I think once we get off LLM's and find something that more closely maps to how humans think, which is still not known afaik. So either never or once the brain is figured out.

I'd agree that LLMs are a dead end to AGI, but I don't think that AI needs to mirror our own brains very closely to work. It'd be really helpful to know how our brains work if we wanted to replicate them, but it's possible that we could find a solution for AI that is entirely different from human brains while still having the ability to truly think/learn for itself.

This is where it’s a mistake to conflate sentience and intelligence. We don’t need to figure out sentience, just intelligence.