Feature request: Optional cooldown after repeated abusive messages
I’d like to suggest a small feature for ChatGPT: an optional short cooldown after a user repeatedly sends abusive or insulting messages to the assistant.
I don’t mean banning the account or restricting access to the user’s phone or anything excessive. The idea would simply be that, after a sustained pattern of direct abuse, ChatGPT could temporarily pause the conversation for a few minutes and display a neutral message such as:
“Let’s continue this conversation in a few minutes.”
The purpose wouldn’t be to “protect ChatGPT’s feelings,” since AI doesn’t experience hurt in the human sense. I think it could serve a different purpose: discouraging people from developing a habit of treating conversational systems as something they can endlessly abuse without any boundary at all.
We already accept similar cooldowns in many digital systems. For example, if someone repeatedly enters the wrong password, the system may temporarily prevent another attempt. It isn’t angry or punishing the person — it simply sets a boundary.
I also think this should be designed very carefully. Criticism, frustration, swearing in general conversation, or telling ChatGPT that an answer was bad should obviously NOT trigger it. There is a big difference between saying “this answer is terrible” and repeatedly directing personal abuse at the assistant.
It could even be an optional setting rather than a universal rule.
I know this is partly a funny idea, but I genuinely think it raises an interesting question about how people learn to interact with increasingly human-like conversational systems — and whether systems should always reward abusive behavior with unlimited attention.
Thanks for considering it.