What Not to Share With ChatGPT: 7 Privacy Rules
Seven simple privacy rules for using ChatGPT and other AI chatbots: what never to type, why it matters, and safer ways to get help.
Short answer: what not to share with ChatGPT or any AI chatbot comes down to seven rules: no passwords or API keys, no identity numbers, no financial details, no health records, no confidential work files, no other people's private data, and nothing you would not want stored. Settings and data policies differ between tools and places, so check each tool's privacy page.
Why Does Privacy Matter With AI Chatbots?
When you type into a chatbot, your text is sent to the company's servers. Depending on the product and your settings, conversations may be stored, reviewed for safety, or used to improve models. Rules for this vary by tool, plan and local law, so do not assume. The safest habit is to share only what you would be comfortable having stored.
The 7 Privacy Rules
1. Never share passwords, keys or codes
This includes passwords, one-time codes, recovery phrases and API keys. If you paste code for debugging, remove the secrets first.
2. Do not share identity numbers or documents
National ID numbers, passport details, driving licence numbers and photos of documents are enough for identity theft in the wrong hands.
3. Do not share financial details
Card numbers, bank account details and full statements should stay out. Describe the situation with made-up numbers instead.
4. Be careful with health and legal records
Medical reports and legal documents are highly sensitive. If you want help understanding one, remove names and identifiers, and remember that a chatbot is not a doctor or lawyer.
5. Do not paste confidential work or school material
Customer lists, unreleased plans, source code under a non-disclosure agreement and exam papers can break company rules or contracts. Read your organisation's AI policy, or ask.
6. Protect other people's data
Chats, emails or photos that contain someone else's name, address or private details are not yours alone to share. Anonymise first.
7. Assume anything you type could be stored
Even with privacy settings, treat the chat as semi-public. If it would hurt you to see it leaked, do not type it.
How Can You Use AI Safely Instead?
Instead of this | Do this |
|---|---|
Pasting a real contract | Paste a version with names and amounts replaced by placeholders |
Sharing your real salary slip | Ask the question with a made-up example figure |
Pasting code with a live API key | Replace the key with |
Uploading a customer spreadsheet | Use five invented rows with the same columns |
Also look for the privacy or data-controls settings in the tool, such as options about chat history and model training. The menus change, so look in the current settings page. For more safe prompts, see what you can do with ChatGPT, and for the wider issue read context windows, bias and privacy.
Common Mistakes
Thinking “it is just a chatbot, so it is private”.
Pasting a whole email thread without checking who is named in it.
Using a work account policy for personal data, or the reverse.
Skipping the settings page and never learning what the tool does with chats.
Key Takeaway
Share only what you would be fine with being stored. Keep secrets, identity, money, health and confidential material out of chats, replace real details with placeholders, and check the tool's data settings.
Ready to start? Learn how to use AI tools responsibly, including privacy, consent and the data you share. Begin with the free course AI Ethics and Responsible Use: Beginner.
Related articles
More in AI Ethics
What Is AI Bias? Real Examples
AI bias is systematically unfair performance for some groups. Learn where it comes from, common real-world patterns, how to check for it and the ethics principles behind fair AI.
4 min readNov 15, 2026How Reliable Are AI Detectors?
AI detectors guess from writing patterns and can flag honest human work. Learn why false positives happen and what students and teachers should do instead of trusting a score.
3 min readNov 5, 2026What Is a Deepfake and How Can You Spot One?
A deepfake is AI-generated or altered video, image or audio made to deceive. Learn the harms, the warning signs, the verification habits that work, and what to do if you are targeted.
4 min readNov 2, 2026Writing Your CV and Cover Letter With ChatGPT
How to use ChatGPT to improve your CV and draft a cover letter without inventing facts: working prompts, a safety checklist and common mistakes.
4 min readNov 16, 2026How to Keep API Keys Safe: .env Files and Secrets
Keep API keys out of your code: use .env files and environment variables, ignore them in Git, and know exactly what to do if a key leaks.
4 min readNov 14, 2026What Is Prompt Injection?
Prompt injection tricks an AI app into following an attacker's instructions. Learn the types, why it is dangerous and the layers of defence that reduce the risk.
4 min readNov 12, 2026
Comments
Sign in to join the discussion.

