April 27, 2023 Joseph Thornton MD Tweet by npj Digital Medicine on detecting academic gpt. Poor sensitivity specificity npj Digital Medicine @npjDigitalMed Topical content… Blinded human reviewers only detected 68% of #ChatGPT generated scientific abstracts, despite incorrectly identifying 14% of real abstracts as being generated. nature.com/articles/s4174… pic.twitter.com/qc0EiP5qsP 4/27/23, 11:25 AM Joseph Thornton Share this: Share on X (Opens in new window) X Share on Facebook (Opens in new window) Facebook Like Loading... Related
Thornton Neuroethics 