Yesterday I decided to see what world problems remain that can be solved with AI. Of course, the clanker found many and honed in on one. Instead of telling it what to do, I asked what kinds of questions should I be asking and what are the answers to them? I did this many times over and eventually it narrowed things down to specific actions. It then downloaded 500 scientific papers and started cleaning data, turning it into a knowledge graph and doing all sorts of things. Then it started running experiments. 🔬

Currently it’s testing past outcomes to see if it can predict future ones regarding a very specific field. Depending on the experiment outcomes itll expand to more scientific literature (100s of thousands of papers).

This current set of experiments is not all that interesting to me but the process is.

Personally, id rather work on material science in hopes of discovering new materials to make science fiction a reality.

Anyway, it was one evening thought that turned into hours of probing an idea.

Curious to see where it leads.