• 10 – 13 April 2025
  • Exhibition Centre Cologne

Is ChatGPT revolutionising the fitness industry?

(c) Shutterstock

Bar ohne Namen

Entschlossen verweigert sich Savage, der Bar einen Namen zu geben. Stattdessen sind drei klassische Design-Symbole das Logo der Trinkstätte in Dalston: ein gelbes Quadrat, ein rotes Viereck, ein blauer Kreis. Am meisten wurmt den sympathischen Franzosen dabei, dass es kein Gelbes-Dreieck-Emoji gibt. Das erschwert auf komische Weise die Kommunikation. Der Instagram Account lautet: a_bar_with_shapes-for_a_name und anderenorts tauchen die Begriffe ‘Savage Bar’ oder eben ‚Bauhaus Bar‘ auf.


Für den BCB bringt Savage nun sein Barkonzept mit und mixt für uns mit Unterstützung von Russian Standard Vodka an der perfekten Bar dazu.





Chat GPT is on everyone's lips, drawing attention to chatbots and AI. What Chat GPT is and can do, how powerful such tools already are and why a critical distance should nevertheless be maintained.

A comment by Andreas M. Bechler, Hashtag Fitness Industry.

At the end of 2022, OpenAI, a US company based in San Francisco, made not only tekkies sit up and take notice with the release of the Chat GPT chatbot. Major competitors such as Google raised alarm and are now working on a rival product. Facebook mother company Meta also wants to get in on the chatbot action and is targeting researchers in particular. Meanwhile, OpenAI continues to work on its own bot with the support of Microsoft and celebrated the Chat GPT 4 release in mid-March 2023. The new version can now also process uploaded images. In addition, the first payment models are also being tested.

This shows that there is a lot of movement in the market for chatbots and artificial intelligence. Sooner or later, they will also affect the fitness market. So it's time to get to grips with them.

What is Chat GPT?

Chat GPT is "only" a so-called chatbot. I can ask a chatbot (usually text-based) questions and it answers me on the basis of the knowledge with which it was previously enriched. The question can also mean a task for the chatbot, which it then processes for the user. But let's just ask ChatGPT itself who it is:

"I am ChatGPT, a large language model based on the GPT 3.5 architecture and trained by OpenAI. I am designed to perform different types of language understanding and text generation tasks, such as writing articles, answering questions or translating texts. I have a vast knowledge of different topics and can interact with users in different ways to meet their needs."

Not all that new. Chatbots that help customers quickly and easily are already integrated into everyday business life. What is special about ChatGPT is, among other things, the immense amount of data with which it was trained. This makes it special in the quality of the individual answer and the breadth of knowledge in general. In addition, the almost human conversation style is certainly still unique at present.

How can I try Chat GPT?

To better understand the tool, it makes sense to try out the gpt chat yourself. All you need is an account, which can be created via a Google or Microsoft account. Then start a "new chat" and ask the chatbot your questions.

But what can Chat GPT do?

Since I also wanted to be convinced of the performance of this new tool, I had some test conversations with the chatbot. The result: I was amazed by the impressive capabilities of this tool - also for the fitness sector. For example, let's have Chat GPT create a training plan for us.


Requirement: Can you create a training plan for me with the goal of building muscle?

Result: However, it is important to note that the plan should be tailored to your individual needs and abilities. It is recommended that you seek advice from a qualified fitness professional before starting a training plan to ensure that it is suitable for you.

Here is an example of a training plan for muscle building:


  • Squats: 3 sets x 8-12 repetitions
  • Deadlift: 3 sets x 8-12 repetitions
  • Bench press: 3 sets x 8-12 repetitions
  • Shoulder press: 3 sets x 8-12 repetitions
  • Triceps extensions: 3 sets x 8-12 repetitions
  • Biceps curls: 3 sets x 8-12 repetitions


  • Rest day or cardio training


  • Leg press: 3 sets x 8-12 repetitions
  • Lunges: 3 sets x 8-12 repetitions per leg
  • Lat pulldown: 3 sets x 8-12 repetitions
  • Rowing: 3 sets x 8-12 repetitions
  • Chest fly: 3 sets x 8-12 repetitions
  • Triceps extensions: 3 sets x 8-12 repetitions


  • Rest day or cardio training


  • Squats: 3 sets x 8-12 repetitions
  • Deadlift: 3 sets x 8-12 repetitions
  • Bench press: 3 sets x 8-12 repetitions
  • Shoulder press: 3 sets x 8-12 repetitions
  • Triceps extensions: 3 sets x 8-12 repetitions
  • Biceps curls: 3 sets x 8-12 repetitions


  • Rest day or cardio training


  • Rest day

In addition to training, it is important to watch your diet and eat enough protein to help build muscle. It is recommended that you consume at least 1.6 grams of protein per kilogram of body weight per day. It is also important to get enough sleep and give your body time to recover.


When you look at this training plan, you have to honestly admit it's not bad. Exercises, numbers of sets and repetitions as well as the embedding in a weekly rhythm are quite comparable to what a graduate of a trainer's licence would answer to a corresponding exercise task. The first advantage is therefore the high quality of the output.

However, there is one small difference: while humans spend a few minutes on this task, Chat GPT spits out the result almost immediately with the submission of the question. The high speed of answering is certainly another advantage.

In conversations with Chat GPT, the extreme breadth of knowledge of this tool also becomes clear. At first, I ask for a training plan, shortly afterwards for a cake recipe for my next birthday and finally I want to know something about the Roman emperor Caesar. Chat GPT can answer all these questions adequately.

What are the limitations of ChatGPT?

To illustrate the limitations of Chat GPT, I again use an interaction with the chatbot. This time, I asked the tool to write a meal plan. To save space, I have shortened the answer by the individual ingredients and the calories listed for each ingredient.


Requirement: Write a diet plan for me, 33 years old, 191 cm tall and 95 kilograms and regular physical activity. There are no allergies and no special diets to follow.

Result: Here is an example of a diet plan for someone who is 33 years old, 191 cm tall, weighs 95 kilograms and has regular physical activity and no allergies or special diets:

  • Breakfast (300 calories): **Listing of ingredients**.
  • Snack (150 calories): **Listing of Ingredients**.
  • Lunch (450 calories): **Listing of Ingredients**.
  • Snack (150 calories): **Listing of Ingredients**
  • Dinner (450 calories): **Listing of Ingredients**

In total, this is about 1,500 calories per day.


I don't have to be a trained nutritionist to realise that 1,500 kcal a day doesn't work for a man of my stature and athleticism. This is where the first limitation becomes obvious. The chatbot can only be as good as the data it has been trained with. The assumption is that it has found the average basal metabolic rate of a man of 1,500 kcal in its data and created a corresponding plan. In any case, the results should be questioned critically.

However, the answer can be improved by asking the questions in stages. If one first asks for a calorie calculation for the individual person (result: 2800 to 3500 calories per day) and then for a diet plan adapted to this, the result is already better. Interestingly, however, if you ask for calorie information for the plan, it is still significantly below the chatbot's own result at 2,250 kcal.

Throughout the dialogue, another point of criticism becomes clear. I don't get any information about the sources from which the chatbot draws its data. Especially in the case of the results that are to be critically questioned, an insight into the original source would be helpful to be able to evaluate the quality of the answer. As a part-time lecturer, I have to say that for my students, the lack of naming of sources would be a no-go. Even when asked, the chatbot unfortunately does not disclose its source.

Another disadvantage arises especially in a business like the fitness industry with always new trends and innovations. "The chatbot can only be as good as the data it has been trained with". However, this also means that the underlying data is always historically in the past. So the chatbot cannot even know current trends because they are not in its data set. This should always be taken into account when interpreting the response.

Of course, the list of disadvantages is not yet complete and could be extended. With the limitations mentioned and the reference to the stage-by-stage questioning (role play), I have relied on two (German-language) YouTube videos by Niko Tauscher from iGroup Internetagentur. I can only recommend both videos for further deepening this topic.


In the second part of the comment, you can read how Chat GPT can still be used successfully in the fitness business.


About the author

Andreas M. Bechler is an author, consultant, lecturer and podcaster in the fitness industry. In addition, Andreas is the spokesperson for the fitness industry working group of the VSD - Verband für Sportökonomie und Sportmanagement e.V. You can get in touch with Andreas via LinkedIn: www.linkedin.com/in/andreasmbechler


About Hashtag Fitness Industry

Hashtag Fitness Industry is the podcast about the German fitness industry with all its facets and players. It is intended to enable a transfer of knowledge between the interview guests and the listener for the benefit of the entire industry. Current trends and developments are as much a part of the podcast as fundamental strategic questions from the daily business of fitness providers. Further information: hashtag-fitnessindustrie.de