The simple answer is no. However, it's naive to deny that people are driven to technology with large amounts of samples and documentation. When trained on it,

The simple answer is no. However, it's naive to deny that people are driven to technology with large amounts of samples and documentation. When trained on it, LLMs can essentially create infinite, customized versions of these. LLMs artificially inflate the amount of documentation and samples for older technologies, and many lack the knowledge of newer ones. This creates a feedback loop until the newer technology is included, but the damage has already been done. That's just a perspective, I don't know that I fully agree with the premise, and certainly don't know a solution. The best solution I can think of is more LLMs that are essentially constantly RAGing from the internet, rather than mostly using pre-trained info. Obviously this part is no longer just the model, but I imagine models could be made to make the RAG process more efficient, or better.