New Report From IRONSCALES Reveals 77% of Organizations Experienced Agentic AI-Related Security Incidents in the Past Year
Publication date unknown · Discovered ironscales.com
Page published
Publication date unknown · First observed: 9 Nov 2026
Coverage timeline
Single-source research — one report is available.
Why it matters
The IRONSCALES survey highlights the need to restrict agent permissions and monitor actions triggered by untrusted messages, particularly when agents can access internal files, inboxes, or payment systems.
IRONSCALES and Censuswide surveyed 501 U.S. IT and cybersecurity professionals: 77% said an AI agent or tool had acted on malicious, misleading, or unauthorized instructions or content in the past year, with 57% reporting a confirmed case. Reported incidents included agents summarizing or acting on malicious messages, while widespread agent access to internal systems increased potential exposure.