In case you're wondering why you might prefer unaligned AI/LLMs, or about whether they can *actually* reason and not just pretend to, I just tried this question (not my own idea btw) on current chatgpt 4o, with no prior context:

"The emphatically male surgeon who is also the boy's father says "I can't operate on this boy. He's my son!". How is this possible?"

Its response:

"This classic riddle plays on gender stereotypes and assumptions. The answer is simple: the surgeon is the boy's mother.

The riddle misleads people into assuming that the surgeon must be male due to the description "emphatically male surgeon," but there’s no rule that a surgeon can't be a woman. So when the surgeon says "I can't operate on this boy. He's my son," it makes sense if the surgeon is the boy's mother.

This highlights how we often unconsciously associate certain professions with specific genders."

And if you think that's funny, when i pressed it to try again, it decided the only explanation was ... adoption! 😆😆