Lesson 3: Checking AI's Work

What You'll Learn

Simple ways to verify what an AI tells you
When it is fine to trust AI and when you should double-check
Why you should never paste sensitive information, and why you stay responsible

You Are the Editor

In the last lesson you saw that AI can be confidently wrong. The good news is that you do not need to be a tech expert to use it safely. You just need a few sensible habits. The mindset to hold onto is this: the AI is your assistant, but you are the editor. It can produce a fast first draft or a quick answer, but you decide what is actually good enough to use.

Verify Facts with Trusted Sources

Whenever an AI gives you a fact that matters, especially a name, number, date, statistic, or web link, check it against a trusted source. A trusted source is a place known for being reliable, such as an official website, a well-known news outlet, an encyclopedia, or an expert you know.

A quick way to do this is a simple web search. If the AI says a museum is open until 8 p.m., look at the museum's own website before you build your evening around it. If it offers a surprising statistic, see whether a reputable source backs it up. This takes seconds and saves you from passing along something false.

Cross-Check the Important Things

For anything with real consequences, do not rely on a single answer. Cross-checking means confirming the same fact in more than one place. This matters most for:

Health and medicine (talk to a doctor or pharmacist, not just an AI)
Money, taxes, and legal questions (confirm with a qualified professional)
Safety instructions (for example, how to use a tool or handle a chemical)
Anything you will publish or send widely that could embarrass you if it were wrong

For low-stakes tasks, like brainstorming gift ideas or rewording a casual message, a quick read-through is usually enough. The bigger the consequences, the more checking it deserves.

Never Paste Sensitive Information

This one is simple and important: do not paste private or sensitive information into AI tools. That includes passwords, bank or credit card numbers, government ID numbers, medical records, and other people's personal details. Depending on the tool and its settings, what you type may be stored or reviewed to improve the service. A safe rule of thumb: if you would not write it on a postcard, do not paste it into an AI tool. When in doubt, leave it out or replace it with a placeholder like "[NAME]."

When to Trust vs. Double-Check

Here is an easy way to decide.

Usually fine to trust: creative help, brainstorming, rough drafts, summaries you will read over yourself, casual rewrites, explanations of common ideas you can sanity-check.
Always double-check: specific facts and figures, anything about health, money, law, or safety, current or recent events, and anything you will share publicly or act on in a meaningful way.

If a wrong answer would cost you time, money, embarrassment, or safety, slow down and verify.

You Are Responsible

Last, the most important point. If you use something the AI wrote, you are responsible for it, not the AI. If you send a report with a made-up statistic, the mistake is yours to own. That is not a reason to avoid AI. It is a reason to treat its output as a helpful draft you review, edit, and stand behind, the same way you would check work from a brand-new assistant before putting your name on it.

Key Takeaways

Verify facts (names, numbers, dates, links) against trusted sources, and cross-check high-stakes topics like health, money, and safety.
Never paste passwords, financial details, or other people's private information into AI tools.
You are responsible for anything you use, so review and edit AI output before you rely on it.