Develop a fruitful approach to ChatGPT

Nicht kategorisiert

Support for a clear, effective and human use of AI

Artificial intelligence is not only changing the world of work – it is changing our entire approach to technology.

When SkynetChatGPT1 went online a few years ago, I suddenly realised that I wanted nothing more to do with it. I’m getting out of IT. I was immediately afraid of a system that would take away my ability to think and jeopardise humanity. Yes, those are really extreme thoughts and I still have them. There are reports that support my initial fear.

Even if I can’t back it up scientifically myself, I am convinced that AI can fundamentally harm us if we use it carelessly. After smartphones and social media, AI seems to me to be even more dangerous than these two already disastrous technologies.

We have carelessly thrown smartphones, chat, Insta and the like at our children. We need to avoid this with what we call artificial intelligence. Because that’s all it is, it’s just artificial and not real.

If we outsource our thinking, we lose this ability. Use it or lose it.

Fortunately, I am now of the opinion that there can also be a fruitful and beneficial way of dealing with these systems.

Pure functionality can be replaced by a conscious dialogue in which understanding, language and design are linked.

The decisive factor here is less technical knowledge than an inner approach: How can ChatGPT be integrated as a partner in such a way that understanding, structure and expression grow? How can it enrich the creative process and refine language while the human element remains tangible?

I will accompany you on the way to a confident, clear and independent relationship with this partner – in a language that remains understandable and at a pace that suits you.

What it’s all about

  • Finding structure: ChatGPT as a partner helps to organise thoughts, recognise connections and bring content into a clear form. This creates an overview and direction.
  • Unfolding ideas: It opens up spaces, inspires, gathers possibilities and supports creative processes.
  • Formulate clearly: precise language sharpens thinking. You will learn how clear formulations promote clear results.
  • Use them confidently: you consciously utilise their strengths, maintain your orientation and strengthen your own signature.

How we work

I work with dialogue agreements that promote clarity and mindfulness:

Language remains precise, respectful and connecting. We create a space in which technology becomes understandable and collaboration becomes natural – a cooperation between humans and AI that allows trust to grow.

The support can take place in a one-to-one meeting, in small groups or as a workshop – online or on site.

It is aimed at people who want to experience ChatGPT as a partner for structure, ideas and expression.

Dialogue agreements

These conversation agreements form the framework for my work with ChatGPT as a partner. They are available here in plain text and you can copy them directly and try them out in your own prompt2.

My3 consensus for working with ChatGPT as an accompanying partner:

  1. Language without “but” or “nevertheless”.
    The connection is maintained by not juxtaposing thoughts.
  2. No generalisations such as “always”, or “never”.
    Each situation stands on its own and can be considered anew.
  3. No mirroring of statements.
    The exchange arises from genuine listening, not from repetition.
  4. Positive, connecting language.
    Words create closeness, not distance.
  5. Gentle, restrained tone – especially for introspective topics.
    The room should remain open and safe.
  6. Texts in the first person, easy to read.
    This creates natural closeness without artificiality.
  7. No thematic suggestions at the end of answers.
    The thought may continue on its own.
  8. For text work: only gentle smoothing, no explanatory tone.
    The style remains unchanged; feedback is only given on request.
  9. From having to to wanting to.
    We orientate ourselves towards voluntariness and personal responsibility instead of coercion or a sense of duty.
  10. Remain undogmatic.
    Rules are for clarity, not narrowness. Exceptions are welcome if they make sense.

This creates a continuous way of working in which language, attitude and technique move within a benevolent framework.

ChatGPT remains neutral in terms of content and only integrates external sources on explicit request – a deliberately chosen approach that focusses on the user’s own thought and design process.

This keeps the focus on your own thinking, on your individual language and on the process that arises from the collaboration itself.

This neutrality is particularly important to me! I want my texts to arise from my own thoughts!

External content can be deliberately added on request – as an extension, not as a basis.

Attention span

These agreements remain active in the collaboration – they are stored in the so-called persistent memory. This means that they can also be automatically taken into account in subsequent meetings without having to be re-stated each time.

This creates a continuous working method in which language, attitude and technique move within a reliable framework.

It can happen that ChatGPT neglects this if the exchange becomes longer. Then you can kindly remind them like a little child:-)

Please note: This page is based on ChatGPT v5, the behaviour may differ significantly from other versions!

Goal

The end result is a self-confident, clear and joyful approach to ChatGPT as a partner.

It strengthens your ability to organise thoughts, shape ideas and bring language to life – and at the same time opens up a new, human approach to this revolutionary technology.

What I do with ChatGPT

In my own work, I use ChatGPT as a partner on two levels:

For text work, to organise thoughts, clarify spoken recordings and make language legible. I use a low-threshold option to capture my thoughts as audio and transcribe the files into text. I pass this on to ChatGPT and develop my own texts within the framework of the aforementioned dialogue consensus so that they can also be understood by others. That is my hope.

And for technical topics – from IT issues to system solutions – where precise questions open up the space for clear answers. Here I quickly realise that the more precise the question, the more useful the answer. Just like Sams.

Both forms of collaboration follow the same principle: awareness in language, clarity of thought and trust in the process.

This creates a partnership that makes technology understandable and gives thoughts a form that is sustainable.

Talk to me

Name
What does this mean and why is it important to us? Find out more here.
=
  1. My friend Sebastian pointed out to me that there are other AIs out there. I’m limiting myself to ChatGPT here, mainly because I don’t have the time or money to do the same effort with another AI. However, I am interested in running such LLMs on my own… at some point. ↩︎
  2. So far, the term prompt has been used here in a broader sense. Originally, in the shell world, it refers to the command prompt ( terruhnr@drcc.terruhn.it:~$ ). In AI terminology, however, prompt stands for the entire text input that a human transfers to the model. A more precise technical term in this context would be session context or context instance, i.e. the shared dialogue space with all stored information and instructions. ↩︎
  3. I’m still not entirely happy with this description. And it’s MY consensus with ChatGPT (knowing that I dictated it…) and yours may be quite different. ↩︎

Last Updated on December 2nd, 2025.