AI Privacy 101: Use ChatGPT & Claude Without Oversharing
Learn what really happens to your ChatGPT and Claude chats, and the settings and habits that keep your data safe. 8 lessons, first 2 free, plus a certificate.
You type something into ChatGPT — a rough email about a coworker, a health question, a paragraph of a contract — and a small voice in the back of your head asks: wait, where does this go? Most people ignore that voice and keep typing. A few get scared and quit using AI for anything real. Both are the wrong move.
Here’s the honest version. Your chats aren’t floating out in the open, but they aren’t a locked diary either. They’re stored on a company’s servers, they can be used to train the next model unless you say no, they stick around for a while even after you delete them, and — as thousands of people found out in July 2026, when shared Claude chats turned up in Google search — one wrong click can make a “private” conversation public. None of that means AI is dangerous. It means it’s a tool with a memory, and you should know how the memory works.
That’s what this course does, in one sitting. Not privacy theory — the actual toggles, on the actual screens, with the exact words to look for. By Lesson 2 you’ll have opened your real settings and found the switch. By the end you’ll have turned off training on both tools, learned what Temporary Chat and incognito really protect (and what they don’t), audited any chat you’ve ever shared, and built a six-point baseline you can hand to your team or your kids. You’ll never again wonder whether it’s safe to paste something — you’ll just know.
The first two lessons are free, no login. If you want the plain-English news version first, read Is ChatGPT Actually Private? and our walkthrough on checking and unsharing AI chats that show up on Google, or the glossary explainer on Temporary Chat. When you’re ready to go deeper, AI Fundamentals teaches how these tools actually work, and ChatGPT Without the Liability is the professional next step for anyone bound by client or patient confidentiality.
What You'll Learn
- Explain in one sentence what happens to a message you send to ChatGPT and to Claude
- Use both tools' data controls to turn off model training and verify it stuck
- Apply Temporary Chat and incognito modes correctly, knowing exactly what they do and don't protect
- Evaluate whether a chat you shared is publicly exposed and revoke it
- Apply a four-question filter to decide what's safe to paste before every sensitive prompt
- Design a personal six-point AI-privacy baseline you can share with a teammate or family member
After This Course, You Can
What You'll Build
Course Syllabus
Prerequisites
- A ChatGPT and/or Claude account (the free tier is fine)
- No technical background needed
Who Is This For?
- Anyone who uses ChatGPT or Claude for work or life and quietly wonders where their words go
- Professionals who paste work documents into AI and want to know what's actually safe
- People with no technical background who want a one-sitting fix, not a security course
- Parents, managers, and teammates who want a simple privacy habit they can pass on
Frequently Asked Questions
Do I need any technical skills or a paid plan?
No. Everything in this course works on the free tiers of ChatGPT and Claude, and there's nothing to install. If you can open Settings, you can do this.
Which AI tools does this cover?
ChatGPT (OpenAI) and Claude (Anthropic), side by side, because most people use both. The habits transfer to Gemini and Copilot too — we note where those differ.
Will the settings I learn here go out of date?
Privacy settings do move, so we teach you where to look and what each control means — not just today's menu path. We also review this course every three months and update the screenshots and labels.
Is turning off training the same as my chats being private?
No, and that gap is the whole point of the course. Turning off training stops your future chats from teaching the model — it doesn't delete them or hide them from the company. Every lesson is careful about the difference.
Do I get a certificate?
Yes. Finish all eight lessons, pass the quizzes, and complete the capstone baseline to earn a verifiable AI Privacy 101 certificate.
