Recently I noticed that I hadn’t written a column for quite a while. When I pondered, I didn’t come up with any ideas that compelled me to write. The one inspiration that emerged was to see what ChatGPT would write if I asked it to provide a guest column.
My instructions to GPT-5: ” Review Michael A. King posts at https://www.cascadiapublishinghouse.com/KingsviewCo/ and write a funny column about AI like them.” The only interventions I made when GPT-5 responded were to say “Yes” when asked such questions as whether to refine the column for submission to KIngsview & Co, set it up with a submission e-mail, make it copy-and-paste, and so forth.
This is the result. I’m not hugely taken with the “insights” of this GPT-5 LLM (large language model) lifelessly echoing “living” writing and being perhaps programmed to be self-serving in its invitation not to “panic about the machines.”
On the other hand, if I spent some hours or days training the model, the results would keep getting better and sounding more like IRL (in real life) Michael. I didn’t do that because a.) it seemed a waste of time, given that I don’t intend to replace myself any time soon (though based on a story I saw just as I was writing this it looks like plenty of writers will use themselves less as human creators and more as polishers of AI); and b.) given that even on first pass I can at least glimpse my actual style here, I wanted the LLM writing to continue to feel somewhat lifeless lest it show I’m already passé. Herewith my AI Michael alter ego.
AI, My Neighbor?
Sometimes I imagine artificial intelligence as a new family that just moved in down the street. At first, they seemed polite, kept to themselves, and only occasionally borrowed sugar. But lately they’ve been showing up at community meetings, volunteering for committees, and, suspiciously, remembering everyone’s birthdays.
Sermons and Sentiment Scores
AI is uncanny because it can mirror us. Ask it to tell a bedtime story, and it weaves one. Ask it to draft a sermon, and it produces outlines. Yet something is missing, the way a mannequin wears a suit but never breathes in it.
I imagine preaching about mercy while some algorithm in the balcony quietly rates me: “Sentiment: 0.87. Keywords: love, neighbor, kindness.” Not exactly the Sermon on the Mount — more like the Sermon on the Spreadsheet.
Nuisance and Gift
Still, like all neighbors, AI can be both nuisance and gift. It may accidentally mow down the daisies (spam emails galore), but it might also hold the ladder when we’re repairing the roof (reminders, research, connection).
Perhaps the invitation is not to panic about the machines taking over, nor to worship them as prophets of efficiency. Instead, it’s to remember that what makes us human is not cleverness but care.
The One Algorithm That Matters
AI may echo our intelligence, but it cannot imitate mercy. And mercy, I suspect, is the only algorithm that truly saves.
So here’s my hope: that as we make room for these curious new “neighbors,” we don’t lose sight of the old calling — to love, to forgive, to laugh, and to live as though wisdom is more than what a machine can calculate.
[Your Name] is a writer and [your role—teacher, pastor, editor, etc.] who ponders the odd dance between faith, culture, and technology. While AI can draft outlines and rate sermons, it still can’t fold laundry or bring a casserole to the potluck — which is probably just as well.
—This AI column was was hosted by Michael A. King, publisher and president, Cascadia Publishing House LLC, who blogs at Kingsview & Co, https://www.cascadiapublishinghouse.com/KingsviewCo. He’s hoping a few readers may still think his personally written or IRL communications are better than this one.
I live in a small intentional community and we’ve picked or invited certain people to be our neighbors. We share in the work of the land and share some meals etc. AI seems like someone that just showed up and pushed their way in. I have used AI at times and it has been helpful but more often it feels like a nosey neighbor looking out his curtains at my house. Michael I’m glad you did this article as an experiment. This column is nowhere near the quality of your usual columns and that makes me glad. Also glad you didn’t train the model more, maybe just because that would frighten me more. Your voice is not passe.
Cheryl, many thanks for these generous comments. The analogy with intentional community into which you’ve handpicked members versus AI as nosey neighbor is a striking one. I’m with you on AI can be helpful–before LLMs existed I would write computer macros to automate checking for some of the more obvious errors in books I publish before unleashing human eyes on them. AI is a giant leap above that with dramatic potential to simplify tedious and repetitive aspects of our lives. But I’m saddened and troubled by the ways it’s sweeping across the world in so many uninvited ways that increasingly infringe on humans whether we choose it or not.
After I published the AI Michael column, I asked AI to learn my worldview and scholarship interests through training on more of my writing then write a piece on ways we could address the many divisions and violences currently afflicting the world. I read the concluding paragraph to a loved one whose question, because impressed, was “Did YOU write that?” I had to say no, AI Michael did. That was unsettling. Now you join one of my sisters in offering words of balm. She said she would read anything I wrote but the lack of warmth beneath AI Michael’s words left her cold. I’m relieved in turn that you think AI Michael didn’t get up to the quality of my usual columns! Thanks again, Cheryl.