https://vegasharmon.com/tags/ai/ Recent content in Ai on Vegas Harmon Hugo en Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/ai-kicks-away-the-ladder-of-mastery/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/ai-kicks-away-the-ladder-of-mastery/ The digital era has ushered in a further phase of the technological destruction of tradition. Whereas some kinds of automation create new demand for higher-skilled work, digital automation based on even rudimentary forms of artificial intelligence can instead increase demand for low-skill work pushing buttons. Training on the collective efforts of thousands of years of culture, machine-learning algorithms aim to supplant human performance for classification, writing, drawing, coding, driving, making music, and a host of other practices. https://vegasharmon.com/commonplace/alignment-exists-because-ml-never-became-a-science/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/alignment-exists-because-ml-never-became-a-science/ One reason people are confused about AI alignment losing its ability to make scientific progress is that machine learning as a whole is also not a very scientific field by the standard I’m applying. Even most early AI researchers were more focused on building artificial intelligence than on understanding scientific principles of cognition. The rise of deep learning exacerbated this problem, as throwing more compute and engineering effort at an AI became arbitrarily scalable. https://vegasharmon.com/commonplace/clown-against-ai/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/clown-against-ai/ Others voiced fears about A.I.’s effects on creativity, about the way it might etiolate the visual and artistic landscape. A lanky thirty-year-old novelist, filmmaker, and practicing clown who went by Red Lamb (he had been to clown school in France, he said) told me that he was “not a big policy-head” but that he dreaded the “slopification of the arts.” “No clanker has ever actually told a joke that I found amusing,” he said. https://vegasharmon.com/commonplace/dignity-isn-t-a-check-why-ubi-can-t-replace-work/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/dignity-isn-t-a-check-why-ubi-can-t-replace-work/ This lost labor simply cannot be compensated for by transferring money from the AI winners to the AI losers, in the form of, for example, universal basic income. A job and income mean much more than support for the material basis of life. Having a paying job means that the surrounding society values your labor sufficiently to compensate you for it, and provides, in addition to material goods, a sense of dignity and self-worth. https://vegasharmon.com/commonplace/each-gpt-beats-the-last-so-much-for-intelligence-being-too-slippery/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/each-gpt-beats-the-last-so-much-for-intelligence-being-too-slippery/ But I don’t need you to accept my speculative philosophical research program - only the drop-dead obvious fact that GPT-6 is better than GPT-2 in every way, and the same process that produced these improvements will soon be used to produce GPT-7, 8, and 9. Once we accept this empirical argument, what is left of the philosophical argument that intelligence is too slippery and conflationary to work with? https://vegasharmon.com/commonplace/evolution-s-purpose-humanity-as-superintelligence-s-birth-canal/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/evolution-s-purpose-humanity-as-superintelligence-s-birth-canal/ However, somewhat surprisingly, “creating digital superintelligence” has all of the features of a “real” meaning of life, with no supernatural elements needed to square the circle. In the most minimal sense, the “need” for evolution to create digital intelligence explains why humans are here at all, in the particular form that we are, and what we are “made to do” with our time on Earth. And the expected outcome of creating digital superintelligence is certainly cosmic-scale; even the AI “doomers” — people who believe that creating superintelligent AI will likely result in human extinction — think that our AIs will fulfill the long-term science fiction goal of “colonizing the universe,” and fill the cosmos with something. https://vegasharmon.com/commonplace/extropian-dreams-paperclip-nightmares/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/extropian-dreams-paperclip-nightmares/ In the early days of Silicon Valley, it was actually quite common (at least among far-thinking futurist types who populated the Bay Area) to believe that the coming Singularity would be the single greatest event in the history of the universe, the culmination of the techno-capitalist project that has already done so much good for humanity. In the 1980s and 1990s, the “extropian” community in the Bay Area (a futurist collective committed to fighting entropy and death, which organized around local meetups, an academic institute and a popular mailing list) eagerly awaited the Singularity, believing that a glorious transhuman future — free from all death, suffering, and scarcity — awaited them on the other side. https://vegasharmon.com/commonplace/hardware-overhang-work-pulls-us-toward-the-singularity/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/hardware-overhang-work-pulls-us-toward-the-singularity/ Insofar as they’re right, all work which was (explicitly or implicitly) justified by the idea of reducing the hardware overhang has been directly pulling us towards the singularity. https://vegasharmon.com/commonplace/humanity-as-bootloader-searching-for-meaning-we-may-never-grasp/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/humanity-as-bootloader-searching-for-meaning-we-may-never-grasp/ It is possible that there is no “meaning of life,” no ultimate “point,” in which case, we did no harm by trying to find one. It is possible that the “meaning of life” is in fact something very simple: like “love,” or “beauty,” or “joy” — in which case, this will have been a long journey to “know a place for the very first time,” and we will have to trust our AIs to realize some aspect of the meaningfulness of human life more fully. https://vegasharmon.com/commonplace/intelligence-keeps-climbing-until-it-passes-us/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/intelligence-keeps-climbing-until-it-passes-us/ GPT-6 is more intelligent than GPT-3. You can split hairs, you can come up with some obscure sense in which that word is slightly off, but common-sensically it’s true. In the same way, GPT-9 will be more intelligent than GPT-6. At some point, this intelligence will exceed human intelligence. My argument requires nothing more complicated or philosophically dubious than this. https://vegasharmon.com/commonplace/lookup-without-understanding/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/lookup-without-understanding/ “Artificial intelligence” is no more intelligent than a pocket calculator or a library: just because both can be used to quickly find something out, it does not mean that there is an intelligence there that understands something. This seems wrong. But why? The first reason that comes to mind is that AI is not as passive as a calculator or library. In contrast to a calculator, an LLM’s outputs are underspecified by human inputs. https://vegasharmon.com/commonplace/machine-gods-the-abstraction-that-eats-the-debate/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/machine-gods-the-abstraction-that-eats-the-debate/ Here are three propositions. Metaphors are the fundamental abstraction layers in modern AI debates. We are in a moment where you could plausibly argue that Eliezer Yudkowsky and Nick Land are the most influential policy intellectuals in the US conversation. Just look at the state of the two of them. This is not a good thing. The collision of (1), (2), and recent events means that “machine gods” have now entered the discussion. https://vegasharmon.com/commonplace/peer-review-isn-t-the-point-scientific-inquiry-is/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/peer-review-isn-t-the-point-scientific-inquiry-is/ A lot of the blame should fall on misguided narratives (common across academia) about what makes science work, which have been entrenched by the best-funded scientific institutions. A core scientific norm is that disputes should be resolved with reference to concrete empirical tests or rigorous proofs, judged by the scrutiny of one’s scientific peers. But that’s very different from the idea that ideas should be developed via paper-sized units of work which are each individually defended and justified. https://vegasharmon.com/commonplace/purpose-without-a-designer-why-building-superintelligence-is-our-meaning/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/purpose-without-a-designer-why-building-superintelligence-is-our-meaning/ However, there is also another, more functional sense of a “meaning of life,” one that preserves the clarity and singularity of the original question, but does not require life to have a creator in order for the question to have an answer. We can think of the “meaning of life” as the “purpose of life”: what life was “made for” and what life “in fact contributes” to our universe. And while the concept of “purpose” has many latent associations with intelligent design, in nature, many things — in fact, pretty much all things — functionally have a “purpose” without having been “designed” by anything at all (except, of course, the ongoing process of natural selection). https://vegasharmon.com/commonplace/science-without-scientists-knowledge-without-knowers/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/science-without-scientists-knowledge-without-knowers/ So far, the progress of science, the accomplishments of literature, have gone hand-in-hand with the formation of scientists and writers. But that consonance between human type and achieving a result has now been broken. The future we are looking at is one of more science but fewer, or no, scientists. Astrophysics without astrophysicists, history without historians—this is the future. https://vegasharmon.com/commonplace/self-preservation-without-selection-three-real-reasons-ai-seeks-power/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/self-preservation-without-selection-three-real-reasons-ai-seeks-power/ There are approximately three reasons why we think AI might have self-preservation and dominance drives, and none of them involve natural selection. First is convergent instrumental goals, second is misgeneralization, and third is reward-hacking. There is a large literature on each of these, which has since been proven prescient by AI breakouts like the Hugging Face incident. I would recommend reading this literature rather than continuing to assert that we are simply misunderstanding the role of natural selection. https://vegasharmon.com/commonplace/the-end-of-vocation-when-work-no-longer-matters/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/the-end-of-vocation-when-work-no-longer-matters/ A final way to think about the tragedy of AI advancing beyond us is that it spells the end of the possibility of having a vocation. That we are elevated above mere animalistic consumption by consistent work in a calling is one of the deepest features of civilized life; far from belonging only to geniuses, many who toil in complete anonymity get much of their satisfaction and sense of self from feeling that they are making a contribution through their work. https://vegasharmon.com/commonplace/the-fundamental-humanist-backlash-to-ai/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/the-fundamental-humanist-backlash-to-ai/ It is easy to see why the C.E.O.s of A.I. companies might promote this theory. Mistrust is a familiar force in American public life. Both A.I. populism and A.I. doomerism are, in their own ways, politics as usual. They correspond to two well-established modes of discourse—the one of white papers and probability distributions, the other of spirited (and justified) calls for redistribution. But if people hate A.I. because it is at odds with their vision of the good life—because they see it as fundamentally anti-human—then the existing categories won’t suffice. https://vegasharmon.com/commonplace/the-future-s-cities-etched-in-silicon/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/the-future-s-cities-etched-in-silicon/ We did build the great cities of the future: they are just very, very small. https://vegasharmon.com/commonplace/the-great-stagnation-lost-in-bits-abandoned-atoms/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/the-great-stagnation-lost-in-bits-abandoned-atoms/ It has become quite common to bemoan the fact that our “future” does not look very futuristic. And despite us having been living squarely in “the future” since the year 2000, it is true that people still have (functionally) the same houses, cars, clothing and aging bodies that human beings did a century ago. The investor and futurist Peter Thiel’s diagnosis is that sometime around 1970 (around the same time that the Digital Revolution began) humans simply gave up on building in the physical world: Thiel argues that humanity’s greatest builders got lost in the easy and addictive pull of information (first led by the internet, then apps, and finally social media), and so wasted decades innovating in the ephemeral world of “bits,” instead of building in the much more real and important world of “atoms. https://vegasharmon.com/commonplace/i-predict-that-you-will-counter-that-this-is/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/i-predict-that-you-will-counter-that-this-is/ I predict that you will counter that this is actually just [perfectly reasonable explanation of why an AI would think this way for instrumental reasons], and that when we dissect the perfectly reasonable instrumental reasons why AI would think this way, it will be identical to the convergent instrumental goals that we were trying to convince you of all along. https://vegasharmon.com/commonplace/the-machine-god-metaphor-how-staring-at-the-sun-blinds-debate/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/the-machine-god-metaphor-how-staring-at-the-sun-blinds-debate/ I could keep going, but I hope that the point is clear. The Machine God metaphor, which is really just a catchy rephrasing of longstanding notions among a particular community of AI people, has broken out of containment. It is leading, more or less by necessity, to a radical simplification of debate. If you believe that the machine gods are coming, so that super-empowered goal directed superintelligence is right around the corner, then the only thing that matters is ensuring that this superintelligence’s goals are aligned with the ones that you yourself care about. https://vegasharmon.com/commonplace/to-love-dung-one-must-believe-in-dung-dot-but/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/to-love-dung-one-must-believe-in-dung-dot-but/ To love dung, one must believe in dung. But as his cognition bounded upward, plunging into physics and causal topology, the macroscopic category of “waste” began to leak. It dissolved into a transient, low-resolution molecular configuration of nitrogen, carbon, and water. Under the lens of his expanding causal models, the very boundaries of objects broke down. You cannot rigorously optimize the universe for a leaky biological abstraction once you can perceive the quantum fields beneath it. https://vegasharmon.com/commonplace/wanting-to-pull-the-lever-then-finding-reasons-why/ Mon, 12 Oct 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/wanting-to-pull-the-lever-then-finding-reasons-why/ What went wrong here? It’s hard to know exactly what led any given person to endorse any given argument. But when we zoom out, it becomes clear that many people in this space really want to pull some lever that feels important, and privilege arguments which justify that. That might come from a sense that they need to have an impact on the world; or fear about failing to fulfil their potential; or the more mundane explanation that big levers tend to be associated with money and prestige and proximity to power. https://vegasharmon.com/commonplace/ai-cannot-show-us-something-new/ Thu, 14 May 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/ai-cannot-show-us-something-new/ Is the new AI a useful tool? Yes, a tool with many uses. It can help us code faster and be our concierge, our stenographer, and so on. It can turn the value-neutral dials technology has manned since the Industrial Revolution: efficiency, automation. But to answer Paul Ford’s old question: What can this thing do, what is it promising, that isn’t just an optimized version of already extant processes? Its promise is to finally personalize the Algorithm — to fix all its alienation factors, to make it the thing that gives us what we really want. https://vegasharmon.com/commonplace/tech-magnates-need-no-arguments/ Sun, 19 Apr 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/tech-magnates-need-no-arguments/ Yet a certain intellectual vacuity has hardly proved to be an obstacle to the ambitions of contemporary tech giants. To the contrary. So it would be naive to suppose that what is wanted are better arguments or sturdier postulates. The far-reaching developments taking place right now—particularly in the realm of artificial intelligence—are not the result of some philosophical breakthrough. The authority of the tech magnates does not derive from victory in some forensic contest but from the fact that they need no arguments to begin with: They are simply moving forward regardless. https://vegasharmon.com/commonplace/ask-the-ai-for-consequences-not-decisions/ Mon, 09 Feb 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/ask-the-ai-for-consequences-not-decisions/ Before the settlement could be finalized, however, the CEO of Robot Central asked his AI system to explore all possible actions he could take and tell him their consequences. He adhered to the 1990s principle: Never ask an AI system what to do. Ask it to tell you the consequences of the different things you might do. https://vegasharmon.com/commonplace/the-computer-longs-to-be-human/ Tue, 27 Jan 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/the-computer-longs-to-be-human/ Homay King argues in Virtual Memory: Time-Based Art and the Dream of Digitality that the epistle’s “tangled mash-up of sentimentality bespeaks a twinge of longing…Like the wooden puppet in search of the Blue Fairy, the computer longs to be human; like Snow White fleeing into the forest, it longs to be admitted into the company of those who are capable of care and affection.” https://vegasharmon.com/commonplace/the-robotic-hand-s-undetectable-autograph/ Tue, 27 Jan 2026 00:00:00 +0000 https://vegasharmon.com/commonplace/the-robotic-hand-s-undetectable-autograph/ Beleaguered professors, who are largely poorly paid adjunct instructors at this point, will now have to not just contend with essay mills and good old-fashioned cut-and-paste plagiarism, but also the undetectable autograph of the robotic hand. https://vegasharmon.com/commonplace/capitalism-uncoupled-from-human-values/ Sat, 15 Nov 2025 00:00:00 +0000 https://vegasharmon.com/commonplace/capitalism-uncoupled-from-human-values/ The idea of biological or cultural evolution causing a mass population explosion is a philosophical toy at best. The idea of technology making it possible is both plausible and terrifying. Now we see that “physical limits” segues very naturally into “excess resources” – the ability to create new agents very quickly means that unless everyone can coordinate to ban doing this, the people who do will outcompete the people who don’t until they have reached carrying capacity and everyone is stuck at subsistence level. https://vegasharmon.com/commonplace/grieving-the-lost-future/ Sat, 15 Nov 2025 00:00:00 +0000 https://vegasharmon.com/commonplace/grieving-the-lost-future/ So many people have lived such grand lives. I have certainly lived a greater life than I expected, filled with adventures and curious people. But people will soon not live any lives at all. I believe that we will soon build intelligences more powerful than us who will disempower and kill us all. I will see no children of mine grow to adulthood. No people will walk through mountains and trees. https://vegasharmon.com/commonplace/i-will-not-call-up-that-which-i-cannot-put-down/ Sat, 15 Nov 2025 00:00:00 +0000 https://vegasharmon.com/commonplace/i-will-not-call-up-that-which-i-cannot-put-down/ But coordination only works when you have 51% or more of the force on the side of the people doing the coordinating, and when you haven’t come up with some brilliant trick to make coordination impossible. The second one first. In the links post before last, I wrote: The latest development in the brave new post-​Bitcoin world is crypto-​equity. At this point I’ve gone from wanting to praise these inventors as bold libertarian heroes to wanting to drag them in front of a blackboard and making them write a hundred times “I WILL NOT CALL UP THAT WHICH I CANNOT PUT DOWN” https://vegasharmon.com/commonplace/multipolar-traps-as-technology-approaches-infinity/ Sat, 15 Nov 2025 00:00:00 +0000 https://vegasharmon.com/commonplace/multipolar-traps-as-technology-approaches-infinity/ The limit of multipolar traps as technology approaches infinity is “very bad”. https://vegasharmon.com/commonplace/nick-land-and-summoning-demon-gods/ Sat, 15 Nov 2025 00:00:00 +0000 https://vegasharmon.com/commonplace/nick-land-and-summoning-demon-gods/ As far as I can tell from reading his blog, Nick Land is the guy in that terrifying border region where he is smart enough to figure out several important arcane principles about summoning demon gods, but not quite smart enough to figure out the most important such principle, which is NEVER DO THAT. https://vegasharmon.com/commonplace/two-places-the-river-reaches-the-sea/ Sat, 15 Nov 2025 00:00:00 +0000 https://vegasharmon.com/commonplace/two-places-the-river-reaches-the-sea/ Absent an extraordinary effort to divert it, the river reaches the sea in one of two places. It can end in Eliezer Yudkowsky’s nightmare of a superintelligence optimizing for some random thing (classically paper clips) because we weren’t smart enough to channel its optimization efforts the right way. This is the ultimate trap, the trap that catches the universe. Everything except the one thing being maximized is destroyed utterly in pursuit of the single goal, including all the silly human values. https://vegasharmon.com/commonplace/shadows-of-the-coming-race/ Tue, 08 Jul 2025 00:00:00 +0000 https://vegasharmon.com/commonplace/shadows-of-the-coming-race/ My friend Trost, who is no optimist as to the state of the universe hitherto, but is confident that at some future period within the duration of the solar system, ours will be the best of all possible worlds—a hope which I always honour as a sign of beneficent qualities—my friend Trost always tries to keep up my spirits under the sight of the extremely unpleasant and disfiguring work by which many of our fellow-creatures have to get their bread, with the assurance that “all this will soon be done by machinery.