while this : "The easiest way to fix this is to improve the quality of life of people in the US." sounds good. I think the practical challenge will be attribution. If the application of AI across society does result in the improvement of the quality of life across people in the US (and other countries) how will people realistically attribute it to AI and therefore reduce the AI trust crisis?

I think the answer is: people have to help improve their own lives, and when AI can help them do that legitimately, then I think they will.