Am I Using ChatGPT Too Much?

I’ve been asking myself this question quietly over the last year, not because anything went wrong, but because something started to feel off. I would reread an article or a post and recognise that it was fine, even good, but it didn’t always feel like it had cost me anything to write. That ease is seductive, and I don’t think we talk enough about what it trades away.

There’s no question that tools like ChatGPT have changed how we create. Writing is faster, friction is lower, and getting something passable onto the page takes far less effort than it used to. For a while, that felt like genuine progress. But alongside that, something else started creeping in. A sameness. A flattening. What people now casually call AI slop. Content that reads well enough but doesn’t really say anything, or worse, says something without having had to think very hard about it.

That’s where my discomfort began, not with the tool itself, but with my reliance on it. I noticed how quickly I would reach for AI to help me write, sometimes before I had really sat with what I wanted to say. It wasn’t that I couldn’t write. It was that it was easier not to wrestle with my own thinking. Over time, that ease started to feel like a quiet erosion of judgement rather than support for it.

I don’t think this means AI is making us worse by default, but I do think it can make us lazier if we’re not paying attention. These systems are very good at producing language. They are not good at responsibility, context, or consequence. Those still belong to us, whether we notice when we’re giving them up or not.

At a certain point, I made a deliberate change. I stopped using AI to write articles for me. Not because it was wrong, and not because I wanted to prove a point, but because I wanted the thinking back. I wanted the slowness. I wanted the discomfort of working out what I actually believed before asking for help shaping it. The writing became messier again, but it also became more recognisably mine.

What surprised me is that I didn’t stop using AI. I just moved it. Instead of putting it at the centre of my creative work, I pushed it out toward automation. Scheduling, sorting, repeatable processes, follow through. Work that doesn’t require my values or my voice, but does take time and attention away from them. That shift gave me more space, not less. More energy for writing, not less reliance on tools.

I don’t think we’re losing our way because of AI. I think we risk losing our way when convenience quietly replaces judgement and speed replaces sense making. Used well, these tools can support better work. Used carelessly, they can hollow it out.

For me, the question has stopped being whether AI can help, and started being where it should help. Writing is something I want to stay close to. Automation is something I’m very happy to hand over. That distinction has made all the difference.

Once we sign this article off, the next part of this conversation is more practical. What automation actually does well, and how to use AI to reduce load without giving up the thinking that matters. That’s the direction I’m heading next.


Next
Next

When Policy Exists but Safety Doesn’t