OpenAI gives ChatGPT users option to turn off chat history
OpenAI gives users the option to turn off their chat history in ChatGPT. Conversations are therefore not used for training OpenAI’s AI models. Chats are permanently deleted from company systems after 30 days.
OpenAI sets the new privacy option available to users from Tuesday. Users can use this to indicate in the ChatGPT settings that their chat history should not be saved. When users do that, their chat conversations are not used to train the AI models that OpenAI creates. Conversations with the chatbot no longer appear in the sidebar of the service.
New conversations are saved for up to 30 days if chat history is turned off, according to OpenAI. According to the company, those conversations are only viewed if it is necessary to check them for abuse. After 30 days, the conversations are permanently deleted. There will also be an option that will make it easier for users to export their ChatGPT data. Users who do so will receive a file containing chat conversations and ‘all other relevant data’ sent to their email address.
OpenAI says it is continuing to work on a ChatGPT Business subscription. That service would be intended for professionals who “need more control over their data”, in addition to companies who want to manage their end users. By default, data from ChatGPT Business is not used to train OpenAI’s models. The new subscription should come out ‘in the coming months’.
The new options follow privacy concerns surrounding the service in a number of countries. ChatGPT is currently being investigated in France, Ireland and Germany, among others. Privacy regulators in Europe are investigating whether OpenAI processes user data without proper permission, which is in violation of the GDPR. By default, conversations with ChatGPT are used by OpenAI to improve their models.
Italy banned ChatGPT last month. The Italian then said ChatGPT could become available again in the country if OpenAI gave consumers the option to object to their data being processed, writes Reuters. OpenAI Cto Mira Murati tells that news agency that the new privacy options were not introduced in response to the ban in Italy. The company says it has spent months giving users more control over data collection.
Source: Open AI