The Turing Test has been replaced by the original genius test now. - oh well
" . . . I thought I didn’t use ChatGPT because I was too clever. I thought that not using ChatGPT made me cleverer. It turns out, though, that it made me very bad at spotting when a text was written by or with the assistance of AI. After the uproar over the Commonwealth Prize, I took a New York Timesquiz entitled ‘Who’s a better writer: AI or humans?’ I got three out of five correct – barely better than a coin toss. On Wikipedia’s ‘AI or not’ quiz, I got seven out of ten, but that was easier because none of the AI articles had footnotes.
I’m not the only Chat non-user who can’t tell when an LLM wrote something. Experimenters in the US last year showed nine subjects a series of articles, half written by humans and half generated by ChatGPT, Claude and other large language models. Asked to guess which of the texts were human, the four subjects who rarely or never used ChatGPT in their daily lives scored ‘at a similar rate to random chance’, while the five who used chatbots almost every day at work collectively misidentified only one in three hundred texts. . . " LRB
Red Testing original genius - is that wise?
No comments:
Post a Comment