Unenforceable rules are bad, but if you tweak the rule to always require some sort of authorship statement (e.g. "I wrote this by hand" or "I wrote this with Claude"), then the honor system will mostly achieve the desired goal of calibrating code review effort.
If it's any consolation, living in Blade Runner will be optional! You'll also have the option of living in full-dive VR where it's permanently 1999. No AI in sight, just print outs of MapQuest directions.
An LLM can trivially instruct someone to take medications with adverse interactions, steer a mental health crisis toward suicide, or make a compelling case that a particular ethnic group is the cause of your society's biggest problem so they should be eliminated. Words can't kill people, but words can definitely lead to deaths.
This isn't a Django problem. Just don't customize the admin. Ever. If the admin can't do what you want, pull out React and Django REST Framework and build something that does what you want.
Point 3 ("WEB SITES ARE DEEMED UNSAFE, EVEN IF FACEBOOK MONITORS THEM") has been addressed. Here are the other two.
"YOU CANNOT BRING YOUR CONTENT IN TO FACEBOOK"
False. Facebook's API allows all sorts of external content to enter Facebook. They're just shutting down their app that does that automatically. There are plenty of third party apps that already solve this problem.
"PUBLISHERS WHOSE CONTENT IS CAPTIVE ARE PRIVILEGED"
False. The Washington Post has chosen to embed their stories within the Facebook canvas pages, but that's not a requirement. The other popular news sites on Facebook, The Guardian and Yahoo, do not do this.