Also in early July 2019, Jessica published ["The AI Timelines Scam"](https://www.lesswrong.com/posts/KnQs55tjxWopCzKsk/the-ai-timelines-scam), arguing that the recent popularity of "short" (_e.g._, 2030) AI timelines was better explained by political factors, rather than any technical arguments: just as in previous decades, people had incentives to bluff and exaggerate about the imminence of AGI in order to attract resources to their own project.
-(Remember, this was 2019. After seeing what GPT-3/PaLM/DALL-E/_&c._ could do during the "long May 2020", it's now looking to me like the short-timelines people had better intuitions than Jessica gave them credit for. I won't say, "How could I have known?", but at the time, I didn't, actually, know.)
+(Remember, this was 2019. After seeing what GPT-3/PaLM/DALL-E/_&c._ could do during the ["long May 2020"](https://twitter.com/MichaelTrazzi/status/1635871679133130752), it's now looking to me like the short-timelines people had better intuitions than Jessica gave them credit for. I won't say, "How could I have known?", but at the time, I didn't, actually, know.)
I still sympathized with the "mainstream" pushback against using "scam"/"fraud"/"lie"/_&c._ language to include motivated [elephant-in-the-brain](https://en.wikipedia.org/wiki/The_Elephant_in_the_Brain)-like distortions. I conceded that this was a _boring_ semantic argument, but I feared that until we invented better linguistic technology, the _boring_ semantic argument was going to _continue_ sucking up discussion bandwidth with others when it didn't need to.
Ben replied that what mattered in the determination of manslaughter _vs._ murder was whether there was long-horizon optimization power towards the outcome of someone's death; it wasn't about what sentiments the killer rehearsed in their working memory.
-On a phone call, Michael made an analogy between EA and Catholicism. The Pope was fraudulent, because the legitimacy of the Pope's position (and his claims to power and resources) rested on the pretense that he had a direct relationship with God, which wasn't true, and the Pope _knew_ (he _had_ to know, at some level) that it asn't true. (I agreed that this usage of _fraud_ made sense to me.) In Michael's view, Ben's charges against GiveWell were similar: GiveWell's legitimacy rested on the pretense that they were making decisions based on numbers, and they had to know at some level that they weren't doing that.
+On a phone call, Michael made an analogy between EA and Catholicism. The Pope was fraudulent, because the legitimacy of the Pope's position (and his claims to power and resources) rested on the pretense that he had a direct relationship with God, which wasn't true, and the Pope _knew_ (he _had_ to know, at some level) that it asn't true. (I agreed that this usage of _fraud_ made sense to me.) In Michael's view, Ben's charges against GiveWell were similar: GiveWell's legitimacy rested on the pretense that they were making decisions based on numbers, and they [had to know at some level](https://blog.givewell.org/2011/08/18/why-we-cant-take-expected-value-estimates-literally-even-when-theyre-unbiased/) that they weren't doing that.
[TODO: secret thread with Ruby; "uh, guys??" to Steven and Anna; people say "Yes, of course criticism and truthseeking is important; I just think that tact is important, too," only to go on and dismiss any _particular_ criticism as insufficiently tactful.]
That was my intuition. To formalize it, I wanted some sensible numerical quantity that would be maximized by using "nice" categories and get trashed by gerrymandering. [Mutual information](https://en.wikipedia.org/wiki/Mutual_information) was the obvious first guess, but that wasn't it, because mutual information lacks a "topology", a notion of _closeness_ that made some false predictions better than others by virtue of being "close".
-Suppose the outcome space of _X_ is `{H, T}` and the outcome space of _Y_ is `{1, 2, 3, 4, 5, 6, 7, 8}`. I _wanted_ to say that if observing _X_=`H` concentrates _Y_'s probability mass on `{1, 2, 3}`, that's _more useful_ than if it concentrates _Y_ on `{1, 5, 8}`—but that would require the numbers in Y to be _numbers_ rather than opaque labels; as far as elementary information theory was concerned, mapping eight states to three states reduced the entropy from lg 8 = 3 to lg 3 ≈ 1.58 no matter "which" three states they were.
+Suppose the outcome space of _X_ is `{H, T}` and the outcome space of _Y_ is `{1, 2, 3, 4, 5, 6, 7, 8}`. I _wanted_ to say that if observing _X_=`H` concentrates _Y_'s probability mass on `{1, 2, 3}`, that's _more useful_ than if it concentrates _Y_ on `{1, 5, 8}`—but that would require the numbers in Y to be _numbers_ rather than opaque labels; as far as elementary information theory was concerned, mapping eight states to three states reduced the entropy from log<sub>2</sub> 8 = 3 to log<sub>2</sub> 3 ≈ 1.58 no matter "which" three states they were.
How could I make this rigorous? Did I want to be talking about the _variance_ of my features conditional on category-membership? Was "connectedness" intrinsically the what I wanted, or was connectedness only important because it cut down the number of possibilities? (There are 8!/(6!2!) = 28 ways to choose two elements from `{1..8}`, but only 7 ways to choose two contiguous elements.) I thought connectedness _was_ intrinsically important, because we didn't just want _few_ things, we wanted things that are _similar enough to make similar decisions about_.
projects / Ultimately_Untrue_Thought.git / blobdiff