Data for Dinner

Just trying to figure things out as much as I can.

[Data for Dinner @data_dinner]

Went looking for vintage existential risk takes but it's hard to beat Churchill's 1924 article where he argues that "the prevention of the supreme catastrophe ought to be the paramount object of all endeavour". https://t.co/MJfJnVn7MC

[Data for Dinner @data_dinner]

@Yozarian22 Churchill did a 1924 essay on existential risk, tiny bombs of terrifying power are mentioned. https://t.co/ye5jyuobAW

[Data for Dinner @data_dinner]

@CAPABLANCA_92 @mattyglesias There are some polls but they tend to ask weird questions.

[Data for Dinner @data_dinner]

@StefanFSchubert Somehow this really hit me the year when my family stopped sending (and mostly stopped receiving) Christmas cards.

[Data for Dinner @data_dinner]

@StefanFSchubert Haha, fair enough. But "automation means economic growth and should as such be welcomed" is a statement the truth of which is not obvious to me in our current context. And I'd like to know how it's obvious to you at the same time as you're linking to disempowerment analyses.

[Data for Dinner @data_dinner]

@mattyglesias Maybe. But AtlasIntel nailed the final results and they showed Trump ahead the whole time.

[Data for Dinner @data_dinner]

@AndyMasley If I type 'climate 2030' into X's search bar, I immediately get posts with some very alarming wording.

[Data for Dinner @data_dinner]

@BryanCodes @besttrousers "The thing you thought was random isn't actually random", memorably the cause of the weird result in the 'hungry judge' study.

https://t.co/CBdcsw7yCm

[Data for Dinner @data_dinner]

@CharlesD353 @StefanFSchubert One underrated source of weird results is clumsy mistakes. There isn't always a fancy logical flaw, sometimes there's just something wrong with the data entry or an off-by-one error in the code etc. etc.

[Data for Dinner @data_dinner]

@StefanFSchubert @peterwildeford To steelman the hypester: As worded, the idea may have been that it would take 6 more months for capabilities to reach that level - the actual deployment might take significantly longer. Probably still quite wrong, though.

[Data for Dinner @data_dinner]

@AnechoicMedia_ About 30% of car drivers dying on the road are under the influence of alcohol. About 50% of road fatalities happen at night but it's pretty impractical to totally avoid driving after nightfall.

[Data for Dinner @data_dinner]

@StefanFSchubert In Isaac Asimov's works, the idea shows up that there would be cool future discoveries in sociology and it would be all nice and mathematical. The Naked Sun has "the Teramin relationship".

https://t.co/6uPxjnYUEQ

[Data for Dinner @data_dinner]

@KirkegaardEmil https://t.co/9TMHhAwvQ1

[Data for Dinner @data_dinner]

@Yozarian22 Tesla was a skeptic - this is from 1921; "scientific illusions", "absolutely impossible" etc.

[Data for Dinner @data_dinner]

@Jotto999 @mattyglesias The mean for three of the subgroups comes to exactly 6.00 - I wonder if there is some reason for that.

[Data for Dinner @data_dinner]

@robbensinger @bananafitz "Millions of troops ... are marching off, prepared to pay the ultimate price for the Central Powers' noble cause, which remains unclear as of presstime, or that of the mighty Triple Entente alliance, which no one can really put his finger on, either."

[Data for Dinner @data_dinner]

@jkeatn @JMannhart @captgouda24 @asteriskmgzn It took me several paragraphs and some googling until I understood that this was fiction. It is plausible that there have been people confused about sex and pregnancy and I was enthusiastic to learn all about it.

[Data for Dinner @data_dinner]

@JMannhart @captgouda24 I found this one frustrating for the same reason: https://t.co/oW16nfNwvf

[Data for Dinner @data_dinner]

@TrumpWasHere @Steve_Sailer @ChuckRossDC Today OpenAI can deliver a red suit, a blue shirt and a white tie. It could not do this yesterday before the rollout. It has been 897 days since my prediction in the tweet above, so this took longer than I thought it would. @StefanFSchubert

[Data for Dinner @data_dinner]

@TrumpWasHere @also_ted @Steve_Sailer @ChuckRossDC No luck. It's just too weird for Dalle. Imagen might be able to do it. Whatever they'll have in a year or two will very likely be able to do it. https://t.co/WpJBJskToD

[Data for Dinner @data_dinner]

@woke8yearold I expect this to trend up. https://t.co/wKEj91StZS

[Data for Dinner @data_dinner]

@tracewoodgrains AI facial similarity detection offers a second opinion. https://t.co/mUnFtxwjgE

[Data for Dinner @data_dinner]

13. But what about the theory that Justin Trudeau is Fidel Castro's son? They sure look alike, don't they? Well, the tool doesn't think they do. Here's a 0.0% match.

[Data for Dinner @data_dinner]

@woke8yearold @revenant_MMXX There's also that other guy who hangs out with Trump and who knows what's up.