Raising the level of abstraction can greatly reduce tedium, and can make code a lot easier to grok.

Introducing LLM generated code doesn't do that in my experience.

Raising the level of abstraction has significant costs though. Anyone who has built a large or complex enough system becomes very wary of abstraction.

I think this is one of the major benefits of LLMs. It's far less tedious to repeat yourself and write boilerplate when doing so is a better engineering decision than adding more layers of abstraction.