FWIW, there seems to be a trend in my corner of the AI safety community (CHAI) to move away from the term ‘values’ and towards the term ‘preferences’, I think for similar reasons.
rosiecam
Karma: 199
FWIW, there seems to be a trend in my corner of the AI safety community (CHAI) to move away from the term ‘values’ and towards the term ‘preferences’, I think for similar reasons.
This is great, I can see it being really helpful for me to consciously think about which of these I’m optimizing for (or am willing to sacrifice) when writing. I got confused by the introduction of the term ‘density’ in the section on trade-offs, as this isn’t represented in the RAIN framework. Is density just a sub-consideration of accessibility or are you considering it in its own right?