martin Randall

Aug 25, 2023 | lesswrong.com | Steven Byrnes |martin Randall |Stephen Bennett |Adam Zerner

Yes it’s possible to be more specific than “good faith” / “bad faith” but that doesn’t mean those phrases aren’t communicating something substantive and useful, right? By the same token, every possible word and phrase and sentence could be elaborated into something more specific. I also agree that there are edge cases, but again, that’s a near-universal property of using language to communicate. Here’s my defense of good faith / bad faith.

Jul 9, 2023 | lesswrong.com | martin Randall |David Hornbein

"Doomer" has become a common term to refer to people with pessimistic views about outcomes from AI. I claim this is not a helpful term on net, and generally will cause people to think less clearly. Reification of identity + making things tribalI finally realized today why politics and religion yield such uniquely useless discussions...

Apr 8, 2023 | lesswrong.com | Eliezer Yudkowsky |Nanda Ale |martin Randall |Charlie Steiner

(Related text posted to Twitter; this version is edited and has a more advanced final section.)Imagine yourself in a box, trying to predict the next word - assign as much probability mass to the next token as possible - for all the text on the Internet. Koan:  Is this a task whose difficulty caps out as human intelligence, or at the intelligence level of the smartest human who wrote any Internet text?  What factors make that task easier, or harder?