The agent security mess
Briefly

The agent security mess
Humans typically ignore 96% of their granted permissions, with only 4% used within 90 days. In contrast, AI agents can trigger security vulnerabilities by exploiting these dormant permissions. Research shows that workers access only 9% of sensitive data they can reach, and a significant portion can modify or delete it. This situation creates a precarious security landscape, akin to a persistent weak layer in snow that can lead to disastrous outcomes when triggered by rapid, unrestrained actions of AI agents.
"According to the Oso blind spot report, corporate workers completely ignore 96% of their granted permissions. Over a 90-day window, only 4% of granted permissions were ever actually exercised."
"With sensitive enterprise data, it's even worse: Workers touch only 9% of the sensitive data they can actually reach, and nearly one-third of users have the power to modify or delete sensitive data."
Read at InfoWorld
Unable to calculate read time
[
|
]