Technology · AI and automation · Component

AI acceptable-use policy

What staff may and may not do with AI tools, which tools are sanctioned, and what material may be given to them.

The deliverable

What it is

Since general-purpose assistants became widely available, much AI use in organisations has been individual and unannounced. A policy does not prevent that; it gives people a clear way to use the tools well and a line they know not to cross.

It works best short, specific and paired with sanctioned tools. A policy that prohibits without offering an alternative tends to move use out of sight, where it is harder to govern.

One level in

What it is made of

Each element is a constituent part of the component. Follow one to see the attributes it carries.

  1. Sanctioned tools

    The AI tools approved for work use, the terms under which each was approved, and any restrictions attached.

    3 attributes: Tool · Approved for · Approved

    Learn
  2. Information rules

    Which classes of information may be given to which tools, keyed to the classification set in Data Core.

    3 attributes: Information class · Permitted tools · Example

    Learn
  3. Disclosure and review rules

    When AI involvement must be disclosed, to whom, and when output must be checked by a person before it is used.

    2 attributes: Disclosure required when · Review required when

    Learn
  4. Literacy expectation

    What each role is expected to understand about the AI tools it uses, and where that understanding comes from.

    3 attributes: Role · Expected understanding · Training route

    Learn

Pair every prohibition with an approved alternative. A ban without one moves the use out of sight.