Ok. I'm responding to the statement that these are "relatively clear concepts". AI will by definition never have "the kind of sentience that living, awake humans have" (and I do acknowledge that this is only a working definition) because the goalposts will continue to move according to whim in the absence of a more formal definition. I remember when one prevailing formal test of self-awareness was "independently claiming to have self-awareness". There were caveats, of course, but that goalpost has evidently moved.