I'm working with the garage door up as I learn how best to integrate AI with MyHub.ai.
(This is version 1 of this homepage, published on my experimental wiki using the permanent versions pattern, where I publish a string of versions as I develop my thoughts in the hope of collecting constructive criticism and ideas along the way. Version control is in the footer.)
Adding an LLM to MyHub.ai has been the idea from the outset (the hint's in the ".ai"), with the longterm goal being "creating a decentralised, open-source ecosystem ... developing cooperatively-owned AI engines to benefit society as a whole" (How Artificial Intelligence will finance Collective Intelligence).
But as anyone who has played with any LLM knows, getting the best out of them is not easy. You need to try stuff out, compare results and try some more. You need, in other words, to experiment.
The quickest way to experiment with integrating an LLM into MyHub.ai was to integrate ChatGPT with it and let MyHub.ai Early Adopters play with it (see ChatGPT integration free trial, Bullshit, Botshit and Bubbles), so I thought I'd share my research as I go. And given that I manage my research notes in Obsidian, sharing them here via a massive wiki was the obvious solution (read more about this massive wiki).
This is one of this wiki's pages managed with the permanent versions pattern described in Two wiki authors and a blogger walk into a bar…