Pablo Villalobos 🔸

655 posts

Pablo Villalobos 🔸 banner
Pablo Villalobos 🔸

Pablo Villalobos 🔸

@pvllss

Trying to thread the needle of prophecy, mostly by clarifying the prophecy.

SF Bay Area Katılım Aralık 2013
427 Takip Edilen655 Takipçiler
Pablo Villalobos 🔸
Pablo Villalobos 🔸@pvllss·
@KelseyTuoc @rwlesq Aren't there some statistical differences in behavior between men and women? Even if transition is medically perfect, maybe it's still possible to identify trans people based on behavior on average. I don't know why anyone should care about that but some people probably will
English
0
0
0
28
Kelsey Piper
Kelsey Piper@KelseyTuoc·
@rwlesq I've thought for a long time that all the arguing over trans people is only coherent in the window (which may be decades-long, but won't be centuries-long) where medical transition is possible but substantially imperfect
English
6
1
118
3.1K
Chasing Ennui
Chasing Ennui@rwlesq·
It's the year 2525. Through some combination of gene therapy and surgery, human testes can be coaxed into generating viable ova rather than sperm. People born with male anatomy can also get an artificial womb implanted in their lower abdomen, allowing them to carry a fetus to term. They continue to have XY chromosomes. Is a person who undergoes this procedure "female?"
English
4
0
15
3.6K
Pablo Villalobos 🔸
Pablo Villalobos 🔸@pvllss·
@YafahEdelman It seems better to be able to optimize your information storage and channels via gradient descent than with hard-coded compaction mechanisms. The bitter lesson again. AlphaGo worked and was only getting better, but AlphaZero was vastly better.
English
0
0
1
36
Yafah Edelman
Yafah Edelman@YafahEdelman·
I don't get why everyone is talking about continual learning so much. The original GPT-3 paper was all about and how models can learn in-context, and they've only gotten better since then.
English
24
3
88
8.8K
Pablo Villalobos 🔸
Pablo Villalobos 🔸@pvllss·
@kingofthecoastt Well, the unemployment rate has fluctuated between 3-8% for decades despite massive changes in salaries, population, LFPR and the labor market in general, isn't this similar?
English
0
0
1
39
Brad Carson
Brad Carson@bradrcarson·
Just my semi-annual reminder that there are 125,000 bills introduced each year in states, and, given some legislatures are every other year, about 225,00 per cycle. AI bills # the same as fluoride bills and kids trans bills. 1% of total on AI is actually < than to be expected.
Adam Thierer@AdamThierer

David Sacks has it exactly right, except that the number of conflicting AI bills out there today is even higher! The total is now well over 1,600 bills across the nation, and there is no end in sight to this madness.

English
1
3
20
1.4K
Pablo Villalobos 🔸
Pablo Villalobos 🔸@pvllss·
@_tokuninai No se yo. Los market makers que mantienen la liquidez en esos mercados de futuros van a empezar a cobrar más comisiones si en cualquier momento puede venirles una de estas. Todos los demás que usan futuros pagan el coste. Llevado al límite a nadie le interesa comprar ni vender
Español
0
0
1
215
特にない
特にない@_tokuninai·
Gente con información privilegiada publica esa información para permitir que los mortales también podamos ganar dinero y este lo llama estafa.
Lorenzo Ramírez@lorenzoramirez

La estafa de la Bolsa: Un grupo de "inversores" compran contratos de futuros sobre el S&P 500 por un valor de 1.500 millones de dólares (esta operación fue tan voluminosa que hizo que todo el índice subiera un 0,3 % en ese mismo minuto). Un cuarto de hora más tarde Trump anuncia «conversaciones productivas» con Irán. El índice bursátil se dispara y los que compraron ganan 60 millones.

Español
2
0
30
2.6K
Pablo Villalobos 🔸
Pablo Villalobos 🔸@pvllss·
Initially I thought space data centers were too vulnerable: anyone capable of destroying a few satellites with missiles could quickly Kesslerize LEO and obliterate all those trillions of capex. But no, turns out Kessler takes decades or more. You'd need thousands of missiles to significantly disrupt a large satellite constellation. Way harder than destroying Earth-based data centers
English
0
0
7
1.2K
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTex·
I was motivated to dismiss space datacenters initially because it's just too good, it makes all sorts of neat earth-bound projects obsolete, and privileges the US and Elon. But the mafs actually checks out, and yes on short timelines. Sorry about that. Not silly at all.
Lisan al Gaib@scaling01

datacenters in space are silly 100kW isn't even enough to power a single GB200 NVL72 but sure let's spend 100 million just for launching the damn thing, while on earth you could buy like 30 GB200 NVL72 for that price

English
22
8
512
50.1K
Pablo Villalobos 🔸
Pablo Villalobos 🔸@pvllss·
@OscarEconomik Todo eso son excelentes argumentos para eliminar o bajar el impuesto permanentemente, pero bajarlo temporalmente en respuesta a una crisis solo hace que sea más difícil planear adecuadamente la respuesta a esta crisis y las futuras
Español
0
0
0
114
Óscar García
Óscar García@OscarEconomik·
Creo que los economistas (me incluyo) analizamos muchas veces la realidad bajo un prisma académico que creemos neutro pero no lo es: "Beneficia tanto a ricos como pobres" ¿Es eso realmente malo? "Es muy costosa" Para el Estado sí ¿En qué se gastaría ese dinero si no fuese en ello?¿No es mejor dejarlo en el bolsillo del ciudadano? "Distorsiona precios" El impuesto es lo que distorsiona el precio, quitarlo lo devuelve al cauce natural.
J.I. Conde-Ruiz 🇺🇦@conderuiz

Bajar los Impuestos a la Energía es una mala idea: ▶️no es una medida Focalizada—beneficia tanto a quien lo necesita como a quien no ▶️es MUY Costosa ▶️distorsiona la Señal de Precios y aumenta la demanda ▶️no beneficia sectores clave como el Transporte, que ya se deduce el IVA

Español
11
31
231
17.5K
ozy brennan 🦙
ozy brennan 🦙@ozyfrantz·
@myhandle @Jake_Etcetera I really don't think the good people of literotica are asking themselves "is this politically correct? will this anger the longhouse?" before they onehandedly pen their magnum opuses
English
1
0
16
221
Jakeup
Jakeup@myhandle·
talking about tits and ass is the *safe and tolerated* expression of male sexuality. if a man wrote the reverse POV of a woman's "I want to sleep with my professor who is also a werewolf" you'd call the cops. men can only write narrative erotica on obscure forums under pseudonyms
LIZZY💥@LizzyStarrrdust

Men tend toward framing sex and sexual fantasies in a very object/body parts+ movement/physicality sense whereas women frame sexuality more from a connection/erotica/narrative driven sense. My assertion is that a great many women don't understand the extent to which men frame it this way and seem to struggle to accept these differences once they do.

English
7
1
145
7.4K
Pablo Villalobos 🔸
Pablo Villalobos 🔸@pvllss·
Cuando un productor destruye inventario o bien ha sido un error de planificación previa y habría ganado más produciendo menos, o bien alguien le está pagando para que produzca de más (como la PAC en Europa). No se hace para mantener los precios a un nivel competitivo, se hace porque el coste de distribuir esos productos es mayor que el precio al que se venderían. Vete a un agricultor y llevate lo que le sobra de cosecha, nadie te lo va a impedir porque quieran mantener los precios
Español
0
0
4
62
Musketaquid
Musketaquid@musketaquid0·
@_tokuninai Que a lo mejor te pensabas que hablaba de que los supermercados tiran lo que no venden, pero es que no es eso, hablo de lo que tiran los productores para mantener los precios a un nivel competitivo. Ya has aprendido algo hoy, mira.
Español
3
0
1
230
Pablo Villalobos 🔸
Pablo Villalobos 🔸@pvllss·
@jachiam0 Science fiction was killed by the difficulty of imagining a coherent future after the singularity. For a while it was possible to write science fiction about the takeoff, now that's no longer fiction
English
0
0
2
193
Joshua Achiam
Joshua Achiam@jachiam0·
Permit me a moment of a spicy, unqualified complaint that is so contestable it is obviously wrong to say. But here is my gripe of the moment, inadequately qualified, surely disprovable. Science fiction is a dead genre; it was killed by space opera and public distrust for tech.
English
14
1
49
7.3K
Pablo Villalobos 🔸
Pablo Villalobos 🔸@pvllss·
Two ideas come to mind: 1. For a physicalist surely sentience is coordinate-independent. Whether a physical process is sentient or nor does not depend on any labeling of states additional to the actual dynamics. If this is the case then sure, the fact that we can assign discrete "coordinates" to a physical GPU and identify suggestive patterns doesn't mean that the GPU is sentient. But it also doesn't rule it out any more than we can rule out say sentience of bacteria. At best it means that computational interpretations give no evidence. 2. You can argue "no computation without mapmaker, sentience doesn't require a mapmaker, therefore computation is not sentient". But the deduction can be done in the opposite direction: "no computation without a mapmaker, sentience appears to be functional/computational, therefore sentience can only be evaluated by a mapmaker". I think this might actually be true, in a deflationistic sense: sentience is another abstraction humans came up with, and we only delude ourselves by mystifying it and giving it ontological and moral weight
English
0
0
1
76
Alexander Lerchner
Alexander Lerchner@AlexLerchner·
🧵1/4 The debate over AI sentience is caught in an "AI welfare trap." My new preprint argues computational functionalism rests on a category error: the Abstraction Fallacy. AI can simulate consciousness, but cannot instantiate it. philpapers.org/rec/LERTAF
English
52
45
268
100K
Pablo Villalobos 🔸
Pablo Villalobos 🔸@pvllss·
It's not a tradeoff between economic efficiency and other goals, it's that parasites are destroying your ability to achieve either of them. The US needs more neoliberalism, and to subsidize the defense production it wants with its massive budget. It needs less corrupt elites
English
0
0
1
45
Pablo Villalobos 🔸
Pablo Villalobos 🔸@pvllss·
Remove internal rent-seekers, subsidize the production or the redundancy you want, and the free market will do its magic. The CHIPS Act fabs are working. SpaceX achieved mass manufacturing of cheap rockets which China still can't replicate 2/n
English
1
0
1
46
Pablo Villalobos 🔸
Pablo Villalobos 🔸@pvllss·
@Manuel_do_rio Well, we call them revolutions but they were very long and diffuse processes in which people slowly copy new ideas and iterate on them. That process of diffusion and compounding growth seems fairly robust to me if the environment is stable enough (eg: regular seasons)
English
1
0
1
30
Manuel del Río Rodríguez
Manuel del Río Rodríguez@Manuel_do_rio·
@pvllss I mean... there's only been two revolutions in human history worthy of the name: the Neolithic and the Industrial. Many orders of magnitude below those, perhaps 1688. Of these, I'd guess that only the first would be quite likely in most worlds like ours, given enough time.
English
1
0
0
10
Manuel del Río Rodríguez
Manuel del Río Rodríguez@Manuel_do_rio·
In the last few years, I've come to realize how incredibly contingent, fragile and absurdly unlikely was our stumbling upon both Liberalism and the Industrial Revolution. And it makes me very pessimistic about us ever winning jackpots again.
English
1
0
4
127
Pablo Villalobos 🔸
Pablo Villalobos 🔸@pvllss·
Right, it might be the case that a low percentage of worlds reach our stage in say 5000 years of history, but maybe as you increase the "simulation period" the chance gets closer to 100% if liberalism is self-sustaining after a certain point, and the preconditions appear often enough
English
1
0
1
9
Manuel del Río Rodríguez
Manuel del Río Rodríguez@Manuel_do_rio·
Precisely, they are good counterexamples. Just goes to show how close you can get to having all the ingredients for Industrial Revolution and Liberalism and still not get the jump. Both require very peculiar conditions to arrive at, sustaining momentum when benefits aren't obvious yet. Like, if you were to repeat history a million times, I'd guess a low percentage of timelines would develop both, and an even lower do it before our timeline.
English
2
0
1
14
Pablo Villalobos 🔸
Pablo Villalobos 🔸@pvllss·
@drethelin Idk, the mechanism sounds plausible to me. The brain does learn and it does have limited capacity
English
0
0
0
31
Pablo Villalobos 🔸
Pablo Villalobos 🔸@pvllss·
There are good ways and bad ways to implement open borders, basically less control at the borders must be balanced by more control inside (who can vote, use public services, and so on). This is doable, the UAE does it, it's also unrealistic in the US, but the right level of analysis is not whether econ-brained people disregard culture, it's how we can reap the benefits of trade with minimal risk to the culture that makes it possible
English
0
0
0
68
Hunter Ash
Hunter Ash@ArtemisConsort·
The econ-brained treat culture and institutions as if they fell out of the sky at random and are now immutably attached to specific plots of land. Look how much more money a Haitian makes in the US! It’s a kind of local approximation in nation-space, and like all local approximations it breaks down if your step size gets too large, if you change too much along any one axis. I don’t think anyone believes, if they honestly reflect, that if we swapped the entire populations of Pakistan and the US, with the laws on the books initially kept constant, the outcomes would be better-predicted by the pre-existing outcomes of the formal nations rather than the people. Yet they support policies (open borders) that only make sense if some version of this claim is true because they’re thinking in terms of marginal units of labor while treating a bunch of coefficients that depend on the character of the people as constants. Taking the partial derivative. Great piece.
Richard Ngo@RichardMCNgo

New post on econ-brained thinking and what it gets wrong about prediction markets, land value taxes, AGI, etc.

English
12
37
481
44K

Keşfet

@KelseyTuoc @rwlesq @YafahEdelman @kingofthecoastt @daniel_271828 @bradrcarson @_tokuninai @OscarEconomik