Abstract: 

Artificial intelligence (AI) algorithms, especially machine learning (ML) programs, are now being employed or proposed for use in:

1. a) scanning résumés to weed out job applicants;
2. b) evaluating risks children face in their families;
3. c) informing judicial decisions about bail, sentencing, and parole;
4. d) diagnosing medical conditions, and not just classifying medical images;
5. e) identifying faces in the crowd for the police;
6. f) caring for seniors;
7. g) driving autonomous vehicles; and
8. h) guiding and directing drones in seeking to kill terrorists.

I will propose what society must require of algorithms that affect human welfare, health, life, and death.  I shall discuss concepts including reliability, openness, transparency, explainability, trustworthiness, responsibility, accountability, empathy, compassion, fairness, and justice.

My analysis will aid researchers in prioritizing problems for AI and HCI research.  It will also assist policy makers and citizens in determining when and how AI technology should be deployed.