Module 1
Your Private Claude Account
Why privacy comes first
Most courses would start by having you type your first question to an AI. We're going to spend ten minutes on something more important first, and here's why.
By the end of this course, you'll be using AI to work on real parts of your business — your pricing, your customers, your books. The conversations you have with an AI tool can include information you'd never post publicly. So before you type a single word, we're going to make sure your account is set up so your business information stays yours.
Doing this now, before your first conversation, means you never have to wonder later what happened to something you typed on day one.
Step 1: Create your free account
The AI we use in this course is called Claude — an AI assistant made by a company called Anthropic. You talk to it in plain English, like texting a very knowledgeable employee.
- Open your web browser and go to claude.ai.
- Click the sign-up option. You can use your email address or sign in with an existing Google account.
- Use your business email if you have one. It keeps work and personal life tidy, and if you ever bring your team on board, it's cleaner to have the business account under a business address.
- Follow the verification steps (Anthropic will confirm your email or phone).
Don't start chatting yet. There's one screen coming up that matters more than everything else in this module.
Step 2: The one toggle that matters
During signup, new users see a popup with Anthropic's consumer terms. Inside that popup is a toggle labeled "Help Improve Claude." It is pre-set to ON.
Find the "Help Improve Claude" toggle in the popup and turn it OFF. Then accept the terms.
Here's what that toggle actually controls, stated plainly:
- With the toggle OFF: your conversations are not used to train Anthropic's AI models. They are retained for about 30 days for technical and security purposes — for example, catching abuse and keeping the service running — and that's it.
- With the toggle ON: your conversations can be used to train Anthropic's AI models, and your data can be retained for up to 5 years.
One honest clarification: turning the toggle off does not mean nothing is ever stored. Your chats still exist in your account so you can read them later, and as noted above there's a short technical retention window. What the toggle controls is whether your words can become part of the training data for future AI models. For business information, that's the line that matters.
Also useful to know: when you delete a chat, it is removed from Anthropic's systems within about 30 days.
Step 3: Verify your setting
Popups are easy to click through. Let's confirm the setting stuck:
- Go to claude.ai/settings/data-privacy-controls (that's Settings → Privacy inside Claude).
- Find "Help Improve Claude."
- Confirm it shows OFF. If it's on, turn it off — the setting applies from that point forward.
Thirty seconds, done. Now your account is configured the way a careful business owner would want it.
Why this matters for a business owner
Think about what will flow through these conversations over the next few months: your margins and pricing strategy. Customer names and situations. Cash-flow questions in a tight month. How you win bids. The stuff you'd only say to your accountant or your spouse.
None of that should become training data for an AI model. Not because anything sinister would happen to it — but because private business information should stay private on principle, and because you may have promised customers and employees (in contracts, or just in trust) that their information stays inside your business.
You made a one-toggle decision that most people never think about. That already puts you ahead of most AI users, including plenty of technical ones.
Claude has an incognito chat mode — look for the ghost icon at the top right of a new chat. Incognito chats are never used for training, regardless of your toggle setting, and don't appear in your chat history. Use one any time you have a one-off question that's especially sensitive.
Anthropic's business and enterprise plans never train on your data — that protection is written into the contract rather than depending on a toggle. When your whole team starts using AI day-to-day, moving to one of those plans is worth considering.
What never to paste into any AI tool
Your privacy toggle protects you from your conversations becoming training data. It does not change this rule, which applies to every AI tool, from every company, forever:
- Customer Social Security numbers or full card numbers. Never. There is no question worth it.
- Employee payroll or HR records. Salaries, medical notes, disciplinary files — these stay in your payroll and HR systems.
- Passwords and API keys. (An API key is a secret code that lets software access an account — you'll meet one in Module 5.) A password pasted anywhere outside a password manager is a password you should consider changed.
- Anything under NDA. If you signed a confidentiality agreement about it, don't paste it.
A simple habit that covers all four: before you paste, ask "would I email this to a stranger?" If not, summarize instead of pasting. "A customer owes me $4,200 and is 60 days late" works just as well as pasting their name and account details — and it's the kind of question AI is great at.
Quick check
Three questions to make sure the important parts stuck. Nothing is recorded or sent anywhere — this is just for you.
You're set up
You now have a free Claude account, configured privately, and a clear rule for what never gets pasted anywhere. In the next module you'll have your first real conversation with Claude — and build your first tool before the module is over.