Can you explain the failure?
2) when it does tell a specific thing it doesn't know, it's a thing it actually knows well
this train of conversation looks monkeypatched by openai bc it reveals a massive vulnerability in reasoning, but is consistently reproducible across all llms