← Close
How HEROMINUTES Uses AI
Version 1.0 · Effective: August 2026
The honest summary: your training programs are written by artificial intelligence. No human reviews your program before you see it. This page explains exactly how that works, what it is good at, and where it deliberately stops.
1. In one minute
- An AI model — Claude, made by Anthropic — writes your program.
- It works only from what you tell the app: your intake answers, your comfort and discomfort logs, your current phase, and your daily check-ins.
- On Comeback and Desk Athlete, it runs in more than one pass — screening happens first, separately from writing the session. Other journeys use a single combined pass.
- No human reviews your program. There is no coach, physiotherapist, or doctor checking it before it reaches you.
- It is fitness guidance. It is not medical advice and does not replace a qualified professional.
2. What happens when you generate a program
Here is the sequence for Comeback and Desk Athlete, the two journeys that use separated passes. On the other journeys, steps 2 and 3 happen together in a single pass — the same information goes in, but one step does both the screening and the writing.
1
You
You answer the intake
Which area is bothering you, how it started, how much you can currently do, your training background, and your goal. Later on, your daily check-in and comfort logs join this picture.
2
AI · Screening pass
Your answers are screened before anything is written
The first pass reads your full profile and produces a structured summary: how much load is reasonable right now, what to avoid, what to watch, and whether anything you reported should be looked at by a professional before you train at all.
3
AI · Programming pass
A second pass writes the session
This pass receives the screening summary and works inside it. It chooses your exercises, sets, reps, tempo and progression, and writes the coach note explaining why. It is given the constraints before it is given the freedom.
4
AI · Safety review
On some journeys, further passes audit the result
An additional review pass checks the written program back against the screening summary, and a final pass reconciles the two. Which journeys use this depends on your plan and is shown in the app as your AI mode.
5
You
You see the program — and you can change it
You can swap an individual exercise for an alternative, rate how each one felt, log discomfort at any point, or stop. Your ratings feed into how your next program is written.
The handover in step 2 is the part that matters. Where the passes are separated, the pass that writes your session never sees your raw answers on their own — it only ever sees them with the screening constraints already attached.
3. Why more than one pass
A single pass has to hold two competing jobs at once: be careful, and be useful. Splitting them means the boundaries are decided before the session is written, by a step that has no interest in making the workout impressive — and on some journeys they are checked again afterwards by a step that did not write it.
Internally these passes are given different areas of expertise to reason from — screening, programming, and safety review. They are reasoning steps, not people. There is no panel of experts, and nobody with a professional qualification is involved in producing your program.
4. Where the AI stops
This is the most important section on the page, and it is the part we are most confident about.
Some answers are signals that no exercise program is the right response — symptoms that need a person, in the room, with hands and equipment. When those appear, the app stops and points you to professional care instead of writing you a workout. It does not program around them, and it does not soften them.
You are not locked out. If you have already spoken to a professional about it, you can confirm that and continue — and the app then serves a standard, non-personalised template rather than building a program around symptoms it flagged. That choice is recorded.
There are two independent layers doing this, on purpose:
- The screening pass above, which can halt the whole generation.
- A second check that runs on your own device, which works even when the AI is unreachable — no internet, an outage, or a plan without AI. It screens your answers for the same kinds of emergency signal before any standard program is shown.
The second layer exists because the first one only protects you when the servers are up. Safety that depends on a network connection is not safety.
If you are experiencing numbness, weakness, loss of bladder or bowel control, a hot or swollen joint, fever alongside pain, chest pain, or an inability to bear weight — stop and seek medical care now. Do not wait for the app, and do not use this app to decide.
5. What is sent, and what never is
Two separate services are involved, and they receive very different things.
Anthropic (Claude) — writes your program
| Sent | Never sent |
|---|
| Intake answers · current phase and history · comfort and discomfort logs · readiness check-ins · your age (not your full date of birth) · biological sex, if you provided it · a generic label rather than your name |
Your name · email address · password · payment details · location · anything identifying you as a person |
Under Anthropic's standard API terms, data sent this way is not used to train their models. See Anthropic's Privacy Policy.
ElevenLabs — speaks the exercise guidance aloud
When you tap the audio button on an exercise, the voice you hear is generated by ElevenLabs. It receives only the generic description of how that movement is performed — the same words for every person doing that exercise — plus short shared phrases like counted numbers and rest cues.
It never receives your identity, your intake, your comfort scores, or anything about your situation. Your own numbers — your sets, your reps, your comfort ceiling — are assembled on your device from those shared phrases, so your personal figures never leave it.
Audio is stored on our own servers once created and reused after that, so the same guidance is not sent out again. See ElevenLabs' Privacy Policy.
6. What AI cannot do
Being straight about the limits is part of using it responsibly.
- It cannot examine you. It has never seen you move. Everything it knows, you typed.
- It does not diagnose. It does not identify what is wrong with you, and nothing it produces is a finding about your body.
- It can be confidently wrong. AI models produce fluent, plausible text whether or not it is correct. Confidence in the wording is not evidence.
- It only knows what it was told. If your intake is incomplete or out of date, the program is built on that.
- It cannot feel what you feel. If a movement hurts, that information is yours alone — and it outranks anything on the screen.
If the program and your body disagree, your body is right. Stop, log it, and speak to a qualified professional.
7. What this gives you
Used within those limits, this is what the approach is actually for:
- It adapts. Your program is rebuilt from your current state, not chosen from a fixed menu. Log more discomfort and the next one is more conservative.
- It is consistent. The same screening runs every time, at 6am and at midnight, on day one and day eighty.
- It explains itself. AI-written sessions carry a note saying why the session looks the way it does, and every exercise has guidance you can read or hear.
- It progresses on evidence, not the calendar. Phases advance on your logged check-ins, not because time passed.
- It knows when to send you elsewhere. The most valuable thing it does is refuse.
8. Your control
- Swap any single exercise for an alternative (available when AI mode is active).
- Rate each exercise as too hard, about right, or too easy — this feeds the next generation.
- Delete your account. Everything identifying you is erased, along with everything you have written. Some training data is kept in an anonymous form that is not linked to you. See the Privacy Policy.
9. Disclaimer
HEROMINUTES is a general wellness and fitness guidance app. It is not a medical device. The AI-generated programs, exercise guidance, audio guidance and any insights in the app are for general fitness purposes only. They are not medical advice, and they are not a substitute for assessment, advice or care from a qualified healthcare professional.
Always consult a qualified professional before beginning any exercise programme, particularly if you have an existing injury or condition, are recovering from surgery, are pregnant, or take medication. Never disregard professional advice, or delay seeking it, because of something this app produced.
You use the app at your own risk. See the full Medical Disclaimer and Terms of Use.
10. Changes
If how we use AI changes materially, this page changes with it, and we will tell you in the app.
11. Contact
Questions about any of this: herominutes@gmail.com
Related documents: Privacy Policy · Terms of Use · Medical Disclaimer