A little off-topic: are LLMs the death knell for new languages, frameworks, tools, processes, etc? I can see how an LLM is going to be such a huge productivity

A little off-topic: are LLMs the death knell for new languages, frameworks, tools, processes, etc? I can see how an LLM is going to be such a huge productivity boost that they’ll be hard to avoid everywhere, but then new stuff won’t have any training data. Will anyone ever go through the effort of everything being 10x - 100x less effective with new tools since there’s no training data?