This is a pretty damning take and I can’t really think of a way to argue against it. So first off, thank you for that perspective!

I’ll probably have to think about it some more, but a) I think I’m inclined to take my statement above back, in that I/we should definitely still aim for understanding, even if it’s not the most “productive” thing to do, and b) I don’t know what’s a “realist” way forward, i.e. not completely denying LLMs are capable tools and never using them, but still understanding what comes out when you do use them. Seems rather difficult to balance the way the current iteration of tools is designed to be used.

[dead]