Designing life science assessments in the era of generative AI (journals.plos.org) 1 points by oceansky 3mo ago ↗ HN
[–] oceansky 3mo ago ↗ "ChatGPT earned a 66% average compared to 87% by doctoral students" That's with GPT-4o, not using reasoning models. Very interesting.
1 comment
[ 2.9 ms ] story [ 11.3 ms ] thread