The issue seems to be the underlying code tells the ai if some data set has too many white people or men, Nazis, ancient Vikings, Popes, Rockwell paintings, etc then make them diverse races and genders.

What do we want from these AIs? Facts, even if they might be offensive? Or facts as we wish they would be for a nicer world?

That’s what I don’t understand. The item is using up warehouse space, marked for me, for five days. Surely they’d make more money just mailing it and restocking the item to sell more during those five days.

Their next story: You cannot live without avocado toast, it’s a fact.

Damn, and I’m fresh out of I’m sorry, I cannot fulfil this request as it goes against OpenAI use policy.

Sounds just like old Twitter. Ban someone who you don’t like politically. If it’s causing to much uproar, day it was an accident/glitch. If it works, repeat. If not, repeat in a few weeks.