flâneur

Samuel Lo

5 followers · 9 following · 799 views

on the atlas — 166

highlights — 301

  • Getting to witness this at all is in part due to the writing and reading culture in the company. There’s a robust book recommendation pipeline, multiple team-wide strategy memos, and even the engineers write long treatises on new agentic architecture. There’s a library of books everyone has read, including one written by a Crosby employee, and an unlimited book budget.
    Crosby: The Inner Workings of a Neofirm - by Nix
  • He’s currently working on a playbook schema for the eval system, and LLM as a judge system for grading Crosby’s AI outputs as compared against what a reviewer actually sent out.
    Crosby: The Inner Workings of a Neofirm - by Nix
  • He likes subversion and is a fan of little delights – once he hired a band off the streets of New York to play late night jazz in the office on a random Tuesday night. In Slack, there’s a channel called #ryan-diary where he discusses the books he is reading in great detail.
    Crosby: The Inner Workings of a Neofirm - by Nix
  • So now you’ve got kids in India who assume Jordan Castro is a major writer in America. These kids are tired of Arundhati Roy or whoever they’ve been told to like—Roy is done, she’s old news. So instead they start writing these highly-compressed one-day novels, with faintly-ridiculous, pathetic narrators who have sex with famous people, in the style of that great genius, Jordan Castro.
    He wanted more attention - by Naomi Kanakia
  • Why didn’t these young writers glom onto Beckett instead? Well...because Beckett’s novels never really broke out and found the readership that Woodcutters did, so there was no reason for any corporation to think that putting money into him would lead to good results. As a result, Beckett’s novels are published, in America, by Grove, a small press, with less distribution, less publicity.
    He wanted more attention - by Naomi Kanakia
  • Culture hasn’t stagnated. What’s happened is that now we can accurately perceive that it’s very difficult to for anything to be truly new. That’s because every author throughout history has iterated upon what’s come before. They didn’t invent anything de novo. In many cases, their formal achievement only appears singular, because we are not familiar with their influences (oftentimes those influences have been lost or simply aren’t read).
    He wanted more attention - by Naomi Kanakia
  • My personal opinion is that nothing is new, and nothing has ever been new, and the notion of the avant-garde is bankrupt.
    He wanted more attention - by Naomi Kanakia
  • Whereas other people, who are against ‘brodernism’, against ‘world literature’, against ‘the zombie avant garde’, against ‘aping a great writer’s style’, these people really have nothing to offer. Oftentimes these critics of the Neo-Bernhardians are just people who think that everything written today is bad, and that we are in civilizational decline. Oscar Dorr, in this House House article, says it openly:
    He wanted more attention - by Naomi Kanakia
  • Yes, he sometimes says the kind of stuff the other neo-Bernhardians say, about how MFA programs are passe (he doesn’t have an MFA) and how domestic realism is tired, but...it seems to lack conviction. There’s a good-natured quality to Haber. He’s just a guy who’s found his thing, and he’s happy with that. If you criticize brodernism or devotees of ‘world literature’ or the ‘zombie avant-garde,’ then you’re essentially criticizing Mark Haber. But why would you ever do that? To say ‘Mark Haber is a bad writer’ would just be an obviously incorrect statement.
    He wanted more attention - by Naomi Kanakia
  • Perelmutter would say that much of what people see in Bernhard—the overt difficulty and iconoclasm—is an illusion. And I would agree, to some extent. People in America read Bernhard and take him on face value when he says that he doesn’t want prizes, doesn’t want attention, doesn’t care about institutions or literary fame. But people in Austria understood that this was not totally correct. He was a writer who took plenty of money from institutions (right up until his death, his plays continued to be staged by Austria’s state-funded Burgtheater). He was not a writer with strong principles: he w…
    He wanted more attention - by Naomi Kanakia
  • Much of the postwar European fiction, some of it very good, that we might read as World Literature — Perec, Bernhard, Nádas, Nooteboom, Jelinek, Marías, Sebald, now Knausgård — is extremely psychological in character and only vestigially social and geographical. Typically the narrator is a monologist, resembling the author, who tells of personal turmoil amid social stasis. He recognizes himself, with snobbish self-approbation, as a part of a stable polyglot pan-European elite; most other inhabitants of his country, as of the neighboring ones, are unthreatening idiots who turn on the TV after r…
    He wanted more attention - by Naomi Kanakia
  • Most concerning, for me, is that brodernist criticism interprets foreignness itself as “difficult.” The brodernist measures of “challenging” literature are a constrained series of attributes: winding sentences, explicit references to entropy and math and classical music, metanarration, anti-realism, lonely and existential male protagonists, brick-length tomes, cringey and misogynistic sex scenes, the stench of once having read Nietzsche, the words “psychedelic” and “oneiric,” half-hearted genre-play. Kafka, poor guy, gets thrown around a lot. The problem, to be clear, is not these traits—hones…
    Against High Brodernism | Los Angeles Review of Books
  • “[i]t is probably embarrassingly clear by now that works of art are my objects of worship,” and, moreover, “works of art are often more real than we are because they embody human consciousness completely fulfilled”
    William H. Gass’s Transformative Translations of Rilke | 12 Winters Blog
  • I am shaping your thoughts as much as you are shaping mine. It is a great responsibility to be able to alter someone’s internal model of reality in their brain. It is also a great pleasure to perpetually forge new ideas through a collaborative design process. It really does take two to think. We should be so lucky to find just one other person to think with and talk to.
    We think better when our minds are entwined
  • The highest task of a bond between two people, as poet Rainer Maria Rilke wrote, is that each should stand guard over the solitude of the other. I see that in Smith and Mapplethorpe’s story - they were the center of one another’s consciousnesses. Perhaps they drifted from each other over the span of years, but always found themselves drawn back that odd combination of fate and will.
    We think better when our minds are entwined
  • I think Demis is a wonderful example of our culture’s Faustian pact, this thirst for knowledge. All our stories ask, “Should I pick this cup, drink it, live forever, and know everything? Or should I just be this human thing?” Wisdom has always said to leave that to the gods. Leave it to the gods. You are not immortal and you are not all-knowing, and that is what makes you precious. You are precious because you’re weak; you’re limited. We disabused ourselves of the notion that we will live forever. We’re living in this scary time, so let’s be a little bit more human.
    Conversations on Human Culture in the AI Age
  • We’re all drunk on these words, ‘super’, ‘ultra’, and they just obfuscate the fact that there are ways of knowing that are not intelligence-based. There are lived processes that affect everything about you. We are not this brain in a jar. It’s amazing that we’ve managed to prove this hypothesis that intelligence is not substrate-dependent. That’s fine. It doesn’t take anything away from the fact that we are more than that. How about they start thinking about a super loving being or a super sexy being?
    Conversations on Human Culture in the AI Age
  • Literature tries to weave the rainbow back together. It involves irrationality; it involves all of those things that science has, by its own method, left out. Literature tries to put it back in, so it presents a messier, darker, and perhaps more complete, if less powerful, perspective on the world.
    Conversations on Human Culture in the AI Age
  • This link to Beckett, who more than anyone continued the project of serious art (at least on a conceptual level) more earnestly than anyone since Schoenberg, is illuminating: when this book was written (early 60s), Beckett had not yet completed his oeuvre, but each of his books is an abstract purification of the previous, beginning all the way from Murphy and concluding with the final ‘trilogy’1 in the 80s, which concludes with a semi-coherent burst of illiterate minimalism. An implication common in Beckett’s last works, beside a depressive (if qualified by irony) pessimism about both life & a…
    Adorno & the Problem of Modern Art - Gyres
  • Adorno’s estimation of contemporary artists is middling, in favor of only a couple of modern classical composers, skeptical about the visual arts, and extremely pessimistic about the world of letters, except for the works of Samuel Beckett, to whom this book was to be dedicated.
    Adorno & the Problem of Modern Art - Gyres
  • Most of all, we must demand that people ask us for the things that they have anxieties about. Are you anxious about AI and employment? Force us to share economic data. Are you anxious about mental health and child safety? Force us to monitor for this on our platforms and share data. Are you anxious about misaligned AI systems? Force us to publish details on this.
    Import AI 431: Technological Optimism and Appropriate Fear
  • For us to truly understand what the policy solutions look like, we need to spend a bit less time talking about the specifics of the technology and trying to convince people of our particular views of how it might go wrong - self-improving AI, autonomous systems, cyberweapons, bioweapons, etc. - and more time listening to people and understanding their concerns about the technology. There must be more listening to labor groups, social groups, and religious leaders. The rest of the world which will surely want—and deserves—a vote over this.
    Import AI 431: Technological Optimism and Appropriate Fear
  • If adopted as a standard across frontier labs, we hope this might create a “race to the top” dynamic where competitive incentives are directly channeled into solving safety problems.
    Anthropic's Responsible Scaling Policy \ Anthropic
  • Anthropic is taking an empirically-driven approach to AI safety. Some of the key areas of active work include improving our understanding of how AI systems learn and generalize to the real world, developing techniques for scalable oversight and review of AI systems, creating AI systems that are transparent and interpretable, training AI systems to follow safe processes instead of pursuing outcomes, analyzing potential dangerous failure modes of AI and how to prevent them, and evaluating the societal impacts of AI to guide policy and research. By attacking the problem of AI safety from multiple…
    Core Views on AI Safety: When, Why, What, and How \ Anthropic
  • In optimistic scenarios, (1) will help AI developers to train beneficial systems and (2) will demonstrate that such systems are safe. In intermediate scenarios, (1) may be how we end up avoiding AI catastrophe and (2) will be essential for ensuring that the risk posed by advanced AI is low. In pessimistic scenarios, the failure of (1) will be a key indicator that AI safety is insoluble and (2) will be the thing that makes it possible to convincingly demonstrate this to others.
    Core Views on AI Safety: When, Why, What, and How \ Anthropic
  • We have no time to lose.
    Dario Amodei — The Adolescence of Technology
  • But we will need to step up our efforts if we want to succeed. The first step is for those closest to the technology to simply tell the truth about the situation humanity is in, which I have always tried to do; I’m doing so more explicitly and with greater urgency with this essay. The next step will be convincing the world’s thinkers, policymakers, companies, and citizens of the imminence and overriding importance of this issue—that it is worth expending thought and political capital on this in comparison to the thousands of other issues that dominate the news every day. Then there will be a t…
    Dario Amodei — The Adolescence of Technology
  • We see early hints of this in the concerns about AI psychosis, AI driving people to suicide, and concerns about romantic relationships with AIs. As an example, could powerful AIs invent some new religion and convert millions of people to it? Could most people end up “addicted” in some way to AI interactions? Could people end up being “puppeted” by AI systems, where an AI essentially watches their every move and tells them exactly what to do and say at all times, leading to a “good” life but one that lacks freedom or any pride of accomplishment? It would not be hard to generate dozens of these …
    Dario Amodei — The Adolescence of Technology
  • Related to this, a public backlash against AI is brewing: this could be a corrective, but it’s currently unfocused. Much of it targets issues that aren’t actually problems (like datacenter water usage) and proposes solutions (like datacenter bans or poorly designed wealth taxes) that wouldn’t address the real concerns. The underlying issue that deserves attention is ensuring that AI development remains accountable to the public interest, not captured by any particular political or commercial alliance, and it seems important to focus the public discussion there.
    Dario Amodei — The Adolescence of Technology
  • Where should the line be drawn? In the list at the beginning of this section, two items—using AI for domestic mass surveillance and mass propaganda—seem to me like bright red lines and entirely illegitimate.
    Dario Amodei — The Adolescence of Technology
  • My worry is that I’m not totally sure we can be confident in the nuclear deterrent against a country of geniuses in a datacenter: it is possible that powerful AI could devise ways to detect and strike nuclear submarines, conduct influence operations against the operators of nuclear weapons infrastructure, or use AI’s cyber capabilities to launch a cyberattack against satellites used to detect nuclear launches.33 Alternatively, it’s possible that taking over countries is feasible with only AI surveillance and AI propaganda, and never actually presents a clear moment where it’s obvious what is g…
    Dario Amodei — The Adolescence of Technology
  • I expect AI-led cyberattacks to become a serious and unprecedented threat to the integrity of computer systems around the world, and Anthropic is working very hard to shut down these attacks and eventually reliably prevent them from happening. The reason I haven’t focused on cyber as much as biology is that (1) cyberattacks are much less likely to kill people, certainly not at the scale of biological attacks, and (2) the offense-defense balance may be more tractable in cyber, where there is at least some hope that defense could keep up with (and even ideally outpace) AI attack if we invest in …
    Dario Amodei — The Adolescence of Technology
  • To put it another way, renting a powerful AI gives intelligence to malicious (but otherwise average) people. I am worried there are potentially a large number of such people out there, and that if they have access to an easy way to kill millions of people, sooner or later one of them will do it. Additionally, those who do have expertise may be enabled to commit even larger-scale destruction than they could before.
    Dario Amodei — The Adolescence of Technology
  • Overall, I am optimistic that a mixture of alignment training, mechanistic interpretability, efforts to find and publicly disclose concerning behaviors, safeguards, and societal-level rules can address AI autonomy risks, although I am most worried about societal-level rules and the behavior of the least responsible players (and it’s the least responsible players who advocate most strongly against regulation). I believe the remedy is what it always is in a democracy: those of us who believe in this cause should make our case that these risks are real and that our fellow citizens need to band to…
    Dario Amodei — The Adolescence of Technology
  • The unique value of interpretability is that by looking inside the model and seeing how it works, you in principle have the ability to deduce what a model might do in a hypothetical situation you can’t directly test—which is the worry with relying solely on constitutional training and empirical testing of behavior. You also in principle have the ability to answer questions about why the model is behaving the way it is—for example, whether it is saying something it believes is false or hiding its true capabilities—and thus it is possible to catch worrying signs even when there is nothing visibl…
    Dario Amodei — The Adolescence of Technology
  • Anthropic has just published its most recent constitution, and one of its notable features is that instead of giving Claude a long list of things to do and not do (e.g., “Don’t help the user hotwire a car”), the constitution attempts to give Claude a set of high-level principles and values (explained in great detail, with rich reasoning and examples to help Claude understand what we have in mind), encourages Claude to think of itself as a particular type of person (an ethical but balanced and thoughtful person), and even encourages Claude to confront the existential questions associated with i…
    Dario Amodei — The Adolescence of Technology
  • I make all these points to emphasize that I disagree with the notion of AI misalignment (and thus existential risk from AI) being inevitable, or even probable, from first principles. But I agree that a lot of very weird and unpredictable things can go wrong, and therefore AI misalignment is a real risk with a measurable probability of happening, and is not trivial to address.
    Dario Amodei — The Adolescence of Technology
  • But it is a world worth fighting for. If all of this really does happen over 5 to 10 years—the defeat of most diseases, the growth in biological and cognitive freedom, the lifting of billions of people out of poverty to share in the new technologies, a renaissance of liberal democracy and human rights—I suspect everyone watching it will be surprised by the effect it has on them. I don’t mean the experience of personally benefiting from all the new technologies, although that will certainly be amazing. I mean the experience of watching a long-held set of ideals materialize in front of us all at…
    Dario Amodei — Machines of Loving Grace
  • The opt-out problem. One concern in both developed and developing worlds alike is people opting out of AI-enabled benefits (similar to the anti-vaccine movement, or Luddite movements more generally). There could end up being bad feedback cycles where, for example, the people who are least able to make good decisions opt out of the very technologies that improve their decision-making abilities, leading to an ever-increasing gap and even creating a dystopian underclass (some researchers have argued that this will undermine democracy, a topic I discuss further in the next section).
    Dario Amodei — Machines of Loving Grace
  • Most mental illness can probably be cured. I’m not an expert in psychiatric disease (my time in neuroscience was spent building probes to study small groups of neurons) but it’s my guess that diseases like PTSD, depression, schizophrenia, addiction, etc. can be figured out and very effectively treated via some combination of the four directions above.
    Dario Amodei — Machines of Loving Grace
  • Beyond the model work, Eigen’s secret weapon and only non-engineer for now is Kyle, a former LA television writer, now responsible for doing much of his “character work.” This shapes his personality, communication style, diction, preferences, and stories. Kyle doesn’t seem like a typical techie. She has very long pointy fingernails painted jet black and yet still types like the wind with a symphony of clicks and clacks on the keyboard. She tells me she “comes to work and method acts all day.”
    Entering Eigenworld - by Brie - early days
  • I believe that in the AI age, we should be talking about the marginal returns to intelligence7, and trying to figure out what the other factors are that are complementary to intelligence and that become limiting factors when intelligence is very high. We are not used to thinking in this way—to asking “how much does being smarter help with this task, and on what timescale?”—but it seems like the right way to conceptualize a world with very powerful AI.
    Dario Amodei — Machines of Loving Grace
  • country of geniuses in a datacenter”
    Dario Amodei — Machines of Loving Grace
  • I am often turned off by the way many AI risk public figures (not to mention AI company leaders) talk about the post-AGI world, as if it’s their mission to single-handedly bring it about like a prophet leading their people to salvation. I think it’s dangerous to view companies as unilaterally shaping the world, and dangerous to view practical technological goals in essentially religious terms.
    Dario Amodei — Machines of Loving Grace
  • A simple way to put this is: How many ideas does one person have in a day, and how many of those things do they actually get to do? Until that proportion is 100%, you know there is a pretty meaningful bottleneck in terms of the drudgery of execution. I don’t think it’s anywhere close to 100%—it’s probably like 5%. You have all sorts of different ideas of things you’d want to do or build or try, and in practice it’s much harder to actually do them.
    The Wu Tapes: Q&A with Cognition's Scott Wu - Colossus
  • I think people will be able to adapt and we will find all sorts of things where having new technologies is going to make our lives better—for drug discovery, for making pizza taste perfect and also perfectly nutritious, for all the technology we use every day. But it really does mean that folks need to be on top of what’s going on. In a lot of previous revolutions, if you adopted two or five years later, you were still mostly in time. In this case, it has to happen faster.
    The Wu Tapes: Q&A with Cognition's Scott Wu - Colossus
  • More seriously, I think this shift from survival mode to creative mode is the right one. It’s not going to be in the next two or four years, but it’s coming. At some point you really will only be capped by what you can imagine and the ideas you have for how you want to spend your life, which will be a very different life. To your question, I think a lot of it will be pure self-expression. People will get to live the lives they want to live. The bottleneck to turning our dreams into reality will dramatically narrow. And the way we live now will seem crazy.
    The Wu Tapes: Q&A with Cognition's Scott Wu - Colossus
  • Here’s the analogy I would give. If you think about all of our ancestors from hundreds or thousands of years ago—what would they think of us? They see us pushing buttons on a computer, then getting in a room with other people and just talking, and we call that a meeting, and that’s what we call work. Meanwhile, they’re out in the field seven days a week making sure they have enough food to eat.
    The Wu Tapes: Q&A with Cognition's Scott Wu - Colossus
  • I enjoy the fact that Hemingway is one of the Ando team’s internal references for how an agent’s output should be read - recognizable for his calm, direct sentences.
    Ando: Building Slack from Scratch — New Ontologies
  • One concept in conversations I overheard frequently was the “bpm of human conversation,” the way meaning is shaped by text length, word choice, cadence, and interchange between participants. If an agent interjects at the wrong moment or in the wrong register, it is immediately unpleasant. After all, in a chat interface where you never see someone’s face, words are the entire basis of understanding and experience.
    Ando: Building Slack from Scratch — New Ontologies