OpenAI is facing a legal challenge from The New York Times, which is demanding access to 20 million private ChatGPT conversations. The Times claims this information could help them investigate users attempting to bypass their paywall. OpenAI argues that this request violates privacy protections and does not relate to the lawsuit’s core issues. They emphasize that releasing such data would risk exposing highly personal conversations belonging to users with no ties to the case. To protect user privacy, OpenAI is developing advanced security features, including client-side encryption for ChatGPT messages. They aim to ensure that only significant safety concerns are escalated to human reviewers. OpenAI has offered The Times alternative options, like targeted searches for specific content, but these were rejected. They maintain that conversations are currently stored securely and will only be accessible to a limited number of legal and security team members. OpenAI is actively contesting the data request, arguing it represents an overreach. They also highlight that The Times originally sought access to 1.4 billion conversations, which was significantly reduced after pushback. The company is committed to safeguarding user privacy and will fight against any attempts to disclose data publicly, despite the legal obligations they must follow.