Yeah I'm in favour of this also, but proving it might be tricky, and people will get unjustly banned.

Then there's the question of situations where people might actually use ChatGPT in a creative way to augment discussion.

Let's say we're talking about the pitfalls of repetitive code over breaking things cleanly into small functions. You have an example of this in mind that highlights a specific pitfall that you encountered, but don't want to share proprietary code, so you might ask ChatGPT (or a future model) to generate some code that demonstrates the same thing, rather than writing it yourself.

I think we're still early enough in the tech that it's hard to create hard-and-fast rules about what kind of content should be allowed; ideally, we'll get to the point where AI can help facilitate and augment human interactions, rather than taking over completely.