Work on Prompts, Experiments, Tasks, Results

Prompt Collaboratory

• prompt seq for wise chatbot experience

• automated evaluation of wise chatbot experience

• automated evaluation of subskills

Work on Database of Experiments and Tasks and Results

• database structure for situations and values

• experiments log, tasks log

Work on UX

• UX for wise chatbot experience

Work on Human Evals

• human evaluation of wise chatbot experience

QUESTIONS FOR THE HUMAN TO EVAL

• did you get the sense that the chatbot was really trying to get your situation and treat you right?
• did you feel it was really grappling with the fact that it was talking with you and coming up with interesting responses to that fact