The One Thing You Should Know Before You Use Any AI Tool at Work
You've probably already used ChatGPT or Claude at least once. You know the basics. Type a question in, get an answer back. Simple enough.
Before you use it for anything work-related, there's one thing that matters more than how the system works or what it can do.
Before you use any AI tool — at work or at home:
Never enter your employer's sensitive information into a free AI tool. This includes customer names or contact information, employee records, financial data, contracts, proprietary processes, or anything you'd consider confidential. Free AI tools are public services. Treat them accordingly. If you're not sure whether something is safe to enter, don't enter it. Ask your manager first.
Free AI tools are public services. This doesn't mean the government runs them. It means you shouldn't type your company's confidential information into them any more than you'd type it into a public forum where strangers could read it.
Here's how it works: when you type text into a free AI tool, that text is processed by servers that store, analyze, and use the information to improve the system. In theory — and depending on the tool — what you type could inform how the system responds to other users. Your company's information could resurface in responses to people outside your company.
This isn't hypothetical. It's happened. Employees have put sensitive data into ChatGPT, and that data has shown up in other people's conversations.
Here are the categories of information you should never enter into a free AI tool:
Customer names, contact information, or account data. Employee records, performance notes, or anything from HR. Financial figures, pricing, or account balances. Contracts, NDAs, or any document marked confidential. Proprietary processes, formulas, or specifications your company doesn't publish publicly. Anything you would put in a folder labeled "confidential."
Use this test: would I feel nervous if a competitor's employee saw this? Would I feel uncomfortable if my manager knew I'd shared it? If the answer is yes, don't type it in.
Think of a public photocopier at a library. You wouldn't leave sensitive documents sitting there while it processes your copies. This is the same instinct. It's not paranoia. It's protecting your company and protecting yourself.
What if you accidentally put something sensitive into an AI tool? Here's what you do: tell your manager immediately. Don't sit on it. Don't assume it was harmless. The sooner someone knows about it, the sooner the company can assess whether anything actually needs to happen. If you wait, and it becomes a bigger issue later, the explanation becomes much harder.
Most of the time, one piece of information going into an AI tool isn't a crisis. But the rule isn't "don't put sensitive stuff in unless it's just a small amount." The rule is don't put it in. The reason is that you can't predict how one piece of information might connect to another piece when aggregated with other data, or what someone on the other side could infer.
Better to be wrong about caution than wrong about judgment.
If you're not sure whether something is safe to enter, don't. Ask your manager first. That's not weakness. That's the right call. Your manager's job is partly to make these judgment calls. Let them do it.
Your company probably has an AI policy that spells this out in detail. If you haven't read it yet, read it. It's not punishment-based guidance. It's protection for you and the company. The clearer you are about what's allowed, the less guessing you have to do when you're in a gray area.
Stay current without the overwhelm.
New technology gets confusing fast, especially when the people talking about it use different vocabularies than the people who actually do the work. We write a weekly newsletter specifically for people in jobs like yours — no fluff, no oversell, just what's changing and what it means for you. You don't have to become a tech person. You just have to know enough to stay relevant.