Mr Z
Why ChatGPT Gives Generic Workout Plans
Francesco Zen, founder of ZTraining · Last updated June 19, 2026
The questions that never get asked
Working with a client, the questions that shape a session aren't about sets and reps — they're about the person. Do you feel confident when you train? Does a loaded squat bar make you nervous? How did you wake up — energised, or already running on empty? Every one of those answers changes what the session should actually look like, and none of them show up in a one-line prompt.
What those answers actually change
- People who feel overwhelmed tend to do better with machines — they let someone express effort without the added demand of stabilising a free-weight movement.
- People with more drive and energy respond better to compound movements, which channel that intensity productively.
- Compound lifts consume more glucose. For someone travelling or skipping a meal, that's a real variable — pushing a heavy compound session on low fuel makes things worse, not better.
A generic plan can produce a good result or a bad one — that's exactly what makes it generic. The outcome depends on whether the inputs happened to match the person, which is closer to chance than to design.
What changes the outcome
This is the gap Mr Z is built to close. Trained on Francesco Zen's 18-year coaching method and running on Claude, it's designed to ask about mobility, confidence, energy state, and constraints before it builds anything — because those are the inputs that turn a workout from generic into yours.
Related: How to get genuinely personalised fitness advice from AI · Can ChatGPT build a real workout plan?