Shhhhh - don't you know that using old models (or in this case, what likely wasn't even a LLM at all) to get wrong answers and make it look like AI advancements are overblown is the trendy thing these days?
Don't ruin it with your "actually, this is misinformation" technicalities, dude.