The Nonlinear Library: LessWrong

Link to original articleWelcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: How I started believing religion might actually matter for rationality and moral philosophy, published by zhukeepa on August 23, 2024 on LessWrong. After the release of Ben Pace's extended interview with me about my views on religion, I felt inspired to publish more of my thinking about religion in a format that's more detailed, compact, and organized. This post is the first publication in my series of intended posts about religion. Thanks to Ben Pace, Chris Lakin, Richard Ngo, Renshin Lauren Lee, Mark Miller, and Imam Ammar Amonette for their feedback on this post, and thanks to Kaj Sotala, Tomáš Gavenčiak, Paul Colognese, and David Spivak for reviewing earlier versions of this post. Thanks especially to Renshin Lauren Lee and Imam Ammar Amonette for their input on my claims about religion and inner work, and Mark Miller for vetting my claims about predictive processing. In Waking Up, Sam Harris wrote:[1] But I now knew that Jesus, the Buddha, Lao Tzu, and the other saints and sages of history had not all been epileptics, schizophrenics, or frauds. I still considered the world's religions to be mere intellectual ruins, maintained at enormous economic and social cost, but I now understood that important psychological truths could be found in the rubble. Like Sam, I've also come to believe that there are psychological truths that show up across religious traditions. I furthermore think these psychological truths are actually very related to both rationality and moral philosophy. This post will describe how I personally came to start entertaining this belief seriously. "Trapped Priors As A Basic Problem Of Rationality" "Trapped Priors As A Basic Problem of Rationality" was the title of an AstralCodexTen blog post. Scott opens the post with the following: Last month I talked about van der Bergh et al's work on the precision of sensory evidence, which introduced the idea of a trapped prior. I think this concept has far-reaching implications for the rationalist project as a whole. I want to re-derive it, explain it more intuitively, then talk about why it might be relevant for things like intellectual, political and religious biases. The post describes Scott's take on a predictive processing account of a certain kind of cognitive flinch that prevents certain types of sensory input from being perceived accurately, leading to beliefs that are resistant to updating.[2] Some illustrative central examples of trapped priors: Karl Friston has written about how a traumatized veteran might not hear a loud car as a car, but as a gunshot instead. Scott mentions phobias and sticky political beliefs as central examples of trapped priors. I think trapped priors are very related to the concept that "trauma" tries to point at, but I think "trauma" tends to connote a subset of trapped priors that are the result of some much more intense kind of injury. "Wounding" is a more inclusive term than trauma, but tends to refer to trapped priors learned within an organism's lifetime, whereas trapped priors in general also include genetically pre-specified priors, like a fear of snakes or a fear of starvation. My forays into religion and spirituality actually began via the investigation of my own trapped priors, which I had previously articulated to myself as "psychological blocks", and explored in contexts that were adjacent to therapy (for example, getting my psychology dissected at Leverage Research, and experimenting with Circling). It was only after I went deep in my investigation of my trapped priors that I learned of the existence of traditions emphasizing the systematic and thorough exploration of trapped priors. These tended to be spiritual traditions, which is where my interest in spirituality actually began.[3] I will elaborate more on this later. Active blind spots as second-order trapp...

Link to original articleWelcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Turning 22 in the Pre-Apocalypse, published by testingthewaters on August 23, 2024 on LessWrong. Meta comment for LessWrong readers[1] Something Different This Way Comes - Part 1 In which I attempt to renegotiate rationalism as a personal philosophy, and offer my alternative - Game theory is not a substitute for real life - Heuristics over theories Introduction This essay focuses on outlining an alternative to the ideology of rationalism. As part of this, I offer my definition of the rationalist project, my account of its problems, and my concept of a counter-paradigm for living one's life. The second part of this essay will examine the political implications of rationalism and try to offer an alternative on a larger scale. Defining Rationalism To analyse rationalism, I must first define what I am analysing. Rationalism (as observed in vivo on forums like LessWrong) is a loose constellation of ideas radiating out of various intellectual traditions, amongst them Bayesian statistics, psychological decision theories, and game theory. These are then combined with concepts in sub-fields of computer science (AI and simulation modelling), economics (rational actor theory or homo economicus), politics (libertarianism), psychology (evolutionary psychology) and ethics (the utilitarianism of Peter Singer). The broad project of rationalism aims to generalise the insights of these traditions into application at both the "wake up and make a sandwich" and the "save the world" level. Like any good tradition, it has a bunch of contradictions embedded: Some of these include intuitionism (e.g. when superforecasters talk about going with their gut) vs deterministic analysis (e.g. concepts of perfect game-players and k-level rationality). Another one is between Bayesianism (which is about updating priors about the world based on evidence received, generally without making any causal assumptions) vs systemisation (which is about creating causal models/higher level representations of real life situations to understand them better). In discussing this general state of rhetorical confusion I am preceded by Philip Agre's Towards a Critical Technical Practice, which is AI specific but still quite instructive. The broader rationalist community (especially online) includes all sorts of subcultures but generally there are in group norms that promote certain technical argot ("priors", "updating"), certain attitudes towards classes of entities ("blank faces"/bureaucrats/NPCs/the woke mob etc), and certain general ideas about how to solve "wicked problems" like governance or education. There is some overlap with online conservatives, libertarians, and the far-right. There is a similar overlap with general liberal technocratic belief systems, generally through a belief in meritocracy and policy solutions founded on scientific or technological principles. At the root of this complex constellation there seems to be a bucket of common values which are vaguely expressed as follows: 1. The world can be understood and modelled by high level systems that are constructed based on rational, clearly defined principles and refined by evidence/observation. 2. Understanding and use of these systems enables us to solve high level problems (social coordination, communication, AI alignment) as well as achieving our personal goals. 3. Those who are more able to comprehend and use these models are therefore of a higher agency/utility and higher moral priority than those who cannot. There is also a fourth law which can be constructed from the second and third: By thinking about this at all, by starting to consciously play the game of thought-optimisation and higher order world-modelling, you (the future rationalist) have elevated yourself above the "0-level" player who does not think about such problems and naively pur...

Link to original articleWelcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: If we solve alignment, do we die anyway?, published by Seth Herd on August 23, 2024 on LessWrong. I'm aware of good arguments that this scenario isn't inevitable, but it still seems frighteningly likely even if we solve technical alignment. TL;DR: 1. If we solve alignment, it will probably be used to create AGI that follows human orders. 2. If takeoff is slow-ish, a pivotal act (preventing more AGIs from being developed) will be difficult. 3. If no pivotal act is performed, RSI-capable AGI proliferates. This creates an n-way non-iterated Prisoner's Dilemma where the first to attack, wins. 4. Disaster results. The first AGIs will probably be aligned to take orders People in charge of AGI projects like power. And by definition, they like their values somewhat better than the aggregate values of all of humanity. It also seems like there's a pretty strong argument that Instruction-following AGI is easier than value aligned AGI. In the slow-ish takeoff we expect, this alignment target seems to allow for error-correcting alignment, in somewhat non-obvious ways. If this argument holds up even weakly, it will be an excuse for the people in charge to do what they want to anyway. I hope I'm wrong and value-aligned AGI is just as easy and likely. But it seems like wishful thinking at this point. The first AGI probably won't perform a pivotal act In realistically slow takeoff scenarios, the AGI won't be able to do anything like make nanobots to melt down GPUs. It would have to use more conventional methods, like software intrusion to sabotage existing projects, followed by elaborate monitoring to prevent new ones. Such a weak attempted pivotal act could fail, or could escalate to a nuclear conflict. Second, the humans in charge of AGI may not have the chutzpah to even try such a thing. Taking over the world is not for the faint of heart. They might get it after their increasingly-intelligent AGI carefully explains to them the consequences of allowing AGI proliferation, or they might not. If the people in charge are a government, the odds of such an action go up, but so do the risks of escalation to nuclear war. Governments seem to be fairly risk-taking. Expecting governments to not just grab world-changing power while they can seems naive, so this is my median scenario. So RSI-capable AGI may proliferate until a disaster occurs If we solve alignment and create personal intent aligned AGI but nobody manages a pivotal act, I see a likely future world with an increasing number of AGIs capable of recursively self-improving. How long until someone tells their AGI to hide, self-improve, and take over? Many people seem optimistic about this scenario. Perhaps network security can be improved with AGIs on the job. But AGIs can do an end-run around the entire system: hide, set up self-replicating manufacturing (robotics is rapidly improving to allow this), use that to recursively self-improve your intelligence, and develop new offensive strategies and capabilities until you've got one that will work within an acceptable level of viciousness.[1] If hiding in factories isn't good enough, do your RSI manufacturing underground. If that's not good enough, do it as far from Earth as necessary. Take over with as little violence as you can manage or as much as you need. Reboot a new civilization if that's all you can manage while still acting before someone else does. The first one to pull the stops probably wins. This looks all too much like a non-iterated Prisoner's Dilemma with N players - and N increasing. Counterarguments/Outs For small numbers of AGI and similar values among their wielders, a collective pivotal act could be performed. I place some hopes here, particularly if political pressure is applied in advance to aim for this outcome, or if the AGIs come up with better cooperation stru...

Link to original articleWelcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: AI #78: Some Welcome Calm, published by Zvi on August 23, 2024 on LessWrong. SB 1047 has been amended once more, with both strict improvements and big compromises. I cover the changes, and answer objections to the bill, in my extensive Guide to SB 1047. I follow that up here with reactions to the changes and some thoughts on where the debate goes from here. Ultimately, it is going to come down to one person: California Governor Gavin Newsom. All of the debates we're having matter to the extent they influence this one person. If he wants the bill to become law, it almost certainly will become law. If he does not want that, then it won't become law, they never override a veto and if he makes that intention known then it likely wouldn't even get to his desk. For now, he's not telling. Table of Contents 1. Introduction. 2. Table of Contents. 3. Language Models Offer Mundane Utility. AI sort of runs for mayor. 4. Language Models Don't Offer Mundane Utility. A go or no go decision. 5. Deepfaketown and Botpocalypse Soon. How hard is finding the desert of the real? 6. The Art of the Jailbreak. There is always a jailbreak. Should you prove it? 7. Get Involved. Also when not to get involved. 8. Introducing. New benchmark, longer PDFs, the hot new RealFakeGame. 9. In Other AI News. METR shares its conclusions on GPT-4o. 10. Quiet Speculations. Are we stuck at 4-level models due to Nvidia? 11. SB 1047: Nancy Pelosi. Local Nvidia investor expresses opinion. 12. SB 1047: Anthropic. You got most of what you wanted. Your move. 13. SB 1047: Reactions to the Changes. Reasonable people acted reasonably. 14. SB 1047: Big Picture. Things tend to ultimately be rather simple. 15. The Week in Audio. Joe Rogan talks to Peter Thiel. 16. Rhetorical Innovation. Matthew Yglesias offers improved taxonomy. 17. Aligning a Smarter Than Human Intelligence is Difficult. Proving things is hard. 18. The Lighter Side. The future, while coming, could be delayed a bit. Language Models Offer Mundane Utility Sully thinks the big models (Opus, 405B, GPT-4-0314) have that special something the medium-sized models don't have, no matter what the evals say. A source for Llama-3.1-405-base, at $2 per million tokens (both input and output). Accelerate development of fusion energy, perhaps? Steven Cowley makes the case that this may be AI's 'killer app.' This would be great, but if AI can accelerate fusion by decades as Cowley claims, then what else can it also do? So few people generalize. Show the troll that AIs can understand what they're misinterpreting. I am not as optimistic about this strategy as Paul Graham, and look forward to his experiments. Mayoral candidate in Cheyenne, Wyoming promises to let ChatGPT be mayor. You can tell that everyone involved it thinking well and taking it seriously, asking the hard questions: "Is the computer system in city hall sufficient to handle AI?" one attendee, holding a wireless microphone at his seat, asked VIC. "If elected, would you take a pay cut?" another wanted to know. "How would you make your decisions according to human factor, involving humans, and having to make a decision that affects so many people?" a third chimed in. After each question, a pause followed. "Making decisions that affect many people requires a careful balance of data-driven insights and human empathy," VIC said in a male-sounding voice. "Here's how I would approach it," it added, before ticking off a six-part plan that included using AI to gather data on public opinion and responding to constituents at town halls. OpenAI shut off his account, saying this was campaigning and thus against terms of service, but he quickly made another one. You can't actually stop anyone from using ChatGPT. And I think there Aint No Rule against using it for actual governing. I still don't know how this 'AI Mayor' w...

Link to original articleWelcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: A primer on the current state of longevity research, published by Abhishaike Mahajan on August 22, 2024 on LessWrong. Note: This post is co-authored with Stacy Li, a PhD student at Berkeley studying aging biology! Highly appreciate all her help in writing, editing, and fact-checking my understanding! Introduction The last time I read about aging research deeply was around 2021. The general impression I was getting was that aging research was increasingly more and more funded (good!). Unfortunately, none of the money led to actionable or useful insights (bad). Over time, you get slightly burnt out by all the negative news. After getting a job in biotech, I kept a hazy eye on the subject but mostly tuned out of it entirely. But, especially today, I am curious: how has the aging field progressed in the last few years? Since 2021, what has changed? In this post, I'll share a list of immediate questions about the state of affairs in aging research, and the answers I've found for them. For each question, I'll offer some basic background knowledge required to understand the question. Feel free to skip that part if you already understand the question! Did the therapeutic focus on sirtuins amount to much? Background Sirtuins are a family of signaling proteins, commonly referred to by their corresponding gene name, SIRT1, SIRT2, all the way up to SIRT7. Their primary role is deacetylation, which is just the removal of a chemical marker (acetyl) on proteins. It was noticed in the 1980s that some sirtuin classes were especially involved in three key activities: modifying histones, which are proteins that tune the accessibility of DNA in the nucleus, transcriptional modification, which determines how DNA is interpreted by the body, and DNA repair, which speaks for itself. And anything involved in modifying and maintaining DNA is something worth paying attention to! Studies in the 2000s showed that the activity of specific sirtuin classes strongly correlated with age; the young had more sirtuin activity, and the old had less. This seemed to be causative in aging; overexpressing certain sirtuin genes led to lifespan increase and downregulation of them led to lifespan decrease. The results were a bit mixed, and the results were for yeast cells - always a red flag - but there was some promise in viewing sirtuins as an aging target. It turns out that editing humans to safely overexpress sirtuin genes is somewhat hard to do (as is expressing any gene in humans). But there was an easy way around that: focus on molecules that are required for sirtuin to do its job. A class of therapeutics grew from this: sirtuin-activating compounds. How do you activate sirtuins? Well, sirtuins are dependent on NAD+, or nicotinamide adenine dinucleotide, to perform their primary function. Increasing cellular NAD+ levels could also be a way to indirectly push for more sirtuin activity. Practically speaking, NAD+ bioavailability is poor, so supplementation with precursors to NAD+, such as nicotinamide mononucleotide (NMN) and nicotinamide riboside (NR), was instead used. There are plenty of other compounds in this category too: resveratrol, fisetin, and quercetin are all names you may hear mentioned. How has this fared? Answer TLDR: The whole sirtuin theory was losing steam by the time I started reading about it a few years ago. It's only gotten worse. Nothing particularly useful has come from sirtuin-focused therapies, and likely nothing ever will. A Cell paper from 2018 found that NAD+ precursor supplementation didn't improve mice longevity. To be fair, they did show that supplementation improves some aspects of health-span, specifically improved glucose metabolism and reduced oxidative stress to the liver in aged mice, so still potentially useful. But nothing revolutionary. Still, human clinical trials ...

Link to original articleWelcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: The economics of space tethers, published by harsimony on August 22, 2024 on LessWrong. Some code for this post can be found here. Space tethers take the old, defunct space elevator concept and shorten it. Rockets can fly up to a dangling hook in the sky and then climb to a higher orbit. If the tether rotates, it can act like a catapult, providing a significant boost in a location where providing thrust is expensive. Kurzgesagt has a nice explainer and ToughSF has a great piece explaining the mechanics and some applications. Tethers make it cheaper to explore space, but how much cheaper? Let's look at the benefits. Tether materials and characteristic velocity The key performance metric for the tether material is the characteristic velocity: Vc=2Tρ Where T is the tensile strength and rho is the density. The stronger and lighter the material is, the faster the tether can spin, boosting payloads to higher speeds and saving more fuel. This quickly leads to thinking about exotic materials. Hexagonal boron nitride! Carbon nanotubes! I'm not immune to this kind of speculation, so I've added an appendix on the topic. But as I argue in another part of the appendix, we already have good enough materials to make a space tether. The potential gain from studying exotic materials is actually pretty small. For what it's worth, I like glass fibers. They're pretty easy to make, the material can be be sourced in space, they can handle large temperature ranges, and they're resistant to atomic oxygen environments and UV [1]. They can also get pretty good performance, S-2 glass fibers have a characteristic velocity close to 2 km/s while the best currently available material clocks in at 2.7 km/s. Now let's look at why the speed of the tether matters. Delta V and fuel savings Rockets have to reach a certain speed in order to orbit any object. For low earth orbit, that's roughly 7.9 km/s; that's over Mach 20 here on Earth. The change in velocity, or delta V (dV), required to reach orbit is the currency of spaceflight. You can essentially map out the solar system based on the delta V needed to reach different places: Source It takes a lot of fuel and engineering to get a payload up to these speeds, making launches expensive [2][3]. Tethers are exciting because they can wait in orbit and offer a rocket some extra delta V. A tether spinning at 1.5 km/s in LEO can grab a rocket moving at 5.8 km/s and release it at 8.8 km/s: Source It takes a while to visualize how these work. Staring at this gif helps: Source Even a small delta V boost saves a lot of fuel. That's because the total fuel required for a mission increases exponentially with delta V requirements, as we can see from the Tsiolkovsky rocket equation: ΔV=Ispg0ln(mimp) I_sp is the specific impulse of the rocket, g_0 is the gravitational acceleration (often just called *g *in Earth's gravity), m_i is the total initial mass of the rocket including fuel, and m_p is the payload mass of the rocket after the fuel has been expended. Note that m_p includes both the literal payload and the mass of the rocket itself. Rearranging to see the exponential: mi=mpexp(ΔVIspg0) m_i is the sum of the payload mass m_p and the fuel mass m_x. We can rewrite the above in terms of fuel mass: mx=mp(exp(ΔVIspg0)1) By offering a free delta V boost, tethers can save literal tons of fuel. If the tether is spinning at a certain velocity V_t, the tether provides a boost twice that size. You can subtract that boost from the dV requirements for the rocket: ΔV'=ΔV2Vt The new initial mass is: m'i=mpexp(ΔV2VtIspg0) The new fuel requirement is: m'x=m'imp=mp(exp(ΔV2VtIspg0)1) As an example, let's imagine a tether orbiting in LEO [4] at an orbital velocity of 7.5 km/s and spinning at 2 km/s. Our rocket only needs to reach 5.5 km/s in order to be boosted to 9.5 km/s. A Starsh...

Link to original articleWelcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Measuring Structure Development in Algorithmic Transformers, published by Micurie on August 22, 2024 on LessWrong. tl;dr: We compute the evolution of the local learning coefficient (LLC), a proxy for model complexity, for an algorithmic transformer. The LLC decreases as the model learns more structured solutions, such as head specialization. This post is structured in three main parts, (1) a summary, giving an overview of the main results, (2) the Fine Print, that delves into various cross-checks and details and (3) Discussion and Conclusions. Structure Formation in Algorithmic Transformers In this work we study the development of simple algorithmic transformers, which are transformers that learn to perform algorithmic tasks. A major advantage of this setup is that we can control several (hyper)parameters, such as the complexity of the training data and network architecture. This allows us to do targeted experiments studying the impacts of these parameters on the learning dynamics. The main tool we use to study the development is the Local Learning Coefficient (LLC) and we choose cases where we have a reverse-engineered solution. Why use the LLC for this purpose? It is a theoretically well motivated measure of model complexity defined by Lau et.al. For an overview of Singular Learning Theory (which serves as the theoretical foundation for the LLC) see Liam Carol's Distilling SLT sequence. For a brief overview of the LLC see e.g. this post. We use the same setup as CallumMcDougall's October Monthly Algorithmic Mech-Interp Challenge. The model is an attention only transformer, trained on sorting numbers with layer norm and weight decay on a cross-entropy loss function using the Adam optimizer. The residual stream size is 96 and the head dimension is 48. It is trained on sequences of the form and to predict the next token starting at the separation token. The numbers in the list are sampled uniformly from 0 to 50, which together with the separation token produce a vocabulary of 52 tokens. Numbers do not repeat in the list. 1-Head Model Let's first look at the case of a 1-head transformer: The model reaches 100% accuracy around training step 100, confirming that a single attention head is sufficient for sorting, as noted in previous work. Once maximum accuracy is reached, the full QK and OV circuits[2] behave as described by Callum for the 2-head model: In the QK circuit, source tokens attend more to the smallest token in the list larger than themselves. This results in a higher value band above the diagonal and a lower value band below the diagonal. The OV circuit copies tokens, as seen by the clear positive diagonal pattern. In addition, we observe a transition around training step 1000, where the LLC decreases while the accuracy stays unchanged. This is supported by a drop in the sum of the ranks[3] of the matrices in the heat maps. It also coincides with the formation of the off-diagonal stripes in the OV-circuit. We speculate that these are simpler than the noisier off-diagonal OV pattern observed at peak LLC, and correspond to the translational symmetry of the problem. We define a Translational Symmetry measure[1] (see purple line in the plot) to capture the degree to which the circuits obey this symmetry. It increases throughout most of the training, even after the other measures stabilize. 2-Head Model Let's now turn our attention to the 2-head transformer in Callum's original setup. We see a lot of qualitative similarities to the evolution of the full QK and OV circuits for the 1-head model. As the LLC begins to drop (around training step 1000), we note the following: QK circuit: Slight changes[5] to the attention pattern, which crystallize into triangular regions late in the training, long after the LLC has stabilized. OV circuit: The heads specialize, splittin...

Link to original articleWelcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Please do not use AI to write for you, published by Richard Kennaway on August 22, 2024 on LessWrong. I've recently seen several articles here that were clearly generated or heavily assisted by AI. They are all dreadful. They are verbose, they are full of "on the one hand" and "on the other", they never make any assertion without also making room for the opposite, and end with "conclusions" that say nothing. Please do not do this. I wrote the above, unassisted, and it states my own views. To illustrate the point I then had GPT-4o rewrite it. Here was the prompt: I have recently seen a substantial increase in the proportion of postings to a certain online forum that were clearly generated by AI, although not marked as such. The writing is verbose and meandering, full of "on the one hand" and "on the other", never making an assertion without also suggesting its opposite, sounding oh-so-measured and even-handed that it ends up saying nothing at all. Please write an article for me, written in the first person, arguing that AI tools in the current state of the art should never be used for this purpose. Below the line is the chatbot's extruded meat product. This is the whole of its first output, unedited, and I did not press it to continue. The first paragraph starts out ok, being merely verbose and repetitive but setting out the basic claim. And then the second paragraph backpedals it past where it began. The third paragraph floats off into the cloud layer, and so on for a total of nine paragraphs and 875 words that would be a waste of anyone's time to read, unless they want to see if their judgement agrees with mine. It did not notice that I was using it in exactly the way I had asked it to argue against, and that the article displays all the faults that it purports to diagnose. The splash image is a photograph from Wikipedia, CC BY-SA 3.0. I'll spare you the image that GPT-4o came up with, but y'know, I do find the AI illustrations that everyone heads their blog posts with these days pretty annoying. (ETA: Well, there was supposed to be an image, I filled out the "link preview" part of the entry form, but no image. But the link above will show you what it was.) Why AI-Generated Content Shouldn't Flood Online Forums Lately, I've noticed something troubling in the online forums I frequent: a distinct uptick in posts that seem like they've been generated by AI, though they're not labeled as such. The language is eerily similar across many of these posts - verbose, wandering, and conspicuously non-committal. Every argument is hedged with a counterpoint, every statement cushioned by a qualifier, making the writing sound balanced on the surface but ultimately empty. The posts stretch on endlessly, piling one vague point onto another, yet they fail to really say anything. It's frustrating, to say the least, and I believe this trend reveals a serious problem with the way AI is being misused in online spaces. I want to be clear: I'm not arguing that AI lacks merit as a tool. It has immense potential in fields like data analysis, automation, and even creativity when it complements human intelligence. However, when it comes to contributing to online discourse - spaces where real people go to express ideas, ask questions, and engage in genuine conversation - I believe AI-generated content, at least in its current form, does more harm than good. First, let's talk about the nature of conversation itself. Forums, at their best, are places where people share perspectives, debate ideas, and solve problems together. What makes these interactions meaningful is that they are infused with human experience, emotion, and authenticity. Even when disagreements arise, they are rooted in real personal stakes - whether it's a passion for a hobby, a struggle to understand a concept, or a desire to ...

Link to original articleWelcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: the Giga Press was a mistake, published by bhauth on August 21, 2024 on LessWrong. the giga press Tesla decided to use large aluminum castings ("gigacastings") for the frame of many of its vehicles, including the Model Y and Cybertruck. This approach and the "Giga Press" used for it have been praised by many articles and youtube videos, repeatedly called revolutionary and a key advantage. Most cars today are made by stamping steel sheets and spot welding them together with robotic arms. Here's video of a Honda factory. But that's outdated: gigacasting is the future! BYD is still welding stamped steel sheets together, and that's why it can't compete on price with Tesla. Hold on, it seems...BYD prices are actually lower than Tesla's? Much lower? Oh, and Tesla is no longer planning single unitary castings for future vehicles? I remember reading analysis from a couple people with car manufacturing experience, concluding that unitary cast aluminum bodies could have a cost advantage for certain production numbers, like 200k cars, but dies for casting wear out sooner than dies for stamping steel, and as soon as you need to replace them the cost advantage is gone. Also, robotic arms are flexible and stamped panels can be used for multiple car models, and if you already have robots and panels you can use from discontinued car models, the cost advantage is gone. But Tesla was expanding so they didn't have available robots already. So using aluminum casting would probably be slightly more expensive, but not make a big difference. "That seems reasonable", I said to myself, "ふむふむ". And I previously pointed that out, eg here. But things are actually worse than that. aluminum die casting Die casting of aluminum involves injecting liquid aluminum into a die and letting it cool. Liquid aluminum is less dense than solid aluminum, and aluminum being cast doesn't all solidify at the same time. Bigger castings have aluminum flowing over larger distances. The larger the casting, the less evenly the aluminum cools: there's more space for temperature differences in the die, and the aluminum cools as it's injected. As a result, bigger castings have more problems with warping and voids. Also, a bigger casting with the same curvature from warping has bigger position changes. Tesla has been widely criticized for stuff not fitting together properly on the car body. My understanding is that the biggest reason for that is their large aluminum castings being slightly warped. As for voids, they can create weak points; I think they were the reason the cybertruck hitch broke off in this test. Defects from casting are the only explanation for that cast aluminum breaking apart that way. If you want to inject more aluminum as solidification and shrinkage happens, the distance it has to travel is proportional to casting size - unless you use multi-point injection, which Tesla doesn't, and that has its own challenges. Somehow I thought Tesla would have only moved to its "Giga Press" after adequately dealing with those issues, but that was silly of me. One approach being worked on to mitigate warping of large aluminum castings is "rheocasting", where a slurry of solid aluminum in liquid aluminum is injected, reducing the shrinkage from cooling. But that's obviously more viscous and thus requires higher injection pressures which requires high die pressures. aluminum vs steel Back when aluminum established its reputation as "the lightweight higher-performance alternative" to steel, 300 MPa was considered a typical (tensile yield) strength for steel. Typical cast aluminum can almost match that, and high-performance aluminum for aircraft can be >700 MPa. Obviously there are reasons it's not always used: high-strength aluminum requires some more-expensive elements and careful heat-treatment. Any hot welds will r...

Link to original articleWelcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: AGI Safety and Alignment at Google DeepMind: A Summary of Recent Work, published by Rohin Shah on August 20, 2024 on LessWrong. We wanted to share a recap of our recent outputs with the AF community. Below, we fill in some details about what we have been working on, what motivated us to do it, and how we thought about its importance. We hope that this will help people build off things we have done and see how their work fits with ours. Who are we? We're the main team at Google DeepMind working on technical approaches to existential risk from AI systems. Since our last post, we've evolved into the AGI Safety & Alignment team, which we think of as AGI Alignment (with subteams like mechanistic interpretability, scalable oversight, etc.), and Frontier Safety (working on the Frontier Safety Framework, including developing and running dangerous capability evaluations). We've also been growing since our last post: by 39% last year, and by 37% so far this year. The leadership team is Anca Dragan, Rohin Shah, Allan Dafoe, and Dave Orr, with Shane Legg as executive sponsor. We're part of the overall AI Safety and Alignment org led by Anca, which also includes Gemini Safety (focusing on safety training for the current Gemini models), and Voices of All in Alignment, which focuses on alignment techniques for value and viewpoint pluralism. What have we been up to? It's been a while since our last update, so below we list out some key work published in 2023 and the first part of 2024, grouped by topic / sub-team. Our big bets for the past 1.5 years have been 1) amplified oversight, to enable the right learning signal for aligning models so that they don't pose catastrophic risks, 2) frontier safety, to analyze whether models are capable of posing catastrophic risks in the first place, and 3) (mechanistic) interpretability, as a potential enabler for both frontier safety and alignment goals. Beyond these bets, we experimented with promising areas and ideas that help us identify new bets we should make. Frontier Safety The mission of the Frontier Safety team is to ensure safety from extreme harms by anticipating, evaluating, and helping Google prepare for powerful capabilities in frontier models. While the focus so far has been primarily around misuse threat models, we are also working on misalignment threat models. FSF We recently published our Frontier Safety Framework, which, in broad strokes, follows the approach of responsible capability scaling, similar to Anthropic's Responsible Scaling Policy and OpenAI's Preparedness Framework. The key difference is that the FSF applies to Google: there are many different frontier LLM deployments across Google, rather than just a single chatbot and API (this in turn affects stakeholder engagement, policy implementation, mitigation plans, etc). We're excited that our small team led the Google-wide strategy in this space, and demonstrated that responsible capability scaling can work for large tech companies in addition to small startups. A key area of the FSF we're focusing on as we pilot the Framework, is how to map between the critical capability levels (CCLs) and the mitigations we would take. This is high on our list of priorities as we iterate on future versions. Some commentary (e.g. here) also highlighted (accurately) that the FSF doesn't include commitments. This is because the science is in early stages and best practices will need to evolve. But ultimately, what we care about is whether the work is actually done. In practice, we did run and report dangerous capability evaluations for Gemini 1.5 that we think are sufficient to rule out extreme risk with high confidence. Dangerous Capability Evaluations Our paper on Evaluating Frontier Models for Dangerous Capabilities is the broadest suite of dangerous capability evaluations published...