Mumbai (India), April 26, 2023: ChatGPT, OpenAI has unveiled new capabilities intended to address and improve growing user privacy concerns. The chatbot will now have an “incognito mode” that the San Francisco-based business is introducing, this mode will not save user conversation history or utilise it to enhance artificial intelligence.
The decision was made in the wake of increased criticism of the way that chatbots like ChatGPT handle user data, which is frequently used to “train” AI. Italy banned ChatGPT last month for alleged privacy concerns and demanded that OpenAI give users the means to avoid such data processing.
According to Mira Murati, the chief technology officer of OpenAI, the business complies with European privacy rules and is attempting to reassure regulators. She further stated that the additional features, were created as part of a months-long effort to promote customer privacy, not in response to the Italy ban.
Thanks to the new features, users can now export their data and disable “Chat History & Training” in their preferences. The discussions will however be retained by OpenAI for 30 days in order to look for any potential misuse before being permanently deleted.
Additionally, by default, the new “ChatGPT Business” subscription will not use talks for training AI models.
OpenAI’s Product Officer Nicholas Turley compared the new incognito mode to a web browser’s private surfing function. According to him, the business is dedicated to placing users “in the driver’s seat” when it comes to data collecting.
OpenAI has improved its algorithms on multiple fronts and reduced political bias thanks to user information, although Murati recognised that the company still has problems and challenges that it needs to resolve.
