- Working With (Not Against) the Machine: How I Use AI Without Losing My Voice - August 28, 2025
- Soundness, Intensity, and the Choice to Resist: Writing with Agency in the Age of AI - August 21, 2025
- Exploring Creativity with AI - August 7, 2025
One of the most common concerns I hear is, “Aren’t you worried AI will replace writers?” My honest answer: not really. But I do think it’s changing how we write—and how we need to think about creativity.
It’s not just about getting the job done faster. For me, it’s about making sure the message still sounds like me, especially when so much of what we read now is shaped by a tool. In my experience, the key isn’t to hand over your voice to a machine, but to learn how to guide it. Working with AI is less like outsourcing and more like managing a very fast (and sometimes clueless) collaborator.
Early on, I found that the quality of what I got from AI depended almost entirely on what I put in. When I started using it to draft social media captions for events, I’d usually feed it basic details: the what, when, where, and a general sense of the event’s objective. I’d then ask it to “generate captions for Facebook.” What came back were standard announcements—generic, flat, and often too similar to things I’d never post myself.
I wasn’t satisfied with those results, so I started refining my prompts. I began specifying who the audience was, what organization I was writing for, and the kind of tone I wanted—formal, informal, casual, or fun. The more specific I was, the better the results sounded—not just accurate, but actually like something I’d want to say.
But there were still moments when the language made me cringe. Some outputs sounded too casual or included phrasing I’d never use—those enthusiasm-heavy phrases like “excited to share” or “thrilled to announce” that sound like they came from a corporate template. I’ve always believed in using simple, clear language, but there’s a line. Without clear direction, the tool sometimes crossed it, and I realized I needed to take more control over how it approached voice and tone.
Eventually, I started building my own approach by feeding past samples of my work—articles, posts, and reports that already carried my tone—directly into my prompts as examples. This gave the tool a clearer sense of how I write, but more importantly, it reminded me of what I sound like. Over time, I also became more sensitive to overly formulaic phrasing. Now, I know what to look for, and I know exactly what to change.
Most of the time, that means scanning for overused words and swapping them out with language that feels more natural—more me. I might keep a structure or an idea the AI suggested, especially if it helps clarify my message. But the way it’s said? That always gets rewritten. Because no matter how efficient the tool is, the final version still has to carry my voice.
I know a piece of writing is truly mine if it conveys my message clearly to readers—whether it’s about women’s empowerment, freelance workers’ rights, or how we protect our creative process in the age of AI. Even when I use these tools to shape a first draft or structure my ideas, I remain guided by the purpose behind the work. I still ask: Is this meaningful? Is this relevant? That’s the part no machine can replace—and that’s the part I hold onto.
If you’re a writer or creative like me, ask yourself: how do you use AI tools right now? Are they just a glorified search engine, a quick assistant, or something closer to a collaborator? The question isn’t whether AI will change how we write—it already has. The question is whether you’ll let it write for you, or whether you’ll use it to write better as yourself.