LLMs are solving MCAT, the bar test, SAT etc like they’re nothing. At this point their performance is super human. However they’ll often trip on super simple common sense questions, they’ll struggle with creative thinking.
Is this literally proof that standard tests are not a good measure of intelligence?
No. It’s the opposite in fact. It shows that ChatGPT is not very intelligent. Just very well-read.
It shows that it’s well-read but not that it isn’t intelligent. It says relatively little about its intelligence (although the tests do require some).