Almost a year after the launch of ChatGPT, OpenAI continues to surprise us with new features and capabilities of its platform.
The latest OpenAI release note actually provides a cool new feature for creating your own custom version of ChatGPT, called GPT.
GPTs are a way to increase your productivity even more. If you’ve been a ChatGPT user, you may have already identified a long list of use cases that will help you in your daily activities: it could be optimizing your tweets, helping you generate ideas for articles, generating variations of a sales pitch, reviewing your code or help you learn Esperanto. GPTs come into play to help you create a “specialized” version of ChatGPT, customized to your needs and very good at one specific thing, rather than being very generalist.
This opens up a new range of possibilities. In this article, we will, step by step, create a new custom GPT and then discuss some limitations of this new feature.
Previous requirements
The prerequisites are pretty simple, you need:
- A ChatGPT Plus license or an Enterprise account (this basically means you can’t create GPT using a free account).
- And an idea for a new GPT (think about your top use cases or ask ChatGPT to help you generate ideas). Here are some use cases shared by OpenAI to give some examples:
The use case
For this article, I wanted to work on a real use case, coming from a problem I faced in the past.
In a previous professional experience, I had to manage GDPR issues facing the client. The GDPR, or General Data Protection Regulation, is a European Union (EU) law on data protection and privacy. Its goal is to give people more control over their personal information and how it is used. However, its implementation can be complicated. The text of the law is more than 150 pages long and is quite technical. As a technician, it can be challenging to navigate legal and regulatory issues. At the time, ChatGPT didn’t exist, so I had to upskill myself and enlist the help of an outside legal expert to support me on the journey (which also increased costs), until I had more autonomy. But even then I was always faced with new issues like: tackling data…
