Everyone's focusing on the big alignment problem in AI (making sure AI in general helps society in general), but not enough people are focusing on the small alignment problem (making sure your AI is working in your interest). The small alignment problem is just as important.