Surprisingly easy, it turned out. In a series of new research tests, Kovács found that a panel of human reviewers could not distinguish between reviews written by humans and those written by GPT-4, the LLM that powers the latest iteration of ChatGPT. In fact, they were more confident about the authenticity of reviews written by AI than they were about reviews written by humans.
Here’s the full story, by Sarah Jenislavski.
Source link