Search
In September 2026, OpenAI's confidential initiative, known as "Project Lily," came to light. This project involved the use of prompt moderators who manually reviewed anonymized user conversations with ChatGPT, aiming to enhance the quality of the model's responses. These moderators were compensated at a rate exceeding $50 per hour. Despite the anonymization of the data, OpenAI acknowledged that there remained a potential risk of personal data exposure, especially in brief exchanges. By default, users authorized the use of their data for model refinement, and the option to revoke this authorization did not apply retroactively, sparking concerns over privacy. Furthermore, the implementation of human review processes within AI firms is more prevalent than one might assume.
-
C114 Communication Network -
Communication Home
