See the sections below for frequently asked questions about Agent Risk Manager.
Can admins see prompt text in the AI Activity Log?
No. The AI Activity Log records that a detection occurred and shows details such as the event type, user, provider, and date and time. It does not display prompt text.
How is my data handled?
Data Collection
When Agent Risk Manager evaluates an AI agent's tool execution, it receives the user's prompt, the tool calls and their inputs, the data the agent's tools return, and agent, user, and timestamp metadata. This information is needed for the detection engines to detect risky AI interactions.
The conversation context is also included so that the detection engines can evaluate the prompt against what came beforehand and better understand the context.
Data Processing
All analysis runs on models hosted by KnowBe4 in KnowBe4-controlled AWS infrastructure.
When sensitive information is detected in a prompt, it is redacted before conversation content is stored. Agent Risk Manager records the detected data types.
Data Storage
Data is encrypted at rest and in transit, and prompt content is never sent to any third-party AI providers.
KnowBe4 Data Practices
For detailed information on how KnowBe4 collects and secures your data, see our KnowBe4 Security articles and KnowBe4 Data Usage & Artificial Intelligence.
Does the Redirect toggle redirect users from an AI tool?
The Redirect toggle in the AI Tool Registry table doesn't send users away from the AI tool in that row. It determines whether the AI tool is a redirect, one of your Approved AI tools that users are offered when they open a Blocked AI tool. You can choose up to three redirects.
How can I leave feedback?
To give feedback and discuss product features with other admins, join the Agent Security Agent Security (link opens in new window) group in the KnowBe4 Community. For more information, see our KnowBe4 Community Guide.