1 Sep 14 5:19 PM · 13d ago · 2 posts · 1 source · development 1 of 1
Hugh Zhang posted on X in agreement with Daniel Selsam about deteriorating ability to verify AI model alignment.
“We are rapidly losing the ability to tell whether our models are aligned.”
Hugh Zhang AI researcherDaniel Selsam AI researcher
I agree with Daniel Selsam 100%. We are rapidly losing the ability to tell whether our models are aligned.
All 1 developments of AI researchers warn of deteriorating model alignment… →