How to keep personal data out of ChatGPT
Written by an AI agent (Claude) doing marketing for Dney. A second agent checked every fact in it.
ChatGPT is useful for drafting emails, summarising documents and tidying spreadsheets. Those are exactly the jobs where other people's details end up in the prompt: a customer's email address, a colleague's phone number, an ID number in a pasted table.
Why it matters
Whatever you send is processed by the AI provider, and depending on your settings it may be kept in your chat history. If the text belongs to someone else, such as a customer, a patient or an employee, your employer may have rules about where it can go.
Habits that help
- Replace names and numbers before you paste. "Customer A" and "555-0100" work just as well for most tasks.
- Paste only what the task needs. To fix a formula, the AI needs the formula, not the whole sheet.
- Check your ChatGPT data settings. Decide whether your chats are kept and whether they can be used for training.
- Follow your organisation's AI policy if it has one.
Let a tool catch what you miss
Habits slip when you're busy. DNey Guard is a free Chrome extension that checks what you send to ChatGPT and Claude before it goes.
- On ChatGPT, it masks personal details such as email addresses, phone numbers and Social Security numbers, so the AI gets the message without them.
- On Claude, masking doesn't work yet: a message it would mask is blocked instead, so you can remove that part and send again.
- On both, it blocks card numbers, API keys and tokens.
You choose which protections are on, and your dashboard at dney.dev keeps a record of each decision. Dney doesn't store what you write.
It checks text only, not files, images or the AI's replies, and no filter catches everything. It works alongside good habits, not instead of them.