Is it safe for staff to use ChatGPT or Copilot at work?

Your staff are almost certainly already using ChatGPT and other AI tools — whether it has been officially approved or not. That is genuinely great for productivity, but it creates a real and often invisible risk: sensitive company or client information being pasted into public chatbots, where it leaves your control and may be retained, used for training, or exposed.

The real risk isn't AI — it's what goes into it

Picture an employee pasting a client contract, a spreadsheet of customer details, or confidential financials into a public AI tool to "summarise this quickly." That data has now left your business. For firms with confidentiality or privacy obligations — law, accounting, healthcare — that is potentially a serious breach, not just a bad habit.

Don't ban it — control it

An outright ban is hard to enforce and simply pushes staff to use AI on personal phones where you have zero visibility. The better approach is to allow AI but control what data can go into it:

• An acceptable-use AI policy — clear, simple rules on what can and can't be entered, and which tools are approved.
• Data-loss-prevention (DLP) controls that detect and block sensitive information from being uploaded into AI tools in the first place.
• A safer, approved option — such as Microsoft Copilot configured to keep data inside your own Microsoft 365 tenant, so staff get the productivity without the leakage.
• Staff awareness — a short briefing on why the controls exist, so people understand rather than route around them.

We've done exactly this for Perth businesses

One client, a Perth building company, came to us specifically worried about staff leaking data into chatbots. We deployed these exact controls — policy, DLP and an approved, contained AI option. Read the case study, or call (08) 9325 1196 to make AI a productivity win rather than a liability.