How to Make Your Chatbot Sound More Like You (or Your Client)

Happy Wednesday! This week:

Let’s get started!

A communications director at a mid-size tech company told me recently that she could spot AI-written content from her team in about three seconds. Not because it was bad — because it was pleasant. Smooth. Inoffensive. Written by no one in particular. "It sounds like a press release from a company that doesn't exist," she said. And honestly, she said, it was obvious.

The paragraph is in italics because I didn’t write it. And I didn’t write it because it never happened. I wrote the little guide below and I couldn’t come up with a lead in, so I asked Claude to write one — specifically, I asked for “a fun anecdote or something to humanize it,” and that’s what Claude came up with. My guess is that you probably figured that out about two sentences in, or at least had a suspicion that something was a little less human than it should be. And honestly? (Sorry, couldn’t resist.) The “and honestly” gave it away at the end.

AI chatbots have this sterile and kind of creepy default AI voice. And when you read something written that way, you doubt the credibility of the writer. I’ve heard stories of people intentionally leaving tpyos (😉) in their writing to make it clear that this wasn’t AI-written. I’ve replaced em-dashes with double-dashes when writing emails, just to make sure no one thinks I did them the disservice of handing off a relationship to Claude. It’s kind of ridiculous — but only kind of.

I’m not here to tell you to not use AI — obviously not, given the raison d'être of this newsletter. You should be using AI to edit drafts, help you flag gaps in your thinking, and beyond. And yes, I’m a fan of using it for first drafts, too — I almost always discard them, but it helps me get past blank page paralysis. But it’d be better if these chatbots could come closer to my voice for all of the above.

So, here’s one way to do that.

First, you need a pre-existing body of writing. (As communicators, you probably have that for yourself. For the leaders and clients we support, that may not be the case yet.) Take that work and toss it into whatever AI chatbot you like and ask it to analyze the person’s writing for, well, everything. Tone, word choice, words you avoid, odd grammar violations or phrasing that you’re fond of, sentence length, paragraph structure — the more the better. And then, ask your chatbot to create a Markdown file that captures all that information. Create a separate markdown file for yourself and each of your clients.

Markdown files are just text files that AI chatbots can read very easily without having to churn through a lot of tokens. Take what your chatbot wrote, review it for accuracy and to see where you can add even more guidance, and save it somewhere on your machine. The next time you write something, upload it as an attachment and tell your chatbot to follow the instructions in the attached markdown file to capture you (or your client’s) voice. Even better, if you use ChatGPT’s custom GPTs, Claude Projects, or Gemini’s Gems, you can add the markdown file to the instructions — you won’t have to upload it every time. (And if you’re onto Claude Cowork, Claude Code, or ChatGPT’s Codex, you can just drop the Markdown file in the folder you’re working in.)

This isn’t a perfect solution — you’ll still get the weird robot voice here and there, and this doesn’t stop AI hallucinations. But the output you’ll get is appreciably better, requiring less editing and rework. You’re putting your name on what you ultimately share. It’s worth taking a few minutes to create this map of your writing that your chatbot can — and usually does — follow. Honestly. 😉

The detailed ones. The ones with examples and edge cases. Wispr Flow lets you speak them instead — clean, structured, ready to paste into any AI tool. Free on Mac, Windows, and iPhone.

Try Wispr Flow free

Most communications teams have access to AI tools and genuinely want to use them — but months after workshops and trainings, little has actually changed about how work gets done. ROI Communication's Melanie Barna argues the problem isn't the tools or the enthusiasm; it's that teams keep adding AI on top of old habits instead of rethinking how they work from the ground up. Her firm's research across multiple communications teams found a consistent pattern: communicators score well on access and attitude, but fall short on process, data, and practical skills. Worse, leaders and their teams often have wildly different reads on how far along they actually are — which makes the gap hard to close.

Why it matters: If you’re familiar with the change management model ADKAR, this probably isn’t a shock. ADKAR is a mnemonic that stands for Awareness, Desire, Knowledge, Ability, and Reinforcement — you can read more about that, here — and I think it fits the moment. We’re all aware of AI and most of us have the desire to adopt it (or, at least, our bosses are demanding it). Most trainings are focused on the knowledge step, and — per ROI’s study — we’re doing well there, too. But we haven’t converted that into ability or reinforcement of those skills.

That’s doable — we just need to shift our efforts toward that.

Your next step: Take off the training wheels. Get the team together — and an outside facilitator — and pick a real workflow that matters, redesign it completely to be AI-first, and measure what changes. Even better: schedule a monthly check-in to iterate on and refine what you’re building — and to identify other workflows and tools to focus on. Ability plus reinforcement — those are the next steps to the change AI can bring.

(Want me to be that facilitator? Let’s talk.)

AI adoption is critical for communications teams — and I’m on a mission to help you get there. Let’s work together to get your team moving — or moving faster.

📋 A prompt playbook. Need a kickstart? My communications-focused prompt playbook has more than 600 prompts to help you get started. Download it, free.

🎓 Self-paced learning. I’ve created six courses to get you up to speed on how AI impacts communications. Get started here.

📅 A strategy call. Whether you're trying to build systematic AI adoption across your team or focused on AI search visibility for your brand or organization, I’m here to help. Learn more and book an intro call.

Most AI tools forget you the moment the chat ends. SureThing doesn’t.

SureThing is an autonomous agent that can draft in your voice, triage what matters, follow up on things you forgot, and report back with what happened next.

Day 1, you onboard it.
Day 30, it knows your clients and patterns.
Day 90, it catches things you missed.

Try SureThing

Looking for a role at the intersection of communications and AI? Here are some opportunities to check out:

Was this forwarded to you by a friend or colleague? Want to get this in your inbox next week? Click here to subscribe!

Quick question: how can I help?

What workflows are you struggling with? Where does AI still feel mysterious or overwhelming? What has worked that you’d like to share with others?

I’m a builder, and I’d love to help you and the rest of The Comms Stack community find great new ways to use AI.

Reply and tell me.
I read every response.

Even a one-sentence reply helps. For example:
“I wish AI could help me with ______.”

Until next Wednesday,

Dan

Advertise | Subscribe