The author just wrote a very poorly worded prompt. I asked it "Is the number of letters in the word ‘prime’ a prime number?" and it instantly got it correct. Maybe they patched the model or improved in the background? Who knows.
I don't know whether the model is being updated, but I do know that repeating the same query will change the seed and you'll get a completely different answer. E.g. you might go from a lengthy yes to a short no. Chatgpt is extremely noon-deterministic.