Alignment Network
So I'm shifting my focus on research and starting to set out a longer term strategy. To do this, I went into a rabbit hole to understand what it is I want to do, and how to go about doing it.
This project started with the idea of predicting what research top researchers would be publishing based on their networks and publication history. That was kind of a clever idea, but not realy what I wanted to do.
In the end I started a large literature review around AI alignment. This has resulted in two things. First, I've developed a network analyzer based on my .bib file which will let me see which articles are more important (in quotes), as well as which authors are connected.
This clearly shows the AI lab cluster of authors - which is actually what I was originally going for. But also highlights some interesting research and some gaps! Which is what I was also going for...
I've also realized I'm not getting what I want out of this 100-days-of-making-ai project. Or at least I'm starting to evolve my thinking. After 50 days, I plan to spend the next 50 days reading and blogging about a one research article in the AI alignment/safty space. I think this is what I really want to get out of this effort!