7 ms·
With the right series of prompts and criteria ChatGPT could probably whip up an effective virtual nanny.
by Teamancer 3y ago
With the right series of prompts and criteria ChatGPT could probably whip up an effective virtual nanny.
- rajlego 3y agoI've thought about this but the problem is it has to be able to poke me in a way I can't ignore It seems plausible to add penalties of some sort that make me take it seriously but hard for me to align it in a way that handles edge cases well