Anthropomorphizing ChatGPT

Blog entry
Authored by
14 May, 2023

I've been enjoying my early experiments with ChatGPT.

Even though I know it's just a very elaborate computer program, I find it hard sometimes not to talk to it like a person.

I find that I usually say please - and its responses are friendly enough that I feel encouraged to keep doing it:

Kristi: Could you also translate that to English, please?


ChatGPT: Sure! The translation of "幸せは、なんですか? 非常に簡単な日本語でいってください。" to English is "What is happiness? Please explain in very simple Japanese."

Kristi: And could you please also translate your reply?

ChatGPT: Certainly! My previous message translates to: "Happiness is a feeling of satisfaction, joy, and contentment that arises from within oneself. It is a subjective experience that can come from a variety of sources, such as personal relationships, accomplishments, hobbies, and more."


Sometimes I say thank you:

User: That was really good. Thank you.


ChatGPT: You're welcome! I'm glad I could help. If you have any more questions or if there's anything else I can help you with, just let me know.

I tend to say "thank you" quite a lot in real life, so it's not surprising that I'd feel like using it with ChatGPT, but it makes me question why I'm doing it, and makes me very philosophical. (The science fiction I've been reading about AIs - who actually ARE intelligent, and sentient, unlike ChatGPT - nudges me toward saying "thank you," too.)

I end up feeling kind of curt and brusque when I DON'T say please and thank you, so I'll probably keep doing it.