One of the first useful and easy-to-use functions of ChatGPT was the creation of texts. As if by magic, the Artificial Intelligence (AI) generated amazing knowledge on a predefined question after a short input (prompt) – and in a comparatively safe style. It didn’t take long for numerous AI-generated texts to flood the internet. They can be seen every day on social media channels such as LinkedIn or in Amazon’s self-publishing databases. However, the quality of this content generally leaves a lot to be desired, as it is easy to see that no technically competent author has put their hand to these works. However, no one wants to be constantly bombarded with meaningless content that comes from a machine. That’s why we’d like to take this opportunity to show you how AI solutions such as ChatGPT can help you as a co-author to put your own thoughts on paper more effectively.

AI is a co-pilot and not a self-runner 

First of all, we can state that generative AI (GenAI) is not yet really capable of generating text formats that are directly flawless in terms of structure, content and style. Particularly with regard to technically complex content and very specific issues, GenAI finds it difficult to immediately generate content that can be published without further ado. However, AI can provide you with excellent support in developing initial ideas, structuring a text in advance and reducing the overall research effort. In short, it helps us to overcome writer’s block and the dreaded blank page. In the first step, however, it is important to feed the AI with the right questions, content and contexts in order to obtain target-oriented results in the second step. Unfortunately, no one will be able to do this intellectual work for you. In our blog post “Prompt correctly to get the most out of ChatGPT”, we have already shown how correct work instructions are entered in ChatGPT and what results they produce. Here we now show what this means in terms of good and informative texts.

Develop a central idea 

An exciting text is characterized by the fact that it contains an interesting central question. This is then answered along a structure that has yet to be developed and filled with content. A question such as “What can AI do?” is formulated too generally and would mean that, in case of doubt, entire books could be written on this topic. But that is rarely the goal. The question must therefore be formulated in more detail. 

Self-experiment

Try out the following: “How can AI support me in writing texts?”

This is much more targeted as an example and you will get the first valuable answers directly when you formulate the question in Chat GPT. However, if you still have no idea what question you want to work on, the following prompt can help: 

Self-experiment

Please enter the question “What are the current challenges in the field of artificial intelligence?” into the input mask. 

In response to this question, Chat GPT lists detailed aspects that can serve as a basis for further questions. These include data quality and availability, bias and fairness, explainability and transparency, security and data protection, scaling and resource management, regulation and ethical considerations, human-AI interaction as well as generalizability and robustness. With specific questions within these detailed aspects, this extends into the area of how initial steps can be taken to solve the problem and how the measures can best be implemented.

Establish a structure and a common thread 

In order for the question to be answered correctly and comprehensibly, the text needs a rough structure or common thread. This defines which statements and arguments are placed at which point in the text. Chat GPT can act as a co-editor here and be tasked with providing a structure. To do this, however, the solution must again know what it is actually about. It therefore makes sense to specify the question and also say which text format is to be created and to what extent. A LinkedIn post is structured differently to a technical article or a white paper. So try the following: 

Self-experiment

“You are an editor and have the task of writing a byline article for an IT trade journal. It should be about how AI can support users when writing texts. The article should have a length of 8,000 characters including spaces. Please create a structure for this article.”

The result will look like this: 

1. Introduction (approx. 800 characters) 

  • Brief introduction to the topic of AI in the writing process. 
  • Overview of the current relevance and development of AI in text creation. 

2. Background and development of AI in writing (approx. 1200 characters)  

  • Brief outline of the history of AI in the writing process. 
  • Examples of early applications and their development to date. 

3. Basic functioning of AI writing aids (approx. 1000 characters) 

  • Explanation of how AI-based writing tools work. 
  • Overview of different types of AI writing assistants. 

4. Practical applications (approx. 1500 characters) 

  • Detailed examples of how AI is used in the writing process (e.g. in creative writing, academic papers, technical documentation). 
  • Case studies or user reports.

5. Advantages of using AI in the writing process (approx. 1000 characters) 

  • Increased efficiency and time savings. 
  • Improving text quality and promoting creativity. 

6. Challenges and limitations (approx. 1000 characters)  

  • Discussion about the limits of AI in terms of creativity and understanding. 
  • Ethics and data protection in dealing with AI tools. 

7. Future perspectives (approx. 1000 characters) 

  • Forecasts for the further development of AI in the field of writing. 
  • Possible new areas of application and technologies. 

8. Conclusion (approx. 500 characters) 

  • Summary of the most important points. 
  • Concluding thoughts on the significance of AI for the future of writing.

ChatGPT answers this question with eight overarching points – from the introduction, background and functions to the conclusion. Each main topic also has two sub-questions and an indication of the scope each structural aspect should have. At this stage, it is easy to add or remove individual topics. If the structure does not meet your expectations, you can simply ask the AI for an alternative. 

Developing content 

To create the content and statements for the individual structure points, you can simply continue working in the current input mask of ChatGPT. At this point, the AI already knows the context of the requests and can build on this. For example, the following can be asked for point 5:

Self-experiment

“Why are there efficiency gains and time savings and how can these be achieved?”

The AI solution then delivers numerous more or less target-oriented contents that save a lot of time for research on the Internet. However, this is also where it becomes clear that the copywriter has a responsibility to assess the quality of the statements and curate the generated content. Particular care must be taken to ensure that statements presented as facts also correspond to the truth. ChatGPT and other AI systems have a tendency to hallucinate. This means that they will generate statements as facts even if they do not correspond to the truth in reality. For example, when asked, the solutions refer to study sources that do not exist.

In principle, however, it is possible to use questioning techniques to generate increasingly in-depth content, which then contributes to answering the key question. However, it will be difficult to create a text with just one prompt that directly covers all the information and arguments in full and satisfactorily. It is up to the copywriter to compile these individual fragments into a meaningful contribution. Further research is also usually unavoidable with regard to vivid and comprehensible examples.

The finishing touch: improved spelling and grammar: 

Modern AI writing tools offer advanced spelling and grammar checks. They detect and correct errors more efficiently than traditional word processing programs. This results in faster processing and a reduction in the time spent on proofreading. AI systems can also make suggestions for better wording or stylistic improvements, which can be particularly helpful for creative writing or technical texts. This supports the writing process and can increase the quality of the final product without requiring additional time for revision.

Self-experiment

Write: “Give me synonyms for the word “meaningful”!”

ChatGPT answers that there are a number of synonyms for the word “meaningful” that can be used depending on the context. Here are some examples: 

  • Purposeful 
  • Useful 
  • Practical 
  • Appropriate
  • Effective 
  • Advantageous
  • Beneficial 
  • Productive 
  • Value-adding
  • Promising 
  • Constructive

Each of these synonyms emphasizes slightly different aspects of “meaningful”, so the choice of the most appropriate word depends on the specific meaning.

Eine Auswahl an kompetenten Co-Textern 

GPT-3 and GPT-4 are advanced AI language models capable of generating human-like text. They can be integrated into various applications to automatically create content, rewrite or improve existing texts. However, there are several AI tools that support text writing, each offering unique features to improve the writing process. Here is a brief overview. 

Grammarly 

Grammarly is known for its advanced grammar, punctuation and style checking features. It is particularly useful for ensuring error-free and clear texts. Unfortunately, it only works in English, so the use of a translation tool such as deepl.com is recommended. With deepl write, this now also has a good grammar checker.

Copy.ai 

Copy.ai uses AI to produce creative and engaging content. It’s great for brainstorming ideas and creating first drafts for different types of copy, such as blog posts, social media content and marketing materials. 

Writesonic 

Writesonic is focused on creating marketing content. It can quickly create high-quality digital ad copy, blog posts, landing pages and more, making it a favorite among digital marketers. 

Jasper 

Jasper excels at creating copy that converts. It’s especially effective for writing product descriptions, email campaigns and website copy that’s optimized for conversions. 

QuillBot 

Known for its paraphrasing tool, QuillBot can help rewrite existing content to avoid plagiarism while retaining the original meaning, which is helpful for refining and diversifying your copy.

Hemingway-Editor 

The Hemingway Editor is not AI-driven, but it is an excellent tool for making your texts clear and concise. It highlights complex sentences and common errors, making it easier to write in a concise and readable form. 

Each tool has its strengths, and the best choice depends on your specific needs, be it grammar checking, idea generation, SEO optimization or creating conversion-oriented copy. 

Copywriting with the support of an AI tool requires some practice and you should always check the results critically. In the long term, however, you will save a lot of time and generate better results. If you would like to find out more about good prompting, just take a look here.

About the author

Mathias Herrmann

CEO at ALLEHERZEN

Mathias Herrmann is an internet entrepreneur going back to the Internet’s early days with a deep interest in digital and future technologies. For over 20 years, he has been helping companies make the most of their data by forging innovative solutions – without forgetting the people behind the data.

Book a free video call of 30 or 60 minutes directly now: