.png)
Ever since late 2022, when ChatGPT first became available to the public, the use of AI in personal and professional settings has been a divisive topic, especially online. Some of the division comes from legitimate concerns about accuracy, reliability, and whether these tools actually understand what they're doing or just sound like they do. The good news is that ChatGPT and many other large language models (LLMs) like Claude and Gemini have made significant leaps in quality over the past couple of years, to the point that the output today is almost unrecognizable from what it was when these tools first launched, and even from what they were producing as recently as last year. In other words, things are moving very quickly, and the gap between "interesting experiment" and "actually useful tool" has narrowed faster than most people expected.
So the question becomes: should personal trainers use AI in designing their programs for clients? The short answer is maybe. There's nuance here, and it's worth considering before calling it a good thing or a bad thing. Many of the points I'll make apply broadly to AI use in general, not just in fitness settings.
First of all, why would a personal trainer use AI in the first place? They should have enough expertise to do everything on their own from scratch, right? In theory, yes. A qualified trainer should be able to design a program, choose exercises, adjust intensity, and coach technique without needing a computer to tell them what to do. But there are nuances. AI tools are very good at creating templates. They can act as an assistant for different tasks, like organizing workout blocks, suggesting new exercises for different goals and populations, or generating variations on a theme when a trainer is trying to keep things fresh for a long-term client. They can also handle some of the repetitive administrative work that takes up time but doesn't require deep expertise, like formatting a program, generating a calendar, or drafting a follow-up email with instructions.
What AI can't do, at least not yet, is effectively coach in the way that a live human can. It can't watch you squat and notice that your knee is caving in on the left side. It can't read your body language and know when you're about to hit technical failure. It can't adjust on the fly when you walk in and say your shoulder's been bugging you all week. It can't motivate you through the last rep when you're convinced you can't do it, or know when to push you harder versus when to back off. Those are human skills, and they're the core of what makes a trainer valuable. So for professional trainers, AI tools probably won't be anything more than useful assistants that can speed up the administrative side of training and free up more time for the actual coaching part.
When should you be worried? If your trainer starts giving you very generic workouts that seem to have nothing to do with your goals or fitness level. This would be a red flag even if AI wasn't in the picture, but in the context of AI use, it can happen when someone just dumps a lazy prompt into a chatbot and hands you whatever comes out without any thought or customization. AI can create functional exercise sessions, but without proper prompting and human guidance, it's going to fall under the umbrella of what people call "AI slop" really quickly. This is why prompting is so important across the board for any AI use. Much of what people describe as slop comes from very poor prompting practices. Without very specific instructions, parameters, and guardrails, the tool will just do its best and give you a super basic output that lacks depth or nuance. It might be safe, and it might technically work, but it won't be tailored to you, and it won't progress in a logical way over time.
Besides, if your trainer is just giving you workouts that a poorly prompted LLM could generate, they probably aren't that good to begin with. A competent trainer should be adding value on top of whatever tools they're using, whether that's AI, a spreadsheet, or a library of exercise videos. The tool is just a tool. The expertise, accountability, and the ability to adapt in real time are what you're paying for.
So, long story short, you probably don't need to worry about whether your trainer is using AI or not. If they're using it irresponsibly, it will show in the quality and structure of their workouts and training programs. You'll notice that the sessions feel generic, that they don't build on each other, that there's no clear progression or logic tying them together. And if you're seeing that, the problem isn't the AI. The problem is the trainer.
To take things a step further, should you personally use AI to create workouts for yourself? My answer, again, is maybe. Our flagship course has many prompt templates and several prompting modules that teach good practices, and if you can craft a highly specific prompt with all the relevant parameters—your current fitness level, your goals, your injury history, your available equipment, your schedule, your preferences—then it can probably create something passable that will get you going in the right direction. But if you give a very general prompt like "Give me a thirty-minute total body workout," you'll get something decent in that it'll get your heart rate up and it'll probably be safe, but it will lack any real direction. It will stand on its own as a one-off session, but it won't build on any past training or set you up for future progress. It won't know what you did last week, or what you should be working toward next month unless it's prompted with that information. It won't know that you're stronger on your right side, or that your lower back gets tight after deadlifts, or that you respond better to higher volume than higher intensity. Those are the kinds of details that turn a generic workout into an actual program, and unless you're feeding all of that context into the prompt every single time, the AI won't have it. All that said, with enough information and ongoing input modern AI tools can put together some pretty impressive programs but they need a lot of information to work with and that level of detail in a prompt takes skill and planning in itself.
That doesn't mean you shouldn't use it. It just means you need to understand what it can and can't do, and be realistic about the role it's playing. If you're using it as a starting point, or as a way to get ideas when you're stuck, or as a tool to fill in gaps when you don't have access to a trainer, that's fine. Just don't expect it to replace the kind of personalized, progressive, adaptive programming that a good coach would provide.
We have a growing free learning center, a growing collection of mini-courses, and our flagship course to help along the way if you'd like some guidance.
Learn to spot generic programming — Whether your trainer uses AI or not, your workouts should feel specific to you. Look for signs of personalization: exercises that match your goals, progressions that build on what you did last week, modifications that account for your limitations if you have any. If every session feels like it could have been pulled from a random fitness blog, something's off. That's not to say basic is bad, but look for patterns in quality.
If you're using AI to program for yourself, write better prompts — Instead of "Give me a workout," try "Give me a three-day-per-week full-body strength program for an intermediate lifter with a history of lower back pain, access to dumbbells and a bench, and a goal of building muscle while avoiding spinal loading." The more specific you are, the better the output will be. Include your experience level, your equipment, your goals, your limitations, and how this workout fits into your week.
Test the output before you commit — If you generate a few workouts or a basic program with AI, do the first session and see how it feels. Does the volume make sense? Are the exercise choices appropriate? Does the session have a logical structure? If something feels off, revise the prompt. Don't assume the first version is the final version.
Use AI as a brainstorming tool, not a replacement for judgment — AI can suggest exercises you haven't thought of, generate variations on a theme, or help you organize a messy idea into a structured plan. But you still need to apply your own judgment about what makes sense for your body, your goals, and your context. Treat the output as a working draft that can modified if necessary.
Focus on the coaching, not the tool — If you're working with a trainer, the value isn't in whether they use AI or write everything by hand. It's in whether they're paying attention to you, adjusting based on your feedback, teaching you proper form, and helping you progress over time. The tool they use to organize the program is far less important than the quality of the coaching itself.
AI slop — Low-quality, generic content produced by artificial intelligence tools when they're given vague instructions or used without human oversight. The output is technically functional but lacks depth, nuance, or relevance to the specific context.
ChatGPT — A conversational AI tool developed by OpenAI, based on a large language model (LLM). It became widely available to the public in late 2022 and is one of the most well-known AI assistants used for generating text, answering questions, and assisting with tasks.
Claude — A large language model developed by Anthropic, designed to assist with text generation, conversation, and reasoning tasks. It's one of several LLMs that competes with ChatGPT in terms of quality and capability.
Gemini — A large language model developed by Google, part of the company's suite of AI tools. It's designed to handle a wide range of tasks, from answering questions to generating content.
Large language model (LLM) — A type of artificial intelligence trained on vast amounts of text data to understand and generate human-like language. Examples include ChatGPT, Claude, and Gemini. These models can write, summarize, answer questions, and assist with a wide variety of tasks.
Prompting — The process of giving instructions or input to an AI tool to guide its output. The quality of the prompt referring to how specific, clear, and detailed it is, has a huge impact on the quality of the response. Good prompting includes context, constraints, goals, and examples.
Progressive programming — A training approach in which workouts are designed to build on each other over time, gradually increasing difficulty, volume, or complexity to drive continued improvement. This is the opposite of random, one-off workouts that don't connect to a larger plan.
Template — A pre-structured framework or outline that can be customized for different users or situations. In fitness, a template might include the basic structure of a workout (exercises, sets, reps, rest periods) that can be adjusted based on individual needs.
