I appreciate the optimism, and I sure hope you’re right. It took me a while to understand and accept the importance of x-risk mitigation, especially for AI, so I should really be more empathetic. And as others have noted, Santos’ presentation wasn’t great in terms of his accuracy and phrasing, so I would love to see people use this speech as a starting point to seek out more accurate assessments of the risks he discussed.
SWK
Karma: 95
AI-Relevant Regulation: CPSC
AI-Relevant Regulation: Insurance in Safety-Critical Industries
AI-Relevant Regulation: CERN
AI-Relevant Regulation: IAEA
Very fair points. I guess since I agreed with the general gist of Santos’ message (and was generally just surprised that he was talking about this stuff in the first place) I looked past the presentation at the time. Looking back, Santos definitely used some language that makes his arguments more difficult to take seriously.
X-risk discussion in a college commencement speech
Thanks for sharing this post, I absolutely agree. Hopefully critics of EA can come to see the genuinely warm-hearted motivations that I think most EAs have.
Thank you for this perspective! Once again, I think you’ve expressed better than I did the connection between humor and humility. I love all of your historical examples as well; it has me thinking that, rather than following current examples of comedy, looking further back in the past might be an even more fruitful approach to getting inspiration for EA-applicable humor that has stood the test of time.
Thanks so much for your insights. Can’t really argue with what you say here: I think you articulated the idea of subtlety and the importance of correct application with humor far better than I did. Admittedly, John Oliver is an extreme example of humor, perhaps so extreme as to be unhelpful as a model for EA. Overall, maybe my use of the word “humor” in this post was too strong. I really liked Tiger Lava Lamp’s comment below on “microhumor,” which Scott Alexander describes as “things that aren’t a joke in the laugh-out-loud told-by-a-comedian sense, but still put the tiniest ghost of a smile on your reader’s face while they’re skimming through them.” This seems to be a more accurate description of the MacAskill and Karnosfky examples I gave. It seems like we both have a sense that something like Alexander’s microhumor can fall within EA’s humor threshold and be an effective tool for EA to an extent.
Love this example, thanks so much for sharing it. I know I mention John Oliver above, but realistically his style is almost certainly too extreme to be replicated in most cases by most people. I agree that Alexander’s microhumor is a perfect example of subtle humor that could potentially be employed in almost any context.
Wow, can’t believe I haven’t seen this before. Thanks for sharing!
Thanks for your comment, and I agree with your points. I definitely should have noted Santos’ unfortunate use of the Doomsday Clock to frame his speech. This was, in hindsight, a sure way undermine his message both for people familiar or unacquainted with x-risk ideas.