Hi my friend 👋
I’ve been thinking about this one thing lately and finally decided to bring more awareness to it.
It’s already been so long since we have access to AI tools and I still keep seeing one major mistake that a ton of people still make with it. And the worst part is that it’s affecting their reputation, without them even realizing it.
Do you know what that is?
It’s that people are letting AI think for them instead of using it to organize their existing thoughts. And it shows in their writing whether it’s a LinkedIn post, an article, or even an email.
When I read obviously AI-written content, it kinda gives me an ick. Because it reveals they haven’t made real thinking behind what they’re trying to communicate AND most importantly, don’t value the reader enough to be useful and concise.
For example, when someone asks ChatGPT or Claude to “write a LinkedIn post about cold outreach in sales,” they end up with content that they themselves don’t relate to enough. Plus, they’re unable to offer real suggestions and guidance with their content, because the writing process itself is what helps you understand the issue deeply enough to solve it. Phew, that was a lot!
So, first, let me show you the example of bad writing I see every day and then I’ll break down a better way to use AI.
How I spot bad AI writing (& why you shouldn’t let it slip)
Bad AI writing always follows predictable patterns:
Phrases like “in today’s rapidly changing landscape,” “as we navigate the complexities of,” “synergies,” “But then, something shifted,” “And honestly? That's ...,” “This is the part most people miss,” “Not because X, but because Y”
Rule of threes like “It's not. X. It's not. Y. It's Z,” “No technical setup. No complicated workflows. Just prompts you can copy, tweak, and start using in minutes.”
Endings like “at the end of the day, it doesn’t come down to this or that, it comes down to…”
Generic, fluffy language instead of clear, direct communication
If I see any of these, my brain just scrolls through the post, no matter who the author is.
And I’m not alone. There are so many people doing the same and the author of these obviously AI-written posts slowly loses credibility.
So please, while I advocate for using AI to take some tasks off your plate, don’t just copy-paste AI outputs and publish them.
My AI usage framework
The right approach is to do the thinking first, then use AI as an editor and organizer:
Step 1. Think through problems yourself first
Don’t outsource the thinking. Always think what you want to communicate, why, and how. All by yourself.
That’s how I approach it personally too. For example, when I’m writing this newsletter, I always make my own decision on what topic I want to cover and what I want to say.
Step 2. Use voice-to-text tools like Wispr Flow
After I have a rough idea in my head, I find it easier to ramble about it and not worry about the structure or how to write it.
For that, I use Wispr and talk as much as I have to. Sometimes it’s a few minutes, other times, it can go up to half an hour or more.
My suggestion is to speak your thoughts naturally and don’t worry about the fluff or mistakes. Wispr polishes all of that and turns it into text, while keeping your manner of speaking.
Once you’re done, go through the text and format it.
Note: I use and recommend Wispr, but you can also use a voice memo feature in any app if that’s your preference.
Step 3. Use AI as an editor
I even highly encourage you to take the text you’ve generated with a voice memo, paste it in Claude or ChatGPT, and ask it to identify unclear language, logical gaps, potential reader questions, or repetitive content.
I usually ask AI helpful editing questions, too:
What questions would someone still have after reading this?
Where is my language unclear?
What fears might this provoke in readers?
Do I have logical arguments/examples?
Where am I being repetitive?
Where can I add more value so the reader leaves the page with all the answers/solutions on this topic?
Step 4. Make your own decisions
And finally, I know, we all love turning to AI when we’re not sure about something in our thinking process.
But it’s important to always remember that AI gives suggestions, and you’re the one who should make decisions on what to keep or change.
Different AI tools will often give opposite advice on the same exact question, too. So, it’s just something to challenge your thinking, and not take it blindly.
Your value isn’t producing fluffy fake words faster. As someone who writes content (for any purpose at all), your value is your ability to think, make judgments, and communicate in your authentic voice using your brain and life experience. AI should only enhance that thinking, not replace it.
If you’re interested in starting a newsletter like this, try out beehiiv (that’s what I use).
See you next week,
Kate 🌟
