Report: “A Watermark for Chatbots Can Expose Text Written by an AI”

From MIT Technology Review:

Hidden patterns purposely buried in AI-generated texts could help identify them as such, allowing us to tell whether the words we’re reading are written by a human or not.

These “watermarks” are invisible to the human eye but let computers detect that the text probably comes from an AI system. If embedded in large language models, they could help prevent some of the problems that these models have already caused.

[Clip]

In studies, these watermarks have already been used to identify A[CI-generated text with near certainty. Researchers at the University of Maryland, for example, were able to spot text created by Meta’s open-source language model, OPT-6.7B, using a detection algorithm they built. The work is described in a paper that’s yet to be peer-reviewed, and the code will be available for free around February 15.

[Clip]

“Right now, it’s the Wild West,” says John Kirchenbauer, a researcher at the University of Maryland, who was involved in the watermarking work. He hopes watermarking tools might give AI-detection efforts the edge. The tool his team has developed could be adjusted to work with any AI language model that predicts the next word, he says.

Learn More, Read the Complete Article (about 760 words)

Filed under: News