I've spent a lot of time asking myself what I can do to reduce AI risk. I think one approach is to produce videos explaining the simple aspects of AI risk for laypeople to understand. I'm looking for feedback on this first video I made: https://www.youtube.com/watch?v=BygErhGbONA

About a month ago, Yann LeCun wrote a tweet that inspired me to argue with him. Yann has what I would describe as an almost child-like optimism around AI risk. He managed to pack so many (in my opinion) wrong things into one tweet, I felt compelled.

I also tried to transcribe it to give everyone a readable option if they didn't want to watch, but after over an hour of editing the YouTube transcript, I felt I couldn't capture all the nuance (and "uhs" and "ums") that way, so here's the plain video.

Please help me by giving me critical feedback on what I could do differently in a future video!