I didn't start out as an AI enthusiast. I started out nervous, then curious, then frustrated, and only later genuinely productive. Today AI is part of how I code, design websites, build a business and run my week. I now even run several AI agents at the same time, each handling part of a task, so that together they produce a better result than any one of them alone.
I'm sharing the journey because most people I speak to are stuck at one of the early stages. If that's you, the good news is that every stage is temporary.
Stage 1: Afraid to touch it
My first reaction to AI was hesitation. It felt powerful and unfamiliar, and I worried that one wrong click or one badly worded question could break something, expose something, or make me look foolish. So I kept it to small, safe jobs: rewriting emails and drafting the odd post.
After more than 20 years in sales and retail, I was used to knowing my tools inside out. This was a tool I didn't understand, and that made me cautious. So I watched other people use it far longer than I used it myself.
Looking back, the fear was out of proportion. You can't break a chat window by asking a question. The real risks are about what you share and what you trust, and both are easy to manage once you know where they are.
And here's what surprised me most: AI doesn't judge you. No question is too basic, and you can ask the same thing five times without anyone rolling their eyes.
Stage 2: Try and forget
Once the fear wore off, I started experimenting. I'd type a prompt, read the answer, and either get something that missed the point or something that was close but needed more work.
Either way, I parked it. I told myself I'd come back to it, and I almost never did. I had dozens of half-finished conversations and no real results to show for them.
The problem wasn't the tool. I was treating it like a search engine: one question, one answer, done. I hadn't yet learned that the first answer is a starting point, and that the value comes from the second, third and fourth exchange.
Stage 3: AI as my teacher
The turning point came when I enrolled in a data analysis certificate. For the first time, I had a clear goal, a real deadline and concepts I genuinely needed to understand.
AI became my tutor. When a formula or a concept didn't click, I asked it to explain it differently, give me an example, or walk me through where my thinking went wrong. It never got tired of the same question asked three ways.
I also learned to use it as a teacher, not a shortcut. Asking it to do the work for me would have taught me nothing. Asking it to explain, check my reasoning and quiz me is what made the learning stick.
Learning to code was a good example. It was rewarding but repetitive, and the same small errors kept tripping me up. Having "someone" check my work and explain my mistakes as I went made all the difference. It felt less like being marked and more like being supported.
Stage 4: The next level
Once I trusted the process, the way I used AI changed completely. It stopped being something I occasionally asked questions of and became a working partner across four areas.
Coding. I'm not a developer, yet I now build working tools with AI writing the code. I describe what I want, test what comes back, explain what's broken, and repeat until it works.
Web design. My own website was built this way, from the first draft to the bilingual pages and booking flow. That experience is now one of the services I offer: websites built with AI for small businesses.
Business builder. As I set up my consulting practice, AI helped me shape the offer, pressure-test pricing, draft service pages and think through branding. It doesn't make the decisions, but it helps me reach better ones faster.
Personal assistant. Day to day, it drafts emails, tailors my documents, prepares me for meetings and helps me plan the week. Small tasks, but they add up to hours saved.
My top 5 lessons
- You won't break it, so start. The fear of doing something wrong costs more than any mistake you'll make. Start with low-stakes tasks, and keep confidential client or company data out until you understand your tools and your employer's policy.
- The first answer is a draft, not the result. Most of my early disappointments came from stopping too soon. Reply, correct, add context, ask again. The quality is in the conversation, not the first prompt.
- Give it context like you would a new colleague. Who you are, who it's for, what good looks like, what to avoid. The more it knows about the situation, the less generic the output.
- Use it to learn, not just to do. Asking "explain why" builds skills you keep. Asking "just do it" builds dependence. The best results come when you understand enough to judge what it gives you.
- You stay accountable. AI can be confidently wrong. Check facts, test the code, read every email before it goes. And don't be shy about asking it to double-check, or even triple-check, its own work. It's a powerful partner, but the judgment and the name on the work are yours.
Where to start
If you're still at the fearful stage, or sitting on a pile of abandoned prompts, you're in good company. Pick one real task from your week, give the AI proper context, and stay with it for five or six exchanges instead of one.
That shift, from asking once to working it through, is what moved me from fear to everyday use. It's also why I now coach individuals and teams to find the practical use cases that fit their own work.
If you'd like help getting started, book an AI coaching session or get in touch.