Question 1

Why only five clauses?

Accepted Answer

Most AI policies run to four pages about responsible use, everyone signs them, and nine months later nobody can say whether a single clause was followed. Five clauses fit on a page, and a page gets read.

Question 2

Why does every clause name a record?

Accepted Answer

A clause nobody can check is unenforceable. Naming the record that shows it was followed is what turns the document into something you can review each quarter instead of a statement of intent.

Question 3

Who should own it?

Accepted Answer

Usually the delivery lead. They own what ships and they are the person asked when a new tool appears. Legal or compliance is common in regulated work, and slower to keep current.

Question 4

What happens when someone breaks it?

Accepted Answer

Treat the first breach as a process failure. Someone had a deadline, the approved tool fell short, and the unapproved one did the job. If the approved tool is bad, the tool list is wrong. If nobody knew the rule, the document is too long or was never circulated. Disregard for a client data term is a different conversation.

Question 5

What changes if our work is regulated?

Accepted Answer

Review and disclosure are set at their strongest whatever else you pick, and the document says where it did that. A generated policy will not state that a legal, medical or financial deliverable ships without review.

Question 6

Is this legal advice?

Accepted Answer

No. It is a starting point for your own policy. WhoWorked is not a law firm, and anything that touches a contract, a client agreement or a subcontractor agreement should be reviewed by a lawyer before you rely on it.

Question 7

What do you do with what I enter?

Accepted Answer

The policy is built in your browser. When you ask to see it, your work email and the policy you built reach our team, and nowhere else. We use it to follow up once.

An AI policy you can check, on one page.

FAQ