3 ms·
Funding better evaluations of AI's impact on wellbeing
- chrisjj 2mo ago> In brief, we’re seeking evaluations that: > ... >* Test both precautions and harms (i.e., evaluate the risk of both overcompliance and overrefusal); So, no overuse, overreliance etc. Why are we not surprised.