I think in little time LLM will have the capability to translate easily code from one computer language to another, for example the LLM could "think" in prolog + lisp instead python. LLM will be able to write thousands of lines of APL or J in just a minute, we will have to accept that python is not the language for future LLMs. Perhaps the power of macros will not be defeated by its complexity because for an LLM a thousand line macro is just another simple code, prolog unification could mean building slots to construct new powerful artifacts that we are not able to have in our minds at the moment. LLM will no longer learn natural language or our most used computer languages, the next step is designing languages for a system that can keep in his head millions lines of code.

I think that in the near future a super object oriented language will emerge that will let us create computer apps instead of programming them directly. Systems analysts, not programmers, will be able to define a computer app with the objects at a very high level and LLMs will create a JAVA like blob that can be run on multiple systems. The advantage is that the objects will be completely bug free and the app size is only limited by the systems resources. Bottom line, most app development will be done by systems analyst. The AI object programmers, a specialty, will help tweak the underlying function of each object.

what llms reason best in might not be a language at all. more like a graph. graphs say more in less space than prose, and they stop the model from wandering off.

Why not assembly or machine code directly?