K-AI Control
Use AI without leaving your data behind
K-AI Control shows which generative AI services are used in the business, applies your rules and warns before sensitive data leaves in a prompt.
The problem
An employee pastes a customer table or an API key into an AI assistant to save time. The data leaves the business without anyone knowing.
What the module does
Usage inventory
Which AI services are used, by whom and how often, based on DNS and the extension.
Policies
Allow, block or allow only work accounts.
Leak prevention
Detection of IBANs, social security numbers, health data, source code or secrets before sending.
Exposure report
A summary for management, without the content of prompts.
How it works
Observe
DNS and the extension list the services in use.
Warn
The extension flags sensitive data before it is sent.
Decide
According to the policy: a simple warning or a block.
What you should know
- The extension and the dashboard are partly built; publication in browser stores is still to come.
- Analysis is done locally in the browser; the content of prompts is not sent to Kayzen.
Frequently asked questions
Questions about this module
Do employees need to be informed?
Yes. This type of tool must be presented to employees and, where applicable, to the works council (CSE) before it is deployed.
Want to try this module?
The pilot programme gives you access to modules as they are delivered, free of charge and with no commitment.