The Green Ledger - Tips for a Sustainable Small Business
(The Third Episode in Our 3-Part AI Series) Over the past two episodes, we've talked about whether AI is worth using and where it creates the most value in a small business. Now it's time to talk about something equally important: How do you use AI without accidentally exposing your most valuable business information? For most small businesses, the biggest AI risk isn't hackers. It's well-intentioned employees trying to work faster. This episode explains the simple safeguards every business should have before using AI with recipes, pricing, customer information, financial data, or proprietary processes. ✔️ What You'll Learn * Why the biggest AI risk usually comes from inside your business * The three types of business information you should never expose * The surprising differences between free, personal, and business AI accounts * Why large companies create AI usage policies—and why small businesses should too ✔️ How AI Tools Use Your Information (What to Check Before You Paste): A quick, plain-English guide to the thing most owners don't realize they're deciding. This isn't legal advice, and these policies change often - treat it as a starting point and check the current settings. The mental model: there are three tiers, not two Most people assume "free = exposed, paid = private." That's the trap. There are really three tiers, and the middle one surprises people. * Free tools - by default, what you type can be used to help improve (train) the AI. You can usually turn this off in settings * Paid individual plans (the "Pro" or "Plus" version) - here's the catch: paying the ~$20/month personal plan usually buys you better features, not more privacy. On most of these, your chats are still used for training by default unless you go into settings and switch it off. * Business / Team / Enterprise plans - these generally do not train on your inputs by default. This is the tier built for company data. People blur two concerns together. One: "they'll use it to train the AI" - that's the one you control, with a setting or a business tier. Two: "a company can process or store it at all" - that's true of almost any cloud tool you already use: your email, your bank's app, your cloud drive. The fixable one is the first - so fix it. What to know about Settings (current as of mid-2026 - verify, as these change) * ChatGPT: Settings → Data Controls → turn off "Improve the model for everyone." For a sensitive one-off, a Temporary Chat isn't used for training. * Claude: Settings → Privacy → turn off the model-improvement toggle. Leaving it off also shortens how long your data is kept (a 30-day window instead of much longer). * Gemini: turn off "Gemini Apps Activity" - just know that doing so also means you won't keep a chat history. Used through a Google Workspace business account, your data is treated as private and not used for training. Two caveats: •Turning training off is not the same as total privacy - the company still processes your request to answer it, and some limited retention remains. •And anything already used to train a model can't be pulled back out later. Which is exactly why the safest move is simply not putting your crown jewels in. The point of the one-page policy from this episode: decide these things once, write down two or three rules, and your team never has to guess. 🛠️ Free Resource 📥 The One-Page AI Policy For a Small Business 📩 Got a Question? Have a resilience, supplier, operations, or sustainability question you want covered on the podcast? Send it in - your question could become a future episode and help other small business owners facing the same challenge. 🔗 Resources & Contact 🌐 www.3pimpactconsulting.com [https://www.3pimpactconsulting.com/] 📩 anca@3pimpactconsulting.com Send us Fan Mail [https://www.buzzsprout.com/2413411/fan_mail/new]
19 episoder
Kommentarer
0Vær den første til at kommentere
Tilmeld dig nu og bliv en del af The Green Ledger - Tips for a Sustainable Small Business-fællesskabet!