Notes

I think I take the structural fidelity of frames more seriously than many around me.

If I pay too much attention to the words I'm speaking or writing, I become nauseous.¹ If I pay too much attention to the words others are speaking or writing, I become nauseous and also relatively misanthropic. This is minimized when I'm operating inside idiolects; my current favorite is Finnegans Wake.

Writing tires me because engaging in conceptual wheat-and-chaff separation, as necessary to dictate words to page, generates nearly irremovable, painful, mental detritus. I don't know how to "garbage collect" while preserving the reproducibility of the generative thought-patterns. Some information is in the artifact, and some is in my head. Their relationship is complicated and not easily compressible!

Relatedly, it's very difficult for me to ingest someone else's "vocabulary."² Locally operating within it is fine. It requires some suspension of disbelief but I've had a lot of practice. Retaining those insights, however, is much harder.

I only feel clean when frames are correct, coherent, and minimal. No one has frames that are correct, coherent, and minimal. The burden of the boundedly-rational agent is to manage this tradeoff. My frames are certainly nowhere near optimal. But translation costs are real.

I'd really like to develop a healthier relationship with confusion. (I wish it wasn't physically aggravating). My best guess at how to do this is to find deep frames I'm satisfied with. Tips appreciated.

Addenda:

The "[t]echnologist" remains . . . valorized? (at least, in certain San Franciscan subcultures). Tech people who have {passion, taste, discernment}, intrepid changemakers seeking to leverage innovation as a force for good. Perhaps it's Lindy: the word was coined in Bentham's Chrestomathia to describe an expert in "Technology," the "connected view" of arts and manufactures, exactly two centuries ago.

Certainly they play an integral role in a modern society. I do, however, take some issue with:

A “technologist” is anyone who thinks critically about the opportunity for technology to improve society.

(1) Too broad. "Technologists" are builders. The democratization of internet tech came a unification of thought and practice. (The Collisons when building Stripe)

(2) Too narrow. Frontend gardeners and Linus Torvalds provide taste as a public good; curated creation raises the quality waterline; artists shouldn't be excluded.

(3) I worry the predominance of tech, the centralization of progress in tech, takes too much mindshare of the thoughtful and ambitious. Startups are vehicles of a certain type. Venture provides stimulus of a certain type. There are other types.

What will it say about our time that the greatest discoveries were made in industry? Hermits and the military made the atom bomb.¹ Billionaires and not-quite-hermits are making superintelligences, mind uploads, aging cures.

Maybe the not-quite-hermits are not-hermits for lack of places to hermit. Technologisms cannibalize. Memetic centralization may be as dangerous as economic centralization.

Valor is worth its weight in gold. To me, society lacks hermit heroes. There's a place to start.

wrote this in ~an hour in early 2025 in response to: what do you imagine the world to look like with an indefinite ASI ban, imposed now? there are some interesting themes but definitely needs reworking

We’re operationalizing ASI-ban as a ban on AI technology that can serve as a drop-in replacement for >99% of human workers that don't run on biological substrates. This broadly lets the compute & AI advances of the next 1-2 years occur, and mostly unclear after that. I would suspect this roughly corresponds to a compute cap of 1e36 FLOPS, which is about 10^14 H100s (10^6 is biggest datacenter today), and factoring algorithmic advances the equivalent datacenter in 2032 might be 10^10 H100-equivalents which would cost 10^4 * $4bn = $40trn. So let’s also say that there’s a ban on >$1-10trn training runs, depending on year, with this value going down over time.

Global GDP is around $100trn today, so the maximum size training run that would be allowed by 2025 is around ~1% of Global GDP, or around 4% of American GDP. This is 2x what is currently being suggested with the Stargate Project, so if Stargate goes well such a procedure would not be entirely out of a question (of course, Stargate is allocated across multiple domains, not just specifically training, but I suspect most of it will go to clusters).

I’d first like to forecast & research what the effects would be on specific narrow domains: energy, software infrastructure, and biology & human augmentation. I’d then like to look at what large macro-trends, sans AI, would be affecting the long-term course of society (climate change and fertility decline come to mind, as does cultural drift), and how AI development to the level that would be permitted would affect then. I would then like to forecast geopolitical developments over the next 5 years, see how that affects long-term AI development, and then start forming general probability distributions across the long-term effects on human society. Ideally the output of this research proposal would be hundreds of pages of well researched forecasting, but here’s an attempt figuring out what those might look like.

In the world without AI development, solar adoption will likely increase across the world, such that the cost of energy starts dramatically decreasing making technologies like solar desalination more possible. The energy gap between the developed and the developing world is such that 1 kWh costs ~1,000-10,000x more in Uganda vs Iceland, and if the cost of solar continues to decrease the energy gap will likely close (the solar panel infrastructure pipeline is relatively robust). Fusion is expected by people in the industry to be entirely net-positive by 2035, and commercial reactors are expected by 2045-2050. I would want to research exactly what sorts of ways in which advanced AI of the sort expected 2-3 years from now would help accelerate the energy production process, but this seems like the sort of thing that will continue sans some global war. So we can expect the cost of energy to substantially decrease within our lifetimes.

Software infrastructure will become extremely fragile and irrobust to both technical and non-technical attack vectors. It is likely that LLMs develop superpersuasive abilities within the next few years, without this requiring superintelligent AI, and also generating training data for finding and utilizing code vulnerabilities is generally quite easy. I would look into what the offense-defense balance would look like for software used by institutions worldwide by 2027 to 2028, and whether or not formal specification / formal verification provides a good solution to e.g. the Linux kernel being broken by a non-state or even non-resourced actor. Likely this does not cripple the world’s ability to communicate, because it is possible to make robust protocols where the stack is verified, but it does make critical software much more vulnerable.

Biological foundation models are well within the realm of the possible, and don’t require absurdly large training runs. The Arc Institute is training them at the moment, and they will likely get 10-100x larger even without national investment. Bio foundation models are incredibly dual-use: it would be very easy to finetune a toxin-producing model that is superhuman at producing toxins, and perhaps similar for pathogen generation. It is worth looking into whether or not there are substantial barriers to creating a biological foundation model that is superhuman at pathogen generation, given current technologies, and whether or not it’s possible to make base bio foundation models that don’t have these negative capabilities (also developing scaling laws, assessing the likelihood of distributed training setups & synthetic datagen). Worth developing policies that restrict access to DNA/RNA manufacturing (these are very poor right now, as per studies by Esvelt et. al.).

Human augmentation seems to be proceeding well. Neuralink is seeing success in its patients, and we will likely see silent speech interfaces as well as noninvasive BCI allowing for much more immersive AR/XR. Cosmetic invasive BCIs (invasive BCIs not required by medical usage) might become technically feasible within the next decade, and it is worthwhile doing a general bandwidth study / seeing if the computer-human bandwidth would follow scaling laws. Much invasive BCI usage is locked behind “brain foundation models”, or compute that allows the BCI to interpret various neuronal signals and brain-waves. Likely good generative models can patch much of this, but it’s unclear whether or not these capabilities are locked behind the 1e36 barrier. More research should be done here, specifically on how AI can accelerate different approaches to BCI etc.

Nanomanufacture might become much more feasible, conditionalizing on a foundation model that is trained on different kinds of chemosyntheses and nano-manufacture models on the micro-scale. Space travel is likely bottlenecked behind rocket design & procurement & manufacturing in the real world, but it’s plausible that biological research acceleration will in fact lead to breakthroughs in astronaut containment etc. It is unclear generally how much manufacturing will get accelerated by tool AI, but this is worth looking into.

RAND has recently put out a report that the U.S.-China relationship will take on many of the contours of the medieval era, placing us in a “neo-medieval age.” This is mostly due to their predictions that the international order will disintegrate, and that technological development will be increasingly stratified. There will be less aid from the developed world to the developing world, less trade, and less international communication. These predictions I find to be largely prescient.

Forecasting chip manufacture is also useful: is it really the case that the US will be able to onboard chip manufacture at the level of TSMC fabs within the next 2 years? They are attempting to do so, and given that ASML is in the Netherlands (under the Western sphere of influence) it is likely we reach this level. Analyzing Chinese supply chains is a worthy piece of work, however, and I place substantial likelihood we reach chip technological parity as well as tool AI capacity parity, placing us in a proper great power conflict.

With a ban on superintelligence, the risk of a single country developing a singleton superintelligence and using that to dominate the rest of the world (or lightcone) is low. Military conflicts between the US and China might result in the destruction of chip supply chains in both countries, leaving a generally leaderless world, but it’s more likely that there are hot proxy conflicts as opposed to hot great power conflicts. However, this does mean that energy costs for the developing world will likely remain high unless e.g. solar panel supply chains become more distributed & robust globally (and even if the US becomes much more isolationist, China still considers it worthwhile to expand its global influence, which is unclear).

Culturally, there seems to have been quite a large right-wing trend in the Western world. Nativism has increased (see AfD in Germany, Trump in USA), and along with it are culturally right elements that are becoming implemented. It’s likely the world takes the fertility crisis seriously by 2035-2040, but the reaction could either be to expand reproductive freedoms (via research and access into embryogenesis and artificial wombs) or to enforce greater fertility via social norms, while still being quite restrictive on biological innovations. Human augmentation will likely either be pushed to the fringes of the world, underground, or in developing nations during this time: as more bio-infrastructure accumulates in the fringes of the world, the risk of e.g. engineered pandemics increase, and during a great power conflict this might lead to more global instability generally. Worth analyzing the likelihood of this.

Other topics: in-depth analysis of nanomanufacture, how infrastructure will develop over time, likelihood of robotic automation of large variety of tasks, whether or not UBI will be implemented as developed nations asymptote towards 99% automation, whether or not the necessary social safety nets necessary for labor automation will still be in place during a great power conflict, and whether or not gradual disempowerment is still a failure mode to worry about. My guess for the latter is not really, and it’s moreso that access to superpersuasive AI agents will allow humans to warp the culture much more without the AIs really taking over: also that it’s unlikely that AIs will e.g. “run the corporations” but might substitute for large portions of middle-managers; and also that really extensive automation is predicated on good societal infrastructure for the AIs to work, given that the AIs are not smart enough to create their own infrastructure and manufacturing supply chains yet and they’re not coordinating global politics (yet).

I started reading The Human Stain from a quote on Twitter. Sadly, I cannot find it as of right now, but it was a pertinent and apt comparison of French and American intellectualism (thanks Uzay).

Aha! Here it is:

. . . and knows something about these ambitious French kids trained in the elite lycées, Extremely well prepared, intellectually well connected, very smart immature young people endowed with the most snobbish French education and vigorously preparing to be envied all their lives, they hang out every Saturday night at the cheap Vietnamese restaurant on rue St. Jacques talking about great things, never any mention of trivalities or small talk—ideas, politics, philosophy only. Even in their spare time, when they are all alone, they think only about the reception of Hegel in twentieth-century French intellectual life. The intellectual must not be frivolous. Life only about thought. Whether brainwashed to be aggressively Marxist or to be aggressively anti-Marxist, they are congenitally apalled by everything American . . .

But in America, no one appreciates the special path she was on in France and its enourmous prestige. She's not getting the type of recognition she was trained to get as a member of the French intellectual elite. She's not even getting the kind of resetnment she was trained to get . . . Her fellow foreign graduate students tell her that she's too good for Athena College, it would be too déclassé, but her fellow American graduate students, who would kill for a job teaching in the Stop & Shop boiler room, think that her uppityness is characteristcally Delphine.

Foils make or break narratives. Delphine Roux—a young intellectual heiress who grew up fed by a silver spoon—is the perfect counterpart to Coleman Silk, an old boxer always an outsider to the circles he traveled in. Both are not quite a part of the American academic elite: Delphine is too French and Coleman is Black and light-skinned (and functionally a Jew). Yet they manage to find themselves on opposite sides of a cultural divide. Professor Silk is an archetypal Tyler Cowen classic 'conservative' (as is Kareem Abdul-Jabbar), a professed believer in meritocracy and fair and equal consideration. Professor Roux is an ardent anti-racist and feminist, and while the two are not inherently opposed to each other, in practice (and as described) those who possess such beliefs are predisposed to clash.

Says Silk to Roux:

"Providing the most naive of readers with a feminist perspective on Euripides is one of the best ways you could devise to close down their thinking before it's even had a chance to begin to demolish a single one of their brainless 'likes'. I have trouble believing that an educated woman coming from a French academic backbround like your own believes there is a feminist perspective on Euripides that isn't simply foolishness. Have you really been edified in so short a time, or is this just old-fashioned careerism grounded right now in the fear of one's feminist colleagues? Because if it is just careerism, it's fine with me. It's human and I understand. But if it's an intellectual committment to this idiocy, then I am mystified, because you are not an idiot. Because you know better. Because in France surely nobody from the École normale would dream of taking this stuff seriously. Or would they? To read two plays like Hippolytus and Alcestis, then to listen to a week of classroom discussion on each, then to have nothing to say about either of them other than that they are 'degrading to women,' isn't a 'perspective,' for Christ's sake—it's mouthwash. It's just the latest mouthwash."

(Silk is a classicist)

When Silk called a pair of chronic absenteers 'spooks', an uproar was raised and he resigned from his faculty post at Athena College. Subsequently, Roux took one of the victims, Tracy, under her wing for a while, but she subsequently failed out of all of her classes and moved out of the city to stay with a half-sister in Tulsa, Oklahoma. When she finds out of his affair with the believed illiterate, divorced, janitor Faunia Farley who just had her two children die in a tragic fire, she labels Silk a predator and attempts to destroy his reputation.

It is important to note that Silk, at this point in time, has invented a Jewish background. His deceased wife was a Jew, and his newfound family never knew of his Black ancestry. He has been cut off from his ancestral family as a result of his successful attempts to 'pass' as white. It is perhaps for this reason that the attempt to oust him based on his perceived racism succeeded, and also a great source of dramatic irony.

Farley is brilliant. She is strong-willed, independent, yet broken from the deaths of her children. She blames herself for the death of her kids because she chose to save her boyfriend from the flaming fire first. She is old, much older than her years of thirty-four would indicate, and potentially older than Coleman.

"I see you, Coleman. You're not closing the doors. You still have the fantasies of love. You know something? I really need a guy older than you. Who's had all the love-shit kicked out of him totally. You're too young for me, Coleman. Look at you. You're just a little boy falling in love with your piano teacher. You're falling for me, Coleman, and you're much too young for the likes of me. I need a much older man. I think I need a man at least a hundred. Do you have a friend in a wheelchair you can introduce me to? Wheelchairs are ok—I can dance and push. Maybe you have an older brother. Look at you, Coleman. Looking at me with those schoolboy eyes. Please, please, call your older friend. I'll keep dancing, just get him on the phone. I want to talk to him."

Her ex-husband, Lester Farley, also blames her for the death of his kids. He's a Vietnam war veteran, and we get some fun segues into a VA therapist's attempts at curing him of his PTSD by repeatedly taking him to Chinese restaurants as exposure therapy. Ultimately, he likely murders the couple of Silk and Farley by driving a car into theirs, sending them toppling over a cliff.

I cannot do justice to the multiplicity of narratives Roth weaves throughout this book. Silk's journey from Black to Jew, Zuckerman's (the narrator, the entire book is framed) realization of Silk's past, Farley grappling with the loss of her children, the other Farley grappling with the demons of Vietnam, digressions into intellectualism through Roux as below:

Narrative structure and temporality. The internal contradictions of the work of art. Rousseau hides himself and then his rhetoric gives himself away. (A little like her, thinks the dean, in that autobiographical essay.) The critic's voice is as legitimate as the voice of Herodotus. Narratology. The diegetic. The differences between diegesis and mimesis. The bracketed experience. The proleptic quality of the text.

Above all, it is fun to read. At one point Coleman is a crow.

The title, from Silk's funeral:

"That's what comes of being hand-raised," said Faunia. "That's what comes of hanging around all his life with people like us. The human stain," she said, and without revulsion or contempt or even condemnation. Not even with sadness. That's how it is. . .

Perhaps the best section of the book deals with Milton's taxonomy of worthy poems to write, namely:

(A) Epic.
 I.
  (a) The diffuse Epic [Homer, Virgil, and Tasso].
  (b) The brief Epic [the Book of Job].
 II.
  (a) Epic keeping the rules of Aristotle.
  (b) Epic following Nature.
 III. Choice of subject ['what king or knight before the conquest'].
(B) Tragedy.
 (a) On the model of Sophocles and Euripides.
 (b) On the model of Canticles or the Apocalypse.
(C) Lyric.
 (a) On the Greek model ['Pindarus and Callimachus'].
 (b) On Hebrew models ['Those frequent songs throughout the Law and the Prophets'].

(this is from his Reason of Church Government, src)

Notably, each category draws from both Classical and Scriptural examples, and some fit better than others. The Canticles as tragedies?? They're psalms! for heaven's sake. Revelation (the Apocalypse of St. John) is slightly more sane in this regard, but really still in quite a different class from Euripides.

Milton was apparently going to write an Arthuriad instead of Paradise Lost. As a fan of Arthuriana, I can't say he made the correct call. . ., but tragedies are preferable to comedies.

Milton's hesitation between the classical and the ro­mantic types of epic is one more instance of something which runs through all his work; I mean the co-existence, in a live and sensitive tension, of apparent opposites. We have already noted the fusion of Pagan and Biblical interests in his very map of poetry. We shall have occasion, in a later section, to notice, side by side with his rebelliousness, his individualism, and his love of liberty, his equal love of discipline, of hierarchy, of what Shakespeare calls 'degree'. From the account of his early reading in Smectymnuus we gather a third tension. His first literary loves, both for their style and their matter, were the erotic (indeed the almost pornographic) elegiac poets of Rome: from them he graduated to the idealized love poetry of Dante and Petrarch and of 'those lofty fables which recount in solemn cantos the deeds of knighthood': from these to the philo­ sophical sublimation of sexual passion in 'Plato and his equal (i.e. his contemporary) Xenophon'. An original voluptuous­-ness greater, perhaps, than that of any English poet, is pruned, formed, organized, and made human by progressive purifica­tions, themselves the responses to a quite equally intense aspiration-an equally imaginative and emotional aspiration-towards chastity. The modern idea of a Great Man is one who stands at the lonely extremity of some single line of de­velopment--one either as pacific as Tolstoi or as military as Napoleon, either as clotted as Wagner or as angelic as Mozart. Milton is certainly not that kind of great man. He is a great Man. 'On ne montre pas sa grandeur,' says Pascal, 'pour etre a une extremite, mais bien en touchant les deux a la jois et remplissant tout l'entre-deux.'

Catholicism cannot exist without paganism, and all that.

Lewis critiques Eliot's belief that 'great poets are the only judges of great poetry' on simple grounds: how do you recognize the great poets? But, Eliot-sim replies, only good men can judge goodness, and only doctors can judge medical skill. Yet all humans are subject to moral law, while not all humans are deemed to be judged to be a poet or not a poet. As for the other, poets have the right to judge the skill of a poet, but not the value of a poet---that is reserved for the readers.

An epic is either Homeric or Virgilian.

The Homeric (Primary) Epic is a performance. Wholly oral, never read or written, meant for an heroic court. Characterized by the Middle English solempne, solemnity yet lacking the implied gloom, oppression, or austerity. The opening feast of Gawain and the Green Knight is a solemnity in this sense. And while the Iliad and Odyssey are too long to be recited as a whole, they are still written in a recitative style.

Aesthetically, the Primary Epic "emphasiz[es] the unchanging human environment." Imagine onomatopoeia, yet imbued in the entire work. "There is no use in disputing whether any episode could really have happened. We have seen it happen." Beowulf takes a more romantic view, "its landscapes have a spiritual quality", whereas the Iliad is objective, possessed with more of a sense of good and evil.

Primary epic does not inherently deal with a great subject. Odysseus is the king of a small country. The telling of the Trojan War is a front for the story of Achilles and Hector.

The truth is that Primary Epic neither had, nor could have, a great subject in the later sense. That kind of greatness arises only when some event can be held to effect a profound and more or less permanent change in the history of the world, as the founding of Rome did, or still more, the fall of man. Before any event can have that significance, history must have some degree of pattern, some design. The mere endless up and down, the constant aimless alternations of glory and misery, which make up the terrible phenomenon called a Heroic Age, admit no such design. No one event is really very much more important than another. No achievement can be permanent: today we kill and feast, tomorrow we are killed, and our women led away as slaves. Nothing 'stays put', nothing has a significance beyond the moment. Heroism and tragedy there are in plenty, therefore good stories in plenty; but no 'large design that brings the world out of the good to ill'.

. . .

Primary Epic is great, but not with the greatness of the later kind. In Homer, its greatness lies in the human and personal tragedy build up against this background of meaningless flux. It is all the more tragic because there hangs over the heroic world a certain futility.

Beowulf animates this despair: heroes fight literal monsters as well as man.

The Virgilian (Secondary) Epic innovates. No longer was the focus on the great and timeless, instead, it was on the great and revolutionary (revolutionary by virtue of interfacing with reality). It is here we take our modern-day understanding of heroes from. Achilles raged against the world, and the world didn't care. He died less than a martyr---he was dispensible. Aeneas raged against the world, and built an empire. It is through his story that the greatest themes are invoked, and the Romans understood their world through the lens of his story.

It's a bit weird that the Romans cared about something less pure than the Greeks, and as a result they grew up. Plausibly this was the genesis of "caring"? When the Romans were "compelled to see something more important than happiness" they invented a hero, Aeneas, who sought more than happiness, who would not bow, who would not accept.

A great deal of what is mistaken for pedantry in Milton (we hear too often of his 'immense learning') is in reality evocation. If Heaven and Earth are ransacked for simile and allusion, this is not done for display, but in order to guide our imaginations with unobtrusive pressure into the channels where the poet wishes them to flow; and as we have already seen, the learning which a reader requires in responding to a given allusion does not equal the learning Milton needed to find it.