OpenAI has tools to watermark ChatGPT text, but doesn't use them ...
OpenAI has had a tool to watermark text generated by the company’s AI assistant ChatGPT for over a year, as reported by The Wall Street Journal and The Verge. OpenAI has also acknowledged the development of this tool in a blog post.
Internal Debate at OpenAI
Despite the assurance that the watermarking would not compromise the quality of ChatGPT's content creation, OpenAI has decided not to implement the tool. The company's internal discussions reveal conflicting opinions about its usage.
User Sentiment and Surveys
Studies conducted by the generative AI (genAI) company indicate a varied response to the implementation of watermarking technology. One survey suggests that four times as many individuals are in favor of tools that can detect AI-generated content rather than against it. Conversely, another survey highlights that 30% of ChatGPT users would potentially reduce their usage if the content produced was watermarked.
