AI is fast and sounds sure of itself. These habits cost seconds and catch most of the expensive errors.
Ask how sure it is
Models give a wrong answer in the same calm tone as a right one. Asking for its confidence, and what would change its mind, shows you where to look harder.
"How sure are you, and why?"
Make it show its sources
Ask for the working and the links, then open them. If a source does not exist or does not say that, you have found the mistake before your client did.
"Show your steps and cite sources."
Keep private data out
Client details, passwords and unreleased work do not belong in a chatbot. A quick check before you paste is the cheapest security you will ever buy.
"Would I be fine seeing this on a billboard?"
Hi, I'm Shaaz.
I build websites and web apps for clients, and AI now touches almost every project I work on. It saves hours, and it has also handed me confident answers that were wrong.
The courses teach what I wish I had known earlier: how to get good work out of AI tools, and how to catch the mistakes and risks before they reach a client. I wrote them for people who need to use AI at work, not study it.
A place where AI answers get checked by people who do the same work as you, so you know what held up in a real project. Details will be posted here.
Questions
Where do I take the courses?
All three courses are on Udemy. Each course card on this page links to its Udemy page.
Which course should I start with?
If you work with clients, start with AI for Freelancers. If your main worry is privacy or scams, start with AI Safety at Work. If you want a daily routine for checking AI answers, start with The 30-Day AI Habit.
Do the courses cover ChatGPT only?
Examples lean on ChatGPT because most people use it, but the habits apply to Claude, Gemini and similar tools.
Who teaches the courses?
Shaaz, a web developer who builds client projects with AI every week and teaches what he learns about using it safely.