Humanity has many solvable problems. Among the biggest are poor epistemics, and thinking that epistemic problems are unique to political enemies. EA, LW, ACX
"Less is More: Recursive Reasoning with Tiny Networks" The news: https://t.co/nZBImS58eI
My thesis from 2½ years ago expected this, and explains why it's extraordinarily dangerous news: https://t.co/uwobP5wjOx
So I'd bet big money that his new predictions:
- Trump gets third term
- Iran is a forever war & the US will do a national draft
- US civil war
are wrong. Even if the ceasefire doesn't hold, Trump has wanted out for many weeks and will give major concessions to escape his war.
Normally I like Diary of a CEO, but Prof Jiang strikes me as off-base. Normally I give deference to people who do correct predictions, but not if the rationale behind the predictions is bad AND the predictions aren't so specific that incorrect rationales couldn't produce them.
I was a big fan of Paul Warburg for a little while. I don't have time to explain, but his latest video will age like milk. (though he could be right about the *people* of China having a sucky future; we'll see)
@AndrewPerpetua Andrew has said that they don't review more videos from the Ukrainian side than the Russian side, and I like what they ask of their volunteers:
@HistoryLegends_
has a new video on the number of KIA on both sides of the Russia-Ukraine war. He starts with negative points because he's always implicitly accepted Putin's "denazification" framing of his definitely-not-a-war. I must've seen 100 of his videos, of which 🧵
@AndrewPerpetua Andrew & co's call for volunteers is still active: https://t.co/4jzL4SBk1U
I found Andrew's YT streams to be less inflammatory & more insightful than HistoryLegends, tho he hasn't done one in >3 months.
Here's Andrew commenting on the loss ratio:
https://t.co/bX3THgDejk
Roughly speaking, for each 4 Russians we see killed in combat we see 1 Ukrainian. Sometimes it is 3 to 1, sometimes 5 to 1. Sometimes 1 to 1. But, on average, about 4 to 1.
Also, if you saw the methodology I used to calculate this loss ratio you would see how insanely favorable it is to Russians. Because I only counted videos I did personally, and the hugely disproportionate majority sometimes literally to the point of exclusivity of videos I personally do are russian videos. And even then the ukrainians kill more.
@AndrewPerpetua A Colonel in the Russian Army? I can believe it, since I noticed a DPR separatist/terrorist (Pavel Gubarev) trusted numbers from the pro-Ukraine side as well. https://t.co/YLKhkDJb6d
Fast forward a couple of years, the same guy now seems to think that the Ukrainian numbers for Russian casualties (including DPR/LPR) are accurate. https://t.co/IVxsK2Tj1v
@AndrewPerpetua Well, I was hoping you could comment on the video on a factual rather than just vibes level, because his new video and web site are likely to attract a lot of attention. If you don't have time, my handle is not for nothing so I'd appreciate a retweet.