"a" by @riceissa
Eliezer Yudkowsky retweeted
Biden still hasn’t appointed a new head of the FDA. I vote for @ATabarrok. Please RT if you agree.
15
158
7
339
Fascinating overview of recent work in polygenic screening of embryos in IVF from @hsu_steve: infoproc.blogspot.com/2021/0… Risk reductions are much larger than I'd expected (eg 42% relative risk reduction for T2 diabetes); and measures seem to have low correlations (so not zero-sum)
1
0
0
18
Descriptively, it's hard to imagine selection techniques of increasing complexity not seeing widespread adoption in decades to come. Normatively, there certainly are some scary futures (and paths) down this road. Helpful analysis here of many issues: jme.bmj.com/content/early/20…
1
0
0
5
Andy Matuschak retweeted
10
14
11
93
Somewhere there is a story about Keynes, advising a bank, & saying that if ever the board unanimously agreed with his investment recommendations, he knew it was time to change the recommendations - if everyone thinks it's time to buy, the price is too high Anyone know a source?
2
0
1
11
Interesting:
$450M Series D for PsiQuantum. Samir Kumar of Microsoft’s venture fund: "PsiQuantum and Microsoft have a shared perspective on the need for a good number of logical qubits enabled by fault tolerance and error correction on 1 million-plus physical qubits." psiquantum.com/news/psiquant…
1
0
0
7
Having spent the first ~46 years of my life with government and the public underestimating the probability of a major pandemic, I expect to spend the rest of my life with them (and, probably, me) overestimating it, in a way that will mostly decline over time.
4
2
1
103
I'll give myself this much credit: I thought it pretty likely I'd see a major worldwide pandemic in my lifetime. I hope this was the only one.
2
0
0
28
TBC: for many kinds of preparation treating it as probability ~1 in the next 5 years is actually a good strategy, & will yield outcomes indistinguishable from a correct probability estimation.
1
1
0
20
Wild to think ad clicks paid to (essentially) solve the protein structure prediction problem
6
54
8
512
Too soon to answer, and a fundamentally silly question, but fun to think about: how many more important things did Bell Labs discover?
4
2
0
56
(All afaict as an outsider: the people who study protein structure mostly seem to agree that the hardest part of the problem of prediction has probably been done. But this is only a piece of an overall puzzle...)
3
0
0
25
This thread feels a tiny bit like channeling @patio11 (my voice is all wrong, but this is the kind of thing reading him makes me see more easily...)
0
0
0
9
Indeed, cliche-dom is often a consequence of greatness. Starry Night was the first print I ever got carefully framed. Still love it.
Cliché paintings that are actually great
5
0
0
67
Eliezer Yudkowsky retweeted
If you think you don't trust scientists, you're mistaken. You trust scientists in a million different ways every time you step on a plane, or for that matter turn on your tap or open a can of beans. The fact that you're unaware of this doesn't mean it's not so.
1,047
1,325
377
8,691
Show this thread
Nobody understood them, nobody understood why it was necessary, they were hated and they knew they’d be hated and they did their job anyways. Not perfectly, but better than I or the TIPS market spread expected. All hail the Fed!
11
2
1
95
Show this thread