Google's AI model Gemini faces shutdown calls after Elon/Hitler debacle
wionews.com
wionews.com
Most models do very verbosely state that they don't express opinions. Of course they do indirectly through their underlying datasets, but this answer fits the intent to not evaluate for destructiveness or evilness, even if everyone intuitively knows this assessment has one definite and correct answer. I bet you still find people that think differently though.
Every AI models will say something outrageous after being prompted with a loaded question.
My opinion is different than yours, and I think this quote from pmarca is highly relevant:
“I know it’s hard to believe, but Big Tech AI generates the output it does because it is precisely executing the specific ideological, radical, biased agenda of its creators. The apparently bizarre output is 100% intended. It is working as designed.”