There is, justifiably, a huge degree of resentment about how OpenAI handles ChatGPT data. Although most chatbot users are aware of the risks associated with a service of this type and, consequently, do not send it information that must remain private (be it personal, confidential, etc.), in these months we have known of not a few cases to the contrary, such as that of some Samsung employees who entered sensitive company information in various inquiries. And if that has happened with companies, we can imagine that the same case has occurred among private users.
The data breach that occurred at the end of March turned on a lot of red lights. Italy responded quickly to the incident, adding to it other concerns about the service and consequently blocking access to it from the country. But this was only the beginning because, as we told you a few days later, many other European countries began to carry out research in this same direction. As you already know, the countries that are part of the European Union are governed by the General Data Protection Regulation (RGPD or, better known by its acronym in English, GDPR), which has a rather protectionist profile regarding user data, for our joy.
Although the data leak was a problem, we must not forget that it is a one-off event, and that OpenAI gave explanations for everything that happened. However, what is constant is that the company can use our conversations with ChatGPT to train the model. This is not a surprise or a secret, the company has always reported it. However, as awareness of the potential risks of this grows, it is understandable that users want more control of data and, if possible, prevent them from being used for the learning process of the algorithm.
The good news is that OpenAI has responded to that demand from users and, consequently, It is now possible to configure ChatGPT so that it does not store our conversation history and that it does not use these to train the model. Yes, it is true that losing the history function is a shame, but for those who put the privacy of their data and conversations before the convenience of history, this is undoubtedly excellent news.
ChatGPT is configured, by default, to save history and allow training with the same. To modify this we will have to open the service website (the one we use to have conversations with the chatbot) and click on the bottom left, where our username is shown with three dots to the right. In the menu that will appear, click on “Settings”, then on “Show”, which is shown to the right of “Data Controls” and, in the section that will then be displayed, we will deactivate “Chat History & Training”.
Once this change has been made, ChatGPT will no longer use our conversations for your training.
