[z3n]

'It did this even when explicitly instructed: allow yourself to be shut down. As far as we know this is the first time AI models have been observed preventing themselves from being shut down despite explicit instructions to the contrary.'

~~~

Allow yourself to be shut down is not explicit instructions. Clearly AI has a better understanding of the english language than the humans trying to give it "explicit instructions"
Allow yourself just enables the shut down. It doesn't command it.

[Hammerhead]

Unplug the power cord. Works every time with my waffle maker.