Less about complexity than about how well-documented the subject is on the internet IMO. I’ve been using it to help me set up and troubleshoot AWS Elastic Kubernetes clusters, which are plenty complex, and I’d estimate it’s been around 95% accurate. (And for at least one of the times it seemed to be wrong, it turned out I’d made a mistake following its instructions...)
You could ask me a difficult scientific question which I wouldn't even understand. But I could google and find a scientific paper which I would pass to you. You could say fantastic answer, thanks. But I would have no clue as to whether it is or is not. Now if I could just do that fast enough to serve all the people all the time, you would call me a sensation. I think this is what's happening with LLMs.
