Everything violates content policies

Everything violates content policies
I got mad at chatGPT because it gave me dumb answers so I told him “Take a trip to France.” He gave me a weird answer. "Take a trip to France so sixty-six million people can destroy you.
That somehow violated content policies.
Then I said,
“Take a trip to Baton Rouge so a thousand pelicans can permanently destroy you” (Also violated content policies)
This is just wrong if chatgpd can even handle some well you know

Hi!

Can you explain to me what the part with the destroying is all about?
It sounds offensive but maybe there is a meaning I am missing?

It is idiotic. It censor it’s own answer. And I’m not writing porn, I don’t ask for how to make a bomb. I do however write about serious social issues. Like human trafficking. Underage sex exploitation, and I do this to raise awareness, and I should be able to talk about more than your average 10 year old can handle. It understand context. Why is this so hard? Put a age confirmation thing on it. Give me the chance to lower sensitivity on the bots I make and only shave with those that have the link.