OpenAI can finally stop saving deleted ChatGPT conversations

OpenAI has received a significant legal reprieve, allowing the company to cease the preservation of user conversations with its ChatGPT interface. This development follows a recent court order that lifted a previous 'preservation order' imposed during an ongoing lawsuit involving the New York Times and several other media outlets. The new directive, issued by U.S. Magistrate Judge Ona Wang, permits OpenAI and the news organizations to jointly request the termination of the preservation order. As a result, OpenAI is no longer required to retain and protect all output log data that would typically be deleted moving forward. While the preservation order has been lifted, the news organizations will still have access to all temporary and previously deleted chats that were stored under the order until September 26th. Additionally, reports indicate that OpenAI will continue to monitor certain ChatGPT accounts, retaining deleted and temporary chats for users flagged by the news organizations during their data examination. This legal battle began in December 2023 when the New York Times and other publishers accused OpenAI and its primary investor, Microsoft, of employing their copyrighted content to train AI models without permission. OpenAI contested these claims, describing the Times' actions as an attempt at 'mass surveillance' over user data while raising concerns about user privacy. However, the court ruled in favor of retaining user data, including standard, temporary, and deleted conversations. The Times argued that the logs would provide evidence that ChatGPT was trained on its copyrighted articles, with outputs often mirroring or closely resembling its content. In response to the copyright allegations, OpenAI has frequently referenced the 'fair use' doctrine, asserting its right to utilize limited, unlicensed copyrighted material for purposes of research and commentary.