What are custom GPTs?

Custom GPTs or AI assistants are customized and individual AI systems that are specially tailored to specific needs, requirements or industries. In contrast to general AI applications, such as the well-known ChatGPT, which cover a wide range of topics and questions, CustomGPTs are trained on specific domains, data sources or task areas. As a result, they offer personalized answers and are better able to respond to context-specific questions and challenges and complete tasks for their users.

For example, a CustomGPT could be developed for a company, a brand or a specialist field and integrate the knowledge, processes and vocabulary of this company in order to subsequently write suitable marketing texts for the company. Employees can use the Custom GPT on their own or share it with the team. This ensures that the tone of voice remains consistent and at the same time significantly increases employee productivity by supporting the AI model. You can find out more about custom GPTs in my article on implementing custom GPTs.

What examples of custom GPTs are there?

Custom GPTs have become known primarily through the GPT Store from OpenAI. Every Chat GPT user can access the GPT Store via the navigation on the right-hand side using the “Explore GPTs” button and then use specific GPTs developed by others.

Here is a small selection of my GPTs

1. branding partner

This GPT creates guidelines for communication based on links and documents that you make available to it. To the GPT.

2nd AI Language Tandem

You can learn languages with this GPT. To the GPT.

3 Frida – For relatives of people with dementia

This GPT plays a role play with you so that you can learn how to communicate with people affected by dementia. To the GPT.

What is the difference between ChatGPT and CustomGPTs?

The difference between CustomGPTs and the “normal” ChatGPT or similar tools such as SwissGPT, Peak Privacy or Perplexity lies mainly in the customization and specialization:

1. Customization for special needs:
   • CustomGPTs are specially trained or customized versions of the general GPT model that are geared towards specific use cases, specialist areas or applications. This means they are optimized for specific tasks or industries and may have been given specific data to be useful in a specialized area.
   • The normal GPT, like GPT-4, is a universal language model that can be used for a variety of use cases without being specifically specialized for a particular task. It offers a wide range of answers based on a very large, general data set.
2. Training and fine-tuning:
   • With a CustomGPT, the model is also fine-tuned to company-specific or subject-specific data. This helps to provide more precise answers that are particularly relevant for a specific niche.
   • The normal GPT, on the other hand, has been trained on a huge amount of data from the Internet and can therefore provide a wide range of information, but sometimes without the specialized depth achieved by additional fine-tuning.
3. Use cases:
   • CustomGPTs are often developed by companies that want to tailor an AI assistant to specific customer requirements or a specific use case, for example for customer support in a specific industry or for specific internal processes.
   • The normal GPT is more versatile and can be used by anyone for a variety of purposes, from general knowledge questions to creative writing tasks, without any special adaptations.

What are the advantages of CustomGPTs?

CustomGPTs offer several advantages compared to the classic ChatGPT. Here are the most important added values of the individual AI assistants.

1. Improved efficiency for specific tasks:
   • CustomGPTs are excellent at efficiently solving certain recurring tasks or handling standardized processes. Thanks to their specialization, they can provide faster and more targeted help than a generally trained model. Users do not have to give the CustomGPT the system prompt every time, but the CustomGPT knows its role and the scenario in which it is located.
2. Reliability and security:
   • A CustomGPT can be customized so that it does not provide any inappropriate information and only uses what is relevant for the respective use case. In addition, data and answers are easier to control.
3. Consistent communication:
   • CustomGPTs can communicate in a particular tone or style that suits the brand – be it formal, casual or particularly customer-friendly. The normal ChatGPT varies more in the way it responds.
4. Specialized expertise:
   • CustomGPTs are specially tailored to specific topics, industries or use cases. As a result, they provide more precise and relevant answers for specific tasks or specialist areas, while the normal ChatGPT provides more general answers.
5. Adaptation to company data:
   • CustomGPTs can be trained on company-specific data. This means they know your products, services and processes better and can respond to them more precisely and appropriately.

CustomGPTs offer a customized solution for recurring tasks or questions, while the normal ChatGPT can provide a general answer to many questions. CustomGPTs are optimal when it comes to having a model that works particularly well in a specific context.

When do I need a CustomGPT or AI-Assistant?

Simply put, a CustomGPT is always suitable if you have recurring questions or processes. In this case, you benefit from the advantage that you do not have to re-enter your prompt or system prompt each time, but your CustomGPT is already in a scenario or role defined by you.

To help you decide whether a CustomGPT is worthwhile, you can ask yourself the following 5 questions. If you answer YES to at least one of these questions, it is worth considering implementing a CustomGPT. The more questions you answer with YES, the higher the probability that you will benefit from the added value of a CustomGPT and that the implementation of your own AI assistant will have a positive impact on your efficiency and quality.

1. do you have specific topics or expertise that is important? If so, a CustomGPT could provide more precise answers.

2. should the model speak specifically to your company or products? An adapted model could be useful here.

3. are there recurring tasks or processes that the model should support? CustomGPTs can help to process such tasks more efficiently.

4. should the model always respond in a certain style? A CustomGPT can be trained to the desired tonality.

5. does your company need specific support with tasks such as customer questions? A CustomGPT could be trained specifically for such requirements.

What are typical applications for CustomGPTs?

The following is a list of possible applications for a CustomGPT:

• GPT for creating social media posts
• GPT for creating newsletters
• GPT for creating applications
• GPT for creating product descriptions for specific target groups
• GPT for training in the form of Q&A or quizzes
• GPT for training in the form of role play (the GPT is a customer, the sales employee has to sell something, the GPT regularly gives tips on how to conduct the sales talk better).
• GPT for replying to e-mails

How do I implement a CustomGPTs?

To create a CustomGPT, you usually proceed in several steps.

1. define goal

First of all, it must be clearly defined what the CustomGPT is to be used for. You determine which specific tasks the AI should take on, e.g. writing blog posts in marketing, answering customer inquiries in a doctor’s surgery or drawing up contracts in a legal department.

2. data collection and prompt

To adapt the GPT model, a very precise prompt is required and, in the best case, sample data. This data should be specific to the task area or industry in which the CustomGPT is used. Examples of data are:

• Blog posts or articles from the company for an AI marketing assistant
• Previous contracts or legal documents for an AI legal GPT

3. fine-tuning of the CustomGPT or AI assistant

In this step, the pre-trained GPT model is further trained with the collected data. You adapt the model by teaching it the company-specific content, technical language or processes. This is done in the form of prompts. For a blog-related CustomGPT, for example, the AI assistant is given previous blog posts and style guides from the company so that it strikes the right tone.

4. test CustomGPT

After the CustomGPT has been customized or created, it is tested. Employees or developers give it typical tasks and check whether the GPT’s answers meet the requirements. This checks whether the GPT responds to requests correctly and in context and whether it writes in the correct style.

5. fine-tuning and iterations

The CustomGPT will be further optimized based on the feedback from the test phase. Additional data may need to be added or adjustments made so that the AI assistant works more accurately or responds better to certain requirements.

6. integration into the workflow

As soon as the CustomGPT is working well, the AI assistant will be integrated into employees’ day-to-day work. This sometimes requires short training sessions, but is generally a minor issue as the CustomGPTs are already developed very closely with the employees.

I have gathered a lot of experience and best practices over the last few months and will be happy to support you in implementing your CustomGPT. Just click on the button and find out more about my offers for the implementation of CustomGPTs.

What is the data protection like with CustomGPTs?

The use of the OpenAI custom GPTs shown above is subject to the same data protection guidelines as the use of the general ChatGPT. This means that all input data is always processed further by OpenAI. This means that the data entered in the input field of the custom GPT is not protected and users can only use this type of AI assistant for applications without sensitive data.

It must also be ensured that no confidential data is passed on to OpenAI when the custom GPT is created. Custom GPTs are personalized by specific prompt instructions and by uploading your own knowledge sources. However, this knowledge is not protected, but is also accessible to OpenAI.

How do I create AI assistants that take data protection into account?

The advantages of individual AI assistants are enormous. Thanks to the ability to customize the AI Assistant to your own needs by giving it documents and prompts as behavioral instructions, AI Assistants are a powerful tool. However, anyone who is concerned about their personal data must choose alternatives to the custom GPTs from OpenAI.

One possible alternative is Peak Privacy. In addition to the general AI chat, such as ChatGPT, Peak Privacy also offers the option of creating your own AI assistants, known as “constructors”. Read more about data protection-compliant implementation in my last article“Implementing AI assistants securely – Secure alternatives to custom GPTs from OpenAI“.

Frequently asked questions about CustomGPTs?

Who has access to the data entered in CustomGPTs?

When using ChatGPT from OpenAI, the input data is further used by the respective companies for training purposes. Basically, the data protection processing here is the same as with the classic ChatGPT. When using secure applications such as Peak Privacy or SwissGPT, the data entered is not used for any other purpose and third parties have no access to it.

What does the implementation of AI Assistants cost?

The exact pricing depends on the scope and use of the AI Assistant. It is important to know that you need a Pro account for the CustomGPTs from OpenAI. Added to this is the effort required to implement or create the system settings. Find out more in the GPT offers.

How long does it take to implement CustomGPTs?

In principle, CustomGPTs or AI assistants, such as the designers of Peak Privacy, can be implemented in just a few hours or days. This is followed by testing and optimization. Depending on the application, this may take a few more days.

How are CustomGPTs shared in the team?

At the end of the creation of a CustomGPT, OpenAI offers three options for publishing the CustomGPT:

1. Only the creators themselves can use the CustomGPT
2. Only those who have the link to the CustomGPT can use it
3. The CustomGPT can be made publicly available in the GPT Store for all OpenAI users.

AI Assistants from Peak Privacy can be shared with other team members via a sharing function. It should be noted that only team members have access to the AI Assistant. This is where the function differs from OpenAI’s GPTs, which can also be shared publicly.

---

My offers for the implementation of CustomGPTs and AI assistants

---

This article is also available as a podcast episode

Attention! The podcast was created entirely by my AI-Assistant based on my contribution – no guarantee for incorrect content.

Finally, a video on the data protection-compliant implementation of AI Assistants