There’s a personal project I’d like to do around this time next year that I think #AI would be able to help with a lot. The problem is, what I would use it for (scalable and wide-ranging research), hasn’t even worked for me at a small scale, hallucinating results that don’t match the sources it provides. So it might force me to spend more time fact checking the results it provides than I might have ever spent actually doing the research myself.