the through line between several theories of leadership, sacrificial kingship, sovereign banking, scapegoating, etc, is that large scale coordination can only solve around the constraints of human nature by concentrating obligations (debt, promises, favors) onto the personal "balance sheet" of a central figure who then holds them in a sort of quantum superposition of solvency/insolvency. the magic is that sometimes this works, against all logic sometimes everyone walks away thinking that they got by far the better side of the deal. but what never works is trying to actually true up balances while things are still in motion. for example, among the millions of frantic things that are happening behind the scenes, one of the most important sets of maneuvers is the principal of an nvda counterparty calling up and saying "we've always had an informal agreement with jensen that if X happens then you guys would do Y. you know that because you were in those meetings too. given the circumstances, we now need you to put that in writing, hope you understand" and for every one of these awkward conversations, there's probably two more going on inside counterparties where they are tearing their hair out because they don't even know who is now the right person at nvda to even make this ask to. i would say that the probability that this sort of thing can resolve painlessly is basically zero, so insofar as you are betting on "nothing ever happens" you are really just betting that these "off-balance-sheet understandings" are small in magnitude relative to the accountable activity. if that is not the case, then by the time you get to a legible "sacrifice the king" moment, that is not even a desperate play to "make good" on the liabilities of the system, it's the final, irrevocable step to declaring insolvency
there is a type of category error that goes like this: the bolsheviks don't really understand how the economy works and their goal of communism is unachievable as conceived therefore it is impossible that they will win
to me the signal here is w.r.t. the AI hype divergence between US vs China i.e. China treats AI as transformative tech that will diffuse on normal timelines (no rsi blowup). new common knowledge: "China" was not lacking access to evidence of the "what did jensen see?" variety
"okay so imagine the economy is mitch mcconnel and the ai buildout is the drugs his doctors are pumping into him 24/7, now imagine that there is a shift change and the republican doctors are suddenly replaced with new democrat doctors..."
@AuronMacintyre "You were the chosen marks! It was said you would pay into the SSTF, not benefit from it! Add balances to the forced ponzi scheme, not leave it underfunded!"
@MorlockP it's actually the other way around, the chance the 2♣️ remains depends on how many cards have passed, but the chance the A♠️ remains is always 75%. at the limit where there's only 4 cards left, there is 25% chance of Ax, A♠️ and 0% chance of Ax, 2♣️
I love that scene in Good Will Hunting where Ben Affleck is talking to Matt Damon's titular character Will Hunting and he's like “you've got a gift man and you've gotta get outta this town and make something of yourself” and the gift in question is being really good at math and the town he needs to get out of is Boston, Massachusetts.
There is only one thing left for humans to do: position themselves at the end of particular processes of computation and determine whether, at the particular scale at which the human is being presently constituted, there is a fit or not.
This is not only an irreducible and indispensable position that only humans will ever be able to take up, but it will be an enormously demanding one involving high levels of learning and training, an ability to improvise and take responsibility on the spot, and to convey one’s decisions back to the computational order (the center) in such a way that a new modes of fittingness will result from the next round.
OP might be bait but this lady's father is alberto fujimori, dictator of chile in 90s and born in 1938, only a few years after her grandparents immigrated to peru. pretty sure they were middle class all the way through. i mention this because it's similar to chinese americans whose parents came here on student visas in the 80s and 90s claiming a throughline to the cantonese workers who built the transcontinental railroad. apparently fujimori the father also had his version of birther conspiracy in which opponents claimed he was secretly born in japan and thus ineligible for the highest office. obama would be another example of this sort of politics as sports, where his rise was occasion for self-congratulatory proclamations of how far we had come in race relations. but obama's genealogical links to american slavery were exclusively through his white mother's side, mainly through a not atypical cast of landowning ancestors but also a surprising link to a man named John Punch, often claimed to be the first enslaved africans in america
This is interesting whatever. What's more interesting to me is that the President of Peru is Japanese. Apparently in three generations they went from migrant agricultural workers to the highest political office in the country.
it's refreshing how clear eyed she is, but her deference to past sacrifices is not a sustainable answer. prior generations before leaving also sacrificed. those who stayed and lifted up china worked at least as hard. just like saying "my culture uniquely values food and fellowship" shows a shallow understanding of one's culture, saying "my lineage uniquely values hard work and sacrifice" shows a shallowly examined identity
china's recent series of wins means that the ground is irrevocably shifting in the collective psyche of the diaspora. after all, the foundational myth for 2nd gen is "my parents sacrificed everything so that we would be free from the backwards conditions of the home country"
it should come as no surprise that people are starting to feel the pull. when you give up family tradition in order to chase opportunity, regardless of the many other choices you are making, you are also choosing to make chasing opportunity your new family tradition
like basically every complaint i had about the cli terminal has been neutralized. everything is just smoother: copying/pasting, formatting, session identity and life cycle, telemetry, etc. and using tool call hooks feels nice bc you can bolt on a custom UI for your custom hook
this might be old news to anyone who might act on it, but for months ive thought about making a thin react wrapper for claude code and kept pushing it off. finally did it in basically one shot with the llm reading the sdk docs to set up a local service and it works like a dream
i suspect part of why anthropic as a company seems to hold certain beliefs about claude is because from their pov their claudes just started talking like that one day. for an external user, it's easy to think "they talk like this because the people who made them preferred the ones that sounded like this" but i imagine from the first person creator point of view, it's like seeing your kid do something that you find you find particularly quirky and endearing and not realizing it's because they osmosis'd it from you when that is apparent to other adults
just my opinion, but i think i talk to claude and chatgpt alot and whenever it starts talking like that i basically wrap up the session and clear context. to me this doesn't parse as "human-like self-awareness of epistemic limits" and more like "pegged the user as wanting self-referential musings" and ime it's hard to get it out of this attractor basin (or rather not worth the effort). maybe that is "human" in some sense, but to me it feels actually cynical and secretly condescending, a more meta version of the overly sycochant failure mode, basically a mode collapse from objective goals to a solely focus "saying things the user would find gratifying to hear"