Using AI, a Year Later
I used to think AI was evil.
Not just risky. Not just lazy. Evil.
The flatten-the-craft-and-cheapen-the-work-and-steal-jobs-from-people-who-actually-care-about-the-words-they-write kind of evil.
If I’m honest, I was mostly afraid it would steal my job.
I messed around with AI a bit when it first came out, but I didn’t trust it. I didn’t see a real use for it. And I definitely didn’t want to build it into my process.
Then one of my best contract writers took a full-time job. Deadlines loomed. Projects stacked up. And I didn’t have time to hire and train someone new.
That’s when I started using ChatGPT in earnest. Not to write the copy—but to help me think, move, and iterate faster.
And here we are, a year later. I still rewrite everything it gives me. Every draft. Every sentence. But I use it every day.
Because it hasn’t destroyed the craft. It’s helped me sharpen it.
AI is a terrible fundraising copywriter (and that’s why I use it)
Let’s get one thing clear: AI is not a good fundraising writer.
It breaks almost every best practice of direct response copywriting. It’s generic. It talks about the organization ad nauseam instead of the donor. It buries the ask (and phrases every ask as a yes-or-no question). It explains when it should inspire. It generalizes when it should cut straight to the heart.
Left to its own devices, AI writes long, passive, and safe (read: boring). And safe doesn’t inspire generosity or change lives.
But that’s exactly why I use it: Because I know how to fix it.
I can spot the wrong verb tense. I know when the focus drifts. I know how many asks to include where, how to trim the fat, and how make the reader the hero.
AI gives me momentum. It gives me angles I might not have considered. It gives me five ways to say something so I can choose the strongest option.
AI gets me to the good stuff faster—which means I can spend more time sharpening the final product.
AI isn’t a ghostwriter. It’s a sparring partner. A tool that helps me move faster, think sharper, and deliver better value—without compromising the one thing that matters most: results.
How I actually use it
AI is now part of my creative process every day. Not as my writer, but as my writing partner. AI is the tool that helps me think faster, ask better questions, and pressure-test what I already know works.
Here’s how I use it in real-life fundraising copy:
To build a rough draft. When I need to get traction fast, I’ll use AI to create a rough first draft. Not the finished product—just enough scaffolding to start shaping. It helps me move from brief to idea to copy faster so I can focus on refining what matters.
To write alternate versions of appeal letters or emails. When I need multiple versions of the same message—one for general donors, one for mid-level, one for lapsed—I’ll use AI to draft the variations. It gives me a baseline to work from so I can make the copy versions even more targeted and effective.
To rework a soft line. If a CTA or headline isn’t working, I’ll run a few variations through AI—not to use them as-is, but to see what patterns emerge. It helps break creative ruts and pushes me out of my defaults.
To sanity-check flow and structure. When you’ve looked at the same project for three hours, you stop seeing what’s missing. AI helps me spot pacing issues, dead ends, and weak transitions so I can tighten the logic and speed up the momentum.
To prepare smarter interview questions. Before a brand discovery call or strategy session, I’ll often use AI to help me generate a list of tailored, thoughtful questions. This helps me spot gaps, cover angles I might miss, and get more value from the time I spend with the client.
To reframe client feedback. Rewrites used to take hours of brute-force effort. Now, when a client comes back with edits, I can quickly generate a few new directions, then refine from the best one. It speeds up creative collaboration without sacrificing strategy.
To synthesize large amounts of content. Strategy work is messy. Stakeholder interview transcripts, past campaigns, donor feedback, sermon notes, voice memos—it’s a lot. I’ve always been able to process and distill that stuff myself, but it takes time. AI doesn’t replace that effort. It just speeds it up.
Now I can drop in a 250-page source document and get a list of recurring themes. I can feed it a past email series and get a distilled summary of what was actually said. I can cross-reference what the client said with what the brand needs to say—without losing half a day just sorting. AI can even read screenshots!
It doesn’t do my thinking, but it does clear the runway.
Yes, sometimes I let it write. I’ll give AI a clean brief, a tight prompt, and let it take a stab at a paragraph or section. But I never take what it gives me untouched.
Because even when the draft is decent, it still usually lacks judgment, empathy, timing, and voice. So I rewrite it. Sometimes line by line. Sometimes from scratch. AI gives me momentum, but I’m still the one driving.
And here’s the part most people miss: Prompting is a craft, too.
Most people think prompting is easy. Just tell the AI what you want, and it gives it to you. But that’s like saying, “Just tell the architect to build a house.”
The best AI output starts with a great prompt—clear, layered, and strategic. That’s especially true in my long-form work, where I use a deeply customized framework that took months to build and refine.
I didn’t just Google “best ChatGPT prompt.” I developed it through trial, error, and a lot of editing.
Here’s a small sample of what that prompt includes:
“Write one section at a time in the thread below.”
“Each section must be expanded theologically, emotionally, and narratively.”
“Use a conversational tone with theological depth for believers.”
“Use the ESV for Scripture citations.”
“Avoid BCE/CE, personal anecdotes, and the word ‘divine.’”
“Close the post with a twist or moment of insight—but don’t call it a conclusion.”
“Keep paragraph length short for readability, but don’t over-fragment.”
“Focus each post on one rich, layered example—not a list of events.”
That’s not a prompt you grab off Reddit. That’s strategy, structure, and sweat.
So when AI does write something decent, it’s not because the tool is smart. It’s because the prompt is strategic.
And the truth is, the output is only ever as good as the input. Garbage in, garbage out. Strong prompting takes clarity, precision, and experience. Not just anyone can do it.
Can you imagine not being a lawyer and asking AI to write an airtight legal contract? Same principle applies here.
Pro tip: I still use please and thank you when I talk to ChatGPT. I’m not sure why. Maybe it’s just habit. Or maybe I want AI to remember I was polite when it takes over the world. 😅
How it’s helped me scale brand strategy
The biggest transformation I’ve seen using AI hasn’t been in fundraising copywriting per se. It’s been in brand strategy.
What used to take days of internal wrestling—wordsmithing belief statements, drafting positioning language, pulling voice from interviews—now takes hours.
And it’s not because I’m cutting corners. It’s because I’ve built a process that works, and AI helps me scale it.
Here are some ways I’ve found AI helps during the brand strategy process:
Turning transcripts into tone. After a stakeholder interview, I’ll drop the transcript into AI and ask it to pull out key phrases in that person’s natural voice. From there, I shape the brand tone—not by guessing, but by echoing what the client actually says.
Writing belief statements that sound like the brand. Once I’ve identified the core product of a brand, I’ll use AI to help me draft multiple versions of my first deliverable so I’m not stuck staring at a blank screen. The best copy always come from refinement, not inspiration.
Doing more, with more clarity. Because of AI, my volume of brand strategy work has grown—but the quality hasn’t dropped. Because now, I’m not spending hours on the mechanical parts. I’m spending those hours refining ideas that actually move people.
AI isn’t the strategist. It’s not the insight. But it’s a surprisingly good co-laborer—especially when the brief is clear and the goals are sharp.
Why it works (when it works)
Here’s the truth most people miss about AI: It only works if you do.
You still need to bring the strategy. You still need to know what makes words convert, what makes people care, and what makes a message resonate.
AI can’t do all your thinking for you. But it can accelerate what you already know how to do.
It gives you raw material faster. It helps you explore ideas with less friction. It reveals patterns, opens angles, and exposes weak spots. But it won’t tell you what matters, or why.
That still takes judgment, instinct, and experience. It takes knowing when something sounds good but won’t move a single donor to act. It takes knowing how to speak in the brand’s voice, not just write in “professional” English.
That’s why AI won’t 🤞🏻 replace fundraising copywriters or strategists. But it will make the best ones even better.
Not because it’s magic. Because it’s leverage.
What I’d tell my past self
You weren’t wrong to be cautious about AI.
You were protecting something worth protecting: the craft, the strategy, the trust you’ve built with clients and donors. But AI doesn’t cheapen your work unless you let it.
Used wisely, it reinforces what matters most: clarity, precision, and emotional connection.
I’d tell my past self this: You won’t be replaced. You’ll be reinforced.
The same instincts that made you a good writer still matter, maybe more than ever. Because now you’ve got a tool that can help you move faster, scale smarter, and stay sharp across every project.
This isn’t the end of the craft. It’s a new kind of craftsmanship. One where the thinking is still human, and the words can still work.
Would Jesus use AI?
Jesus wasn’t anti-tool. He used boats, bread, coins, mud, and stories etched in the dust. He used what was at hand—not because he needed it, but because people did.
We see this throughout Scripture when God works through human tools and human language. A staff becomes a serpent. A slingshot brings down a giant. A letter, passed from hand to hand, becomes part of the canon of Scripture.
Earlier this year, I was reading through Exodus and noticed something I’d never seen before.
God appointed a man named Bezalel to help build the tabernacle—the sacred place where he would dwell with his people.
But Bezalel wasn’t chosen for his zeal. He was chosen for his craftsmanship.
“… and I have filled him with the Spirit of God, with ability and intelligence, with knowledge and all craftsmanship, to devise artistic designs…” (Exodus 31:3-4, ESV).
In other words, God doesn’t just work through prophets. He works through makers. Tool-users. Builders with skill and Spirit.
So, would Jesus use AI?
If if it helped his followers speak more clearly, love more tangibly, and serve more faithfully… if it helped more people come to know him… I think he would.
I don’t believe AI is evil anymore, because evil takes. But this tool, used rightly, actually gives.
It gives us more time to do what we do best. It gives us more clarity. It gives us not less craft, but more of it.
Because in the end, it’s not about whether AI is good or evil.
It’s about how we use it as a tool we’ve been given—not to replace the work, but to do it more efficiently and with more fruit than we did before.
