Following and shaping workplace AI rules

When you finish, you can find your workplace rules, follow them, and ask smart questions when rules are missing.

Find the rules first

Many workplaces now have an AI policy, or a section of an existing policy about technology and data. It may be in a staff handbook, an intranet page or an email from IT. Look before you start, not after a problem.

If you cannot find one, ask your manager three questions: which AI tools are approved, what information may go in, and whether AI use must be labeled. Write the answers down.

Typical rules you may see

Common rules include: use only approved tools, never enter customer or employee personal data into unapproved tools, check AI output before using it, do not use AI to make final decisions about people, and label AI-assisted work where needed.

Rules exist to protect customers, colleagues and you. Following them is part of doing the job well, like locking a door. If a rule blocks a good idea, raise it with the right person rather than working around it.

Shadow AI

Shadow AI means using unapproved tools for work without telling anyone. It often starts with good intentions: someone wants to save time. But it can leak company secrets or customer data, and the company may not even know it happened.

Weak: pasting a client contract into a free app you found. Strong: asking your manager, "Can we get an approved tool for summarizing documents, and what data would be allowed?" You solve the same need without the risk.

Be part of the solution

Workers who understand AI risks help their teams. You can suggest a first small trial, offer a list of safe tasks, and share what worked. You can also speak up kindly when something looks risky.

A worked example: Priya, an office manager in Pune, notices colleagues pasting customer emails into a chatbot. She does not blame anyone. She asks IT for an approved tool and shares a one-page list of safe and unsafe examples. The team adopts it within a week.

Practice now

Search your workplace or school for an AI policy. If none exists, write the three questions you would ask and who you would ask.
Write a one-page list for your team: five safe tasks, five tasks to avoid, and who to ask when unsure.

Use any AI assistant you have. Do not paste private, client or patient information into it.

Check yourself

1. You cannot find an AI policy at work. What is a good first step?

2. What is shadow AI?

3. A rule blocks a good AI idea. What is the best response?

Ask the AI tutor

Ask anything about this lesson, in any language. The tutor is an AI: it can be wrong, and it never needs your personal details.

Take the final exam

AI-drafted. Checked by an automated validator and a separate AI reviewer; a person has not read every lesson. Translations and tutor answers are AI-made. Found a mistake? Tell us. AI can be wrong. General information only, not professional advice. Not affiliated with any AI company named. Current University is not accredited.