💻
💻 Technology
OpenAI hires people to read private ChatGPT conversations
According to 404 Media's investigation, OpenAI is running an internal project codenamed "Project Lily" in which employees review ChatGPT user conversations and evaluate model responses. They check naturalness, helpfulness, and whether the AI sounds too flattering or "too AI." Reviewers don't see usernames, and conversations pass through a filter removing personal data — but the system is not foolproof and private information has reached staff. Similar practices are used by Google, Anthropic, and Chinese AI labs.
Comments
No comments yet
Comments
No comments yet — be the first to weigh in 👇
No comments yet. Be the first!