Six months ago, I almost lost a client because I spent three days writing a market research report that my competitor handed in within a few hours. Same quality. Maybe even a little better, honestly. And the thing that stung the most? They weren’t smarter than me. They just weren’t being stubborn about AI.
That was my wake-up call. And if you’re still on the fence about whether to actually use AI tools in your daily work — not just toy around with them, but really build them into how you operate — I want to share what changed for me. Because the shift wasn’t overnight, and it wasn’t without a few embarrassing mistakes along the way.
“I wasn’t afraid of AI taking my job. I was afraid it would make me look lazy. Turns out, the people who looked lazy were the ones refusing to adapt.”
1- The mindset trap I was stuck in
For a long time, I treated AI like a cheat code. Something you used when you were desperate or didn’t care about quality. I told myself that real writers write, real coders code, real marketers think. AI was for people who didn’t want to do the work.
That’s a comfortable story. It’s also completely wrong.
The better analogy, I realized, is a calculator. Nobody says a physicist is “cheating” by using one. Nobody says an architect lacks skill because they use CAD software. The tool doesn’t replace the thinking — it removes the friction around it. And once I saw AI that way, everything changed.
What “working with AI” actually looks like day-to-day
I want to be specific here, because vague advice like “use AI for productivity” is basically useless. Here’s what my actual workflow looks like now:
Morning brain dump
I open Claude and paste in my messy, stream-of-consciousness notes from the day before. I ask it to help me find the three most important tasks buried in there. It doesn’t decide for me — I do. But it surfaces structure I couldn’t see when I was too close to the chaos.
First-draft anything
Whether it’s an email to a difficult client, a project proposal, or a blog post outline, I don’t start from a blank page anymore. I give AI a rough direction and ask for a starting point. Then I rewrite it heavily. My voice comes through because I’m editing, not reading. The blank page paralysis? Gone.
Research as a conversation
Instead of opening fifteen browser tabs and losing the thread, I use AI to have a back-and-forth dialogue about a topic. I follow up, push back, ask for sources to verify. It’s faster, and weirdly more engaging than scrolling through search results.
The tools I actually use (not just the famous ones)
People always ask what tools I recommend. Here’s my honest stack:
Claude (claude.ai), Notion, AI Perplexity, Otter.ai, Gamma (presentations)Runway (video)
Claude is my main thinking partner — it handles longer, more nuanced tasks and doesn’t lose the plot halfway through a complex request. Notion AI is where I do most of my writing and note-taking, and having AI right inside those pages removes a lot of context-switching. Perplexity is my go-to when I need sourced, current information fast. Otter handles meeting transcription so I’m not scrambling to write notes while also trying to actually listen. Gamma saves me from the death-by-PowerPoint grind. And Runway is genuinely wild for anyone doing video content — I use it for quick edits I used to outsource.
The point isn’t that you need all of these. Pick one and actually use it for two weeks. That’s it.
2- How to actually start — a simple approach that works
Most people overcomplicate this. Here’s what I’d tell someone starting from zero:
- Pick one recurring task you find annoying
Not the most important task. The one you dread. For me it was writing weekly status updates. Tedious, repetitive, low-stakes enough to experiment on.
- Give AI more context than you think it needs Don’t just say “write a status update.” Tell it who your audience is, what the tone should be, what happened this week, what you’re worried about. The quality of output scales directly with the quality of your input.
- Edit aggressively: Don’t send the first draft. Never. AI outputs are starting points, not finished products. Your job is still to think — AI just does the scaffolding.
- Notice what it got wrongThis is how you get better at prompting. Every time AI misses the mark, ask yourself: what context was missing? What did I assume it knew? Over time you get a feel for how to communicate with these tools.
- Gradually expand to harder tasksOnce you trust the tool in low-stakes scenarios, start applying it to bigger work. That’s when the real time savings kick in.
2. Mistakes I made that you can skip
- Treating every output as final. I sent an AI-written email to a client once without reading it properly. It was technically fine but tonally off — too formal for our relationship. Client noticed. Awkward.
- Using it for things I should think through myself. AI is great at producing answers. It’s not great as a substitute for sitting with a hard problem. Strategic decisions still need your own head. Use AI for the execution, not the judgment.
- Switching tools too often. Every few weeks there’s a new “best AI tool.” I wasted months hopping between them. Pick something solid, go deep on it, then expand. Breadth before depth is a trap.
- Not fact-checking. AI is confident even when it’s wrong. I once used a statistic from an AI response in a client presentation. The stat was plausible but I couldn’t verify it. That was a bad five minutes when someone asked for the source.
3- The unexpected thing that happened
Here’s the part I didn’t see coming: using AI made me better at the actual skills I was worried it would replace.
When you’re editing AI output instead of writing from scratch, you develop a sharper eye for what’s generic versus what’s genuinely yours. I became a better writer because I spent more time in editing mode, cutting the fluff, finding the voice. The same happened with my thinking on strategy — because I was using AI for the legwork, I had more mental energy for the parts that actually required judgment.
I also started to understand my own work better. When you have to explain a task to an AI clearly enough for it to help, you often realize you didn’t fully understand the task yourself. That forced clarity has real value.
4- The honest reality check
Worth knowing before you dive in
- AI tools cost money. Most worth using have paid tiers. Budget for it like you would any professional tool.
- There’s a learning curve. The first two weeks feel awkward. That’s normal. Keep going.
- AI makes confident mistakes. Your critical eye is still essential — maybe more so.
- Privacy matters. Don’t paste sensitive client data or confidential info into general AI tools without checking their data policies first.
None of that should stop you. But going in with realistic expectations means you won’t give up when it’s not magic on day one.
The people I’ve watched thrive over the past year aren’t necessarily the most technically skilled. They’re the ones who got curious instead of defensive. Who asked “how can this help me?” before asking “should I be worried about this?”
You don’t need to overhaul your whole workflow this week. Just pick that one annoying task. Give an AI tool a real try on it — not a five-minute test, but a genuine two-week experiment. See what happens. I’d bet you don’t go back.
And if you’re already using AI in your work and have a tip or a tool I missed — drop it in the comments. I’m always looking to improve the stack.