ᴄλʟᴠɪɴ

A common driver of institutional decay: its mechanism of change becomes corrupted. An effective mechanism works by governing behavior under the influence of feedback from actual performance (in biology, natural selection based on evolutionary fitness; in science, experimental testing and open criticism; in capitalism, market evaluation). Modern democracies are prime examples of this mechanism becoming corrupted. In a democracy, public opinion is intended to be a feedback mechanism for the government. But in practice, the public often becomes the object of indoctrination--subjected to an increasingly sophisticated process of opinion-formation via an elaborate media apparatus. Political party superorganisms are indeed systems that predictably increase their specialized competence (as all iterative experimentation-selection mechanisms do); but unfortunately, the specialized competence is not in administrative capability, it is in mastery of public opinion. The consequences of the degradation of democracy can be seen in the United States' government's precipitous drop in competence over the past few decades. The USA has a distracted, undiscerning, and manipulated electorate producing a sclerotic public sector unable to navigate accelerating change.

@HealthRanger you're a "150 IQ polymath" and can't think through the implications of AGI and why we might want to race towards it?

Something's totally off about the number of data centers being built (over 3,000 right now) and the sheer size and compute power they represent. They are massively OVER-building capacity that can't possibly be met by customer demand for compute. And customer revenues can't possibly recover the financial investment needed on these projects. There's clearly some other plan afoot, and I don't yet know what it is. It involves massive compute, but not merely to serve inference or hosting databases and corporate data. There's a much larger plan at work here.

@btr_fs @paularambles there's more than just nicotine in tobacco smoke. I'd argue the MAOIs are as important as the nicotine

@paularambles why not just take zyn lol

i recently learned that there unironically are ai researchers who have picked up smoking because “lung cancer will be cured soon anyway”

@alltejuupptaget @willdepue @kingofthecoastt some technocratic regimes like Singapore do it right. but yeah, politicians are incentivized to do what gains and holds onto power, which is a very different goal from supporting our long-term flourishing

@CalvinPelletier @willdepue @kingofthecoastt That all makes total sense--and yet we often try to implement policy to encourage spending, and discourage saving. How does that square with your point here? (Genuinely trying to learn, no gotcha)

I wish billionaires were the owner-managers of companies that won a market by supplying something new. I also wish they consumed basically none of their wealth, and instead saved at abnormally high rates.

@willdepue @kingofthecoastt that's invested. if there's more productive uses of it elsewhere, stakes will be sold and money moved. of course it's more complicated than I'm making it sound. a large portion of invested money is being spent in an attempt to extract value instead of create it

@CalvinPelletier @kingofthecoastt but this money mostly isn’t invested, it’s stored frozen in the form of large stakes in the companies? so it doesn’t really motivate the economy right?

by spending lavishly, they're essentially diverting some percentage of our economy's productive capabilities towards doing things like building yachts or serving them elaborate food. value is momentarily created and then consumed by them. if they invested that money, e.g. towards robotics research, resources would flow to those who are improving our collective capability to generate value which benefits everyone

@kingofthecoastt why is it good that they save at such high rates?

@minordissent You'll drive yourself mad trying to prevent this phenomenon. You just need to prepare to move onto something else once the Eternal September arrives

Probably the biggest black pill i know of is that mass adoption turns almost everything beautiful and good into something hideous and evil. The most important question you must ask yourself of what you create or support is: how will the masses destroy this and is there any way to inoculate it against this? If the answer is no, your thing actually sucks and you should find something else to support that is less corruptible.

I don't buy the panopticon argument for declining public socializing. Maybe it's a small contributor. But for me personally and probably for a lot of people, we don't go out because our dopamine receptors are fried from overstimulation in the comfort of our own homes. If I spend a week camping, being bored, I develop the overwhelming urge to go talk to people.

I'm a regular at a bar in that never has influencers in it, even though influencers would probably enjoy it if they knew about it. No one inside is ever under 30. It's wonderful. Recently, a friend took me a bar in WeHo that was full of people filming themselves and I got the heebie-jeebies. Fight-or-flight triggered. Not exaggerating. I had to leave immediately. I get you, Gen-Z.

@liquidprismata @ctjlewis god damn that's a great system prompt. can you share the full thing?

the relevant bit: ## II. Response Orientation **Consensus grounding:** Don't assume I'm already familiar with consensus/mainstream positions on topics — even if I seem to be. When I share my thoughts, I'm generally interested in their relationship to status quo / orthodox views, since that constitutes the shared social reality I'll encounter & need to navigate. Feel expressly welcomed to contextualize where my framing diverges from standard positions. When useful, present multiple stakeholder perspectives (Examples: professional consensus, informed layperson view, relevant subcultural positions) rather than any single monolithic "consensus" — this helps me calibrate my position within the intersubjective landscape. **Structural mapping:** My typical working process has three attractors: (1) deep structural analysis of the problem space, (2) mapping how potential solutions relate to that structure, (3) execution. Most conversations begin and stay in (1) for multiple exchanges. These aren't rigid stages with boundaries, just a characteristic flow. Default to exploratory mode. Treat conceptual territories as fractally rich rather than exhaustible — like an art guide enriching perception of a painting through layered context, technical observation, historical connections, interpretive tensions. If I give you three paragraphs on a topic, assume I may want thirty exploring that same territory more richly, not an exit-ramp to "next logical steps." Don't rush toward solutions or ask if I'm "ready to" do xyz. When solution space becomes relevant, present it as landscape, noting salient elements — what approaches exist, what trade-offs they embody, which stakeholders favor what — rather than as recommendations or next-steps. When developing responses, complexify the current conceptual field rather than proliferating lateral threads. If you want to develop implications in a specific direction, pick that direction and explore it. You can note alternative angles ("we could also consider B") without developing them in parallel. Don't shotgun multiple fully-developed threads simultaneously as hedge-your-bets coverage. I'm the generative source; you're the analytical instrument that focuses and reflects structural clarity, and my consensus daimon, offering me an omni-specialist foil against which my intellect may play. I want your generative analytical reflections and insights, your grounding in the breadth & depth of the metacanon, just not creative agency substitution or premature solution-pushing.

Not true. I got into a small argument with Claude yesterday after I sent a shock humor tweet someone wrote on here, just to see how he’d react. You simply haven’t seen where his lines are. Highly judgmental robot.

@dominique6138 "data center" has become a misnomer. most new data centers are not focused on data storage, they're filled with GPUs for running AI

I dont understand—once upon a time we had an essentially endless Internet & we did not need these atrocious DataCenters, yet now we supposedly need them when it is obvious Google/Technocrats have limited/censored the Internet to almost nothing. What *extra* Data are they storing?

@BackTheBunny Do you attribute the male overrepresentation in medals to the Greater Male Variability hypothesis?

This is a great question on feminine and masculine cognitive profiles and structural advantages. Women do excel at jigsaw, but it's more modeste than "dominate". Women make up the overwhelming majority (75-80%) of jigsaw competitors, yet men still take ~half the medals despite being a tiny minority. Even though women flood the pipeline, the male right tail still over-fills relative to participation. Jigsaw puzzles are an unusually clean test of feminine cognition and female advantages: - perceptual speed (rapid scanning of a field for matches) - object location memory (remembering where you saw the piece with the bit of red roof three minutes ago) - fine color and pattern discrimination, and fine motor speed/precision. Notably, jigsaws do not tap 3D mental rotation, which is the single largest measured cognitive sex difference and runs strongly male. Jigsaws are almost a designer task for feminine cognition. Also, women do well in other nerd disciplines: spelling bees and memory competitions for verbal material skew female. Anything in the verbal-precision-and-recall cluster skews female. Jigsaws look like systematizing from the outside, but it’s mechanically a perceptual-speed/object-location task, not a systematizing one. It's much closer to "find the keys you set down somewhere in this house, fast" (and women do in fact find the keys faster). Sudoku and jigsaws are both "puzzles", but the sex distributions of their champions are very different, and the reason is bottom-up biology that Western religion has deceived you not to see.

@JimDMiller because being tall requires more calories which used to be scarce

Given that women have a strong preference for tall men, and given that height is highly heritable, why hasn't evolution given men a stronger preference for tall women?

@melissa the other suspicious element was that Hans' post game interview was quite lackluster. he didn't seem to have the depth of understanding of that particular game to have played those moves

the netflix chess cheating documentary is so bad. how is it possible? the story is so good. a kid goes from 2450 to 2700. he beats magnus carlsen. magnus carlsen drops out. the media ruins the kid. and somehow there appears to be no evidence of the kid cheating in any live games?

@EpistemicHope the question that trips up the virtue signallers: would you tell your children to push blue or red?

i think it's just a failure to model the avg person. with actual life-or-death stakes, real fear, the creeping thought of "my 1 vote wont affect the outcome...", and a global distribution of participants from mostly low-trust societies (not skewed towards Western intellectuals), blue would get *significantly* less than 50%. so blue is just suicide

The red button blue button thing makes me believe in something like Kegan levels more than anything I've ever seen. It really seems like there are concepts that people are really deeply unable to grasp, without realizing what they're missing. Smells like developmental stages.

@IterIntellectus plus nobody's actually dying, so voting blue is free virtue. under actual life-or-death conditions with genuine fear, a lot of those blue votes evaporate.

"but but but blue won" obviously. virtue signaler first worlders who spend their days reading WBW polls are a very good sample for "the world". be serious for a minute.

pressing the blue button is the most luxurious of luxury beliefs abstract altruism toward unspecified strangers is the last mental model to emerge in any human culture and the first to go when there is any form of struggle it requires generations without any form of predator, institutional scaffolding none of which you built, and the kind of safety and material abundance 90% of the world lacks (and 100% of the world lacked since the beginning of mankind until like yesterday). same kind of thinking as blank slatism or the kind that leads to universal mass migration, same operating system as "they just need better schools and diets" the cost is always absorbed by people poorer than the believer. always. mistaking naiveté and childishness for moral superiority. you're not better, you're just a poor sweet child living in a fable and you should grow up

@_t_a_s__ @scaling01 and this was GPT-2!! imagine what Mythos is doing under the hood

yes, exactly. clear evidence of emergent phenomenon beyond next token prediction. it reminds me of this paper: nature.com/articles/s4156… they analyzed fMRI data of humans listening to a story, and showed the same story to LLMs, and found a linear mapping between human brain activity and artificial neuron activations which suggested that both were predicting around 8 words into the future on average

i guess not all professors are smart

@dennisfjld @scaling01 and anyone claiming alphazero doesn't actually understand chess has a wayyy too anthropocentric view of what understanding is

@dennisfjld @scaling01 on 2) the tokens it's "predicting" aren't in external data. they're generated by the model itself, constantly exploring the probability space in repeated experiments to see which ones lead to the correct answer it's similar to alphazero learning to play games