Designing life science assessments in the era of generative AI

journals.plos.org

1 pointoceansky1 comment

"ChatGPT earned a 66% average compared to 87% by doctoral students" That's with GPT-4o, not using reasoning models. Very interesting.