“GPT-4 Outperforms GPT-3.5 and Google Bard in Answering Sleep Medicine Board Exam Questions”

https://pubmed.ncbi.nlm.nih.gov/38117307

The abstract compares the performance of three artificial intelligence chatbots, GPT-3.5, GPT-4, and Google Bard, in answering sleep medicine certification board exam questions, and finds that GPT-4 performs significantly better than the other two.