You’ve seen the headlines. Maybe you’ve poked around in ChatGPT, typed something in, thought “huh, that’s kind of neat,” and then went back to building programs in a spreadsheet. Or maybe you haven’t touched it at all because it feels like something for tech people, not trainers.
Either way, nobody has sat down and explained AI in a way that actually matters for your work. Not what it is in theory. What it does, what it can’t do, and why any of it should matter to someone who spends their days writing programs and coaching people.
That’s what this post is for. No jargon. No hype. Just a clear picture of what’s happening and where it fits into your business.
”AI” Is a Big Word for a Simple Idea
When people say “AI” right now, they’re almost always talking about one specific thing: large language models, or LLMs. These are software systems trained on enormous amounts of text — books, articles, websites, code — and they’ve gotten very good at predicting what a useful, coherent response to a question looks like.
That’s worth sitting with for a second. These systems aren’t sentient. They don’t “understand” things the way you do. They’re extremely sophisticated pattern-matching engines that have learned from billions of examples of human writing and reasoning.
The major players you’ll hear about: ChatGPT (made by OpenAI), Claude (made by Anthropic), and Gemini (made by Google). They all work on the same basic principle, they each have different strengths, and they’re all getting more capable every few months.
Here’s what they’re not: they’re not robots that will replace you, they’re not magic, and they’re not going to spontaneously transform your business. They’re tools. Powerful tools, but tools. And like any tool, the results depend on how you use them.
Three Things AI Can Do That Actually Matter to You
Once you strip out the hype, AI is good at three things that are genuinely useful for running a coaching business.
1. Write
AI can draft. Client emails, training programs, social posts, check-in messages, newsletter content — anything that requires words on a page, AI can produce a strong first draft. The key word is draft. The output will be good, sometimes very good, but it’ll need your eye and your voice to finish it.
The practical upside: instead of staring at a blank screen, you’re editing and refining. That’s a much faster place to start.
2. Analyze
Feed AI structured information, and it can pull insights from it. Paste in your client’s last 8 weeks of training data and ask it to identify patterns. Give it a summary of your schedule and ask where you’re losing revenue. Hand it your intake form responses and ask for a quick overview of the client’s goals and history.
This isn’t magic — AI is just doing what a sharp assistant would do if you handed them a pile of notes and asked them to organize it. But having that capability available instantly, for free, changes how you can work with information.
3. Automate
This one takes a bit more setup, but the payoff is real. Tools like Zapier and Make (more on these in Post 2) let you connect AI to your existing systems so that when something happens, something else happens automatically.
A simple example: a new client fills out your intake form → AI drafts a welcome email using their specific goals and background → that draft lands in your inbox for a quick review and send. You spend 30 seconds instead of 20 minutes. Multiply that across every new client, and the math gets interesting fast.
What AI Is Bad At (and Why That Matters)
Building real trust in a tool means understanding where it fails. AI has genuine limitations, and trainers who ignore them end up frustrated or, worse, relying on bad outputs without realizing it.
AI is non-deterministic and can hallucinate. This is the term for when AI generates something confidently incorrect. It will cite studies that don’t exist, give you wrong information about a drug interaction, or invent facts that sound plausible. For general knowledge, this matters. Always verify anything high-stakes before acting on it.
AI doesn’t know your clients as well as you do. Out of the box, AI has zero context about the people you work with. It doesn’t know Sarah’s knee history, or that Marcus is crushing PRs lately and ready to push harder, or that your programming style is built around RPE not percentage maxes. Without that context, its outputs will be generic. Generic is fine as a starting point, not as a finish line.
AI can’t be in the gym with your and your client in real-time. The embodied, intuitive part of coaching — reading someone’s energy, seeing a compensation pattern before it becomes an injury, knowing when to back someone off versus push — that’s not something AI can replicate. It doesn’t have eyes or presence. It can’t feel the room.
AI models needs really clear instructions. Give AI a vague prompt and you’ll get a vague answer. The trainers who get the most out of these tools are the ones who learn to communicate clearly with them — which is a skill we’ll cover in depth in Post 3.
None of this should scare you off. It should just calibrate your expectations. AI makes you faster. You are what makes it good. Think of it like having a training partner for your business — it can spot you, it can track your reps, it can help you recover faster. But you still have to lift the weight.
The Bottom Line
AI is powerful, it’s becoming part of how modern coaching businesses run, and the trainers who figure it out now will have a real advantage over the ones who don’t.
But it’s a tool. A good craftsperson learns their tools — when to use them, how to get the most out of them, and what they can’t do. That’s all this series is about: getting you comfortable enough to actually use this stuff.
In the next post, we’ll walk through the actual landscape of AI tools worth your time — the ones trainers are using right now for programming, content, automation, and client management.
Series: AI for Fitness Professionals
- AI Guide for Personal Trainers ← you are here
- The AI Tool Landscape: What’s Actually Worth Your Time
- Prompting 101: How to Actually Get Useful Output From AI
- 5 AI Workflows That Save Trainers Hours Every Week
- Why Generic AI Isn’t Enough: What to Look For in Fitness-Specific AI
Going deeper: Once you’ve worked through this series, The Next Generation picks up where it leaves off — on what’s changing in fitness software at the architecture level, and what that means for trainers who get in early.