What will Manifolders mostly use LLMs for, by EOY 2025?

What LLM API will work unmodified until the end of 2025?

By 2025 end, will it be generally agreed upon that LLM produced text/code > human text/code for training LLMs?

By 2027, will it be generally agreed upon that LLM produced text > human text for training LLMs?

Voice control of Cursor + LLM interactions by mid 2025

Will prompting of generalist models be the dominant method of interaction with LLMs on 1 January 2028?

Will there be a free, public way to generate LLM text that evades jan2024 llm detector 'binoculars' by the end of 2024?

In 2030, will most human-computer interactions happen through a LLM-interface?

By 2029 end, will it be generally agreed upon that LLM produced text/code > human text/code for training LLMs?

Will we have a popular LLM fine-tuned on people's personal texts by June 1, 2025?