OpenAI's o1 model sat this year’s Korean SAT exam.
It got only one question wrong.
Top 4% score.The test requires reading comprehension, critical thinking, and logical reasoning to answer questions on the given passages (+images).
The topics cover a wide range of domains - e.g., science and literature.
It's not testing wikiIt was also written by professors who were locked in a hotel for a month - so o1 hasn’t seen the test before.