OpenAI possesses resources to embed a watermark in ChatGPT text, yet refrains from their application

For over a year, OpenAI has had access to a feature to implant a watermark in text produced by their artificial intelligence assistant, ChatGPT, as revealed to The Wall Street Journal via The Verge. OpenAI has acknowledged the development of such a feature in a recent blog post.

As per reports from The Journal, the watermarking of text by ChatGPT is indicated not to compromise the caliber of the generated content. However, OpenAI has consciously opted against deploying this feature due to internal divergence pertaining to its utilization.

Allegedly, the artificial intelligence firm conducted a research study that exhibited a considerable preference for tools capable of authenticating AI-generated content compared to those opposed. Simultaneously, another survey disclosed that 30% of ChatGPT users would decrease their usage if the content was marked with a watermark.

About Author

AndyC

Andy Curtis is an award-winning security consultant, researcher and public speaker. He has been working in the computer security industry since the early 1990s, having been employed by state and federal government, leading healthcare and banking providers across three continents. He has given talks about computer security for some of the world’s largest companies, worked with law enforcement agencies on investigations into hacking groups, and is a regular voice on TV and radio explaining IT security threats.

See author's posts