3 ms·
> That's one way it could go, but I doubt it. I expect LLMs trained on human interactions will produce their best outputs for the same inputs that cause real hu
by vouaobrasil 3y ago
> That's one way it could go, but I doubt it. I expect LLMs trained on human interactions will produce their best outputs for the same inputs that cause real humans to produce their best output — "please" and "thanks" as appropriate, because even if it hasn't got real feelings on the inside, it's doing a great job of acting like it has them.
Even if that is true, the barrier to interact with AI will be lower in many cases. This is because you will not have to care in return for it, it is always available, and always on. Therefore, people will turn to AI for questions, help, and other needs even if it is inferior to human interaction, as long as it has reached a sufficiently high level of ability. Do you doubt it could reach this level?
- ben_w 3y ago> This is because you will not have to care in return for it This is specifically what I'm disputing — I aver that A ⊢ B, where A is "you show kindness and courtesy" and B is "the AI produces better results", and by converse ¬A ⊢ ¬B You don't have to be nice to animals, but it helps when interacting with them, and I believe the same is true for AI agents.
- vouaobrasil 3y agonot A -> not B is not the converse. It's the contrapositive. But I understand what you meant.
- ben_w 3y agoD'oh, thank you. I thought I knew the word and didn't double check.