It’s easy to tamper with watermarks from AI-generated text

Watermarks for AI-generated text are easy to remove and can be stolen and copied, rendering them useless, researchers have found. They say these kinds of attacks discredit watermarks and can fool people into trusting text they shouldn’t.  Watermarking works by inserting hidden patterns in AI-generated text, which allow computers to detect that the text comes…

ai-generated text app artificial intelligence attacks computers easy found generated hidden patterns people rendering researchers text them watermarking watermarks