Okay, I need your help! I thought I'd seen a paper shared that showed that people who use #LLMs become quite accurate in recognizing AI generated text after all. Does this ring a bell?
I seem to remember 90% accuracy and even a list of tells they got right/wrong.
But my usual ways of findings things again are failing me to the point that I worry this might be a false memory. Or perhaps I saw a misrepresented summary of this study with novice and experienced teachers? https://www.sciencedirect.com/science/article/pii/S2666920X24000109 🆘
