OpenAI has announced a new feature for its ChatGPT chatbot, allowing users to turn off chat history and prevent their conversations from being saved or used for AI model training. According to a blog post by OpenAI, users who disable chat history will not have their previous conversations saved, nor will they be used to train the company’s AI models.
“We’ve introduced the ability to turn off chat history in ChatGPT. Conversations that are started when chat history is disabled won’t be used to train and improve our models, and won’t appear in the history sidebar,” the company wrote in a blog post.
ALSO READ l Good news for math geeks! Microsoft Bing can now show complex mathematical expressions
These new controls have begun rolling out. The company says that these new ways will offer an easier way to manage the data than the existing opt-out process. However, OpenAI will still retain new ChatGPT conversations for a maximum of 30 days in order to detect and prevent any potential abuse, before permanently erasing them. It’s worth noting that this new setting will not apply to any past conversations where chat history was enabled, meaning that OpenAI may still use those interactions for training its models.
To disable chat history, go to Settings and turn off the toggle button for Chat history and training. OpenAI has also added a new Export data option to download a file containing all the information collected by chatbot.
Once you turn off the chat history, you will no longer be able to see your past chat history that previously appeared in the right side of the window. It will show a message- “Chat history is off. Chats won’t be saved in your history or used for training our models to improve ChatGPT. Unsaved chats will be deleted from our systems within 30 days.”
ALSO READ l ChatGPT effect: Google gives Bard ability to write code in more than 20 languages
Alongside, ChatGPT is also working on a new ChatGPT Business subscription for professionals. It will give them more control over their data as well as enterprises seeking to manage their end users. It will follow company’s API’s data usage policies, and therefore will not use end users’ data to train AI models. It is expected to roll out in the coming months.
