How did I get back here?
As a higher education professional, I spend a significant amount of time writing. Generative AI has certainly helped, especially for brainstorming, creating outlines, rearranging my original material, and creating accompanying visuals. I’ll admit, there are areas where I’d like to make greater use of gen AI. But my selective use is not from dissatisfaction or trepidation with its outputs. I lean heavily into its use for topics that I feel comfortable speaking about, because I trust that I can spot its shortfalls. However, as an educational researcher, when I’m exploring something relatively unfamiliar, I turn to reputable, trusted scholars and databases.
I’m not sure how relatable this is, but some of my output slows down in the rewriting and polishing stage. I can see why a sentence or a word does not work, but I sometimes stumble over the fix. That’s why human editors are so valuable in producing quality work. I trust feedback from my colleagues far more more than I would a machine editor. Even as I am writing this, the built-in AI grammar tool is overexuberant with its suggestions for my post, but its suggestions don’t work for what I want to say and how I want to say it. With the advent of generative AI, I have become more forgiving of the flaws in my writing. A hiccupy, stuttering human-generated sentence does not always have to be fixed.
My current position (and this is subject to change) is that if a sentence sounds clunky, then I am going to release the temptation to agonize over how to rearrange it. I am certainly going to try to fix it, but I will resist the temptation to make it as flawless as possible. It may mean sitting with sentences or expressions that do not sound polished. I’m speaking specifically about writing as inquiry (Richardson & St. Pierre, 2005) or writing as discovery. For me, the flow of discovery is stunted when I am caught up in the imperfections of a first draft.
And that’s why I’ve found myself returning to academic blogging in an age of generative AI. There is an abundance of content and output created and shaped by generative AI, and this has made some processes easier for me. Generating images to illustrate my text, I lean heavily into that. But I don’t want my writing as inquiry voice to be cleaned up by gen AI. Just as I love reading work that sounds like there is a person who composed it, with a style and cadence that evokes a living, breathing, imperfect human on the other side, I want to embrace that imperfection in flow and expression, especially in the inquiry stage of a project.
The following is the gen AI rewriting my post to make it pithy:
- In my current position, I’m working in a time where generative AI is becoming more and more common and useful. Gen AI is great for brainstorming and creative exercises, and its outputs are so diverse. It has been very helpful for my writing, especially for brainstorming, creating outlines, rearranging my original material, and creating accompanying visuals. There’s been some trepidation from some people because of the high potential for misuse of this kind of technology, but so far it seems mostly trustworthy and I like the results.
I can say I definitely do not like the results. It’s not what I sought to say. At all. For writing as inquiry, exploring and understanding, a run-on sentence, a sentence fragment, faulty parallelism, and misplaced commas, I’ll make a welcome peace with the imperfections of the first draft outputs and try to obsess less over the clunky mess.
Inspiration
Richardson, L. & St. Pierre, E. A. (2005). Writing: A method of inquiry. In N. K. Denzin, & Y. S. Lincoln (Eds.), The Sage handbook of qualitative research (3rd ed., pp. 959–978). Sage.
Galicia S. T. Blackman 