Good question. SIAI hasn't been filming anything this week, so I could extrapolate from that (with low-moderate confidence) that they generally don't film the

Good question. SIAI hasn't been filming anything this week, so I could extrapolate from that (with low-moderate confidence) that they generally don't film the talks he gives. I'm not sure if NYU or Yeshiva filmed anything themselves, though if I had to guess I'd say no. If you're just wondering what to expect, Eliezer will probably talk about cognitive bias (specifically the biases which make it hard to think about existential risk), and then describe unfriendly AGI as an existential risk, and possibly follow with some of the solutions SIAI has for subproblems of building a friendly AI (e.g. timeless decision theory, reflective equilibrium (metaethics), etc.). No promises, but that's the kind of talk he seems to give these days.