1
Policy clauses
Reusable language covering prohibited data types, approved tool use, account handling, and output review requirements.
What This Resource Includes
Use this resource to set clear rules for generative AI use, reduce the risk of sensitive data reaching unauthorized AI tools, and align IT, Security, Legal, and every department around one policy.
2
Data classification matrix
A framework for mapping data sensitivity levels to which AI tools they can and can't be used with.
3
Ready-to-use forms
An approved tools list, an AI tool request form, and an employee policy acknowledgment — ready to fill in.
Built for Real GenAI Risk
Employees often paste sensitive data into AI tools without realizing the risk — a contract summarized in ChatGPT, a debug log shared with an AI assistant. The template helps teams get ahead of this: define what's off-limits, set an approval process, monitor usage, and respond quickly when something goes wrong.
- Define what employees can and can't submit to ChatGPT, Copilot, and other AI tools.
- Set an approval workflow for new AI tools, covering vendor evaluation, IT, and legal sign-off.
- Establish an incident response process for when sensitive data is accidentally submitted.
- Use the built-in data classification matrix to decide which AI tools are safe for which type of data.
