General LLMs like ChatGPT, Bard, etc. are all the rage right now. However, where do you see the industry heading towards from your perspective and experience in the next couple of years?
Do you see them:
(1) Being modular - where multiple niche LLMs are available across verticals (similar to Harvey for law) which are used in tandem with a base model which is trained on general knowledge of conversations?
(2) A very large general model that is trained across multiple categories (jack of all trades-kinda LLMs)
(3) Something completely different from the above perspectives
Would love to learn :)
