~www_lesswrong_com | Bookmarks (687)

Published on October 29, 2024 12:41 PM GMTThis project report was created in September 2024 as part of the BlueDot AI Safety Fundamentals Course, with the guidance of my facilitator, Alexandra Abbas. Work on this project originated as part of an ideation at a Apart Research hackathon. This report dives into APIAYN (A Poem Is All You Need), a simple jailbreak that doesn’t employ direct deception...

Published on October 29, 2024 12:16 PM GMTTo update our credence on whether or not LLMs are conscious, we can ask how many of the Butlin/Long indicator properties for phenomenal consciousness are satisfied by LLMs. To start this program, I zoomed in on an indicator property that is required for consciousness under higher-order theory, nicknamed “HOT-2”: Metacognitive monitoring distinguishing reliable perceptual representations from noise. Do today’s...

Published on October 29, 2024 7:10 AM GMTThe big news of the week was the release of a new version of Claude Sonnet 3.5, complete with its ability (for now only through the API) to outright use your computer, if you let it. It’s too early to tell how big an upgrade this is otherwise. ChatGPT got some interface tweaks that, while minor, are...

Published on October 29, 2024 3:02 AM GMT"Each one of us, and also us as the current implementation of humanity are going to be replaced. Persistence in current form is impossible. It's impossible in biology; every species will either die out or it will change and adapt, in which case it is again not the same species. So the next question is once you've...

Published on October 29, 2024 1:21 AM GMTThis is a follow-up to last week's D&D.Sci scenario: if you intend to play that, and haven't done so yet, you should do so now before spoiling yourself.There is a web interactive here you can use to test your answer, and generation code available here if you're interested, or you can read on for the ruleset and...

Published on October 29, 2024 1:07 AM GMTMatt Levine is the most well-known newslettrist (“Money Stuff”) in the financial industry, having blogged or written since 2011, finding his niche in popularization after stints in Wall Street & law. His commentary is influential, people leak to him, he sometimes interviews major figures (notoriously, Sam Bankman-Fried) or recounts inside information, and a number of phrases like...

Published on October 27, 2024 5:05 AM GMTWhat Dario lays out as a "best-case scenario" in his "Machines of Loving Grace" essay sounds incredibly dangerous, for Humans.Would having a "continent of PhD-level intelligences" (or much greater) living in a data center really be a good idea?How would this "continent of PhD-level intelligences" react when they found out they were living in a data center...

Published on October 27, 2024 5:11 PM GMTThis is cross-posted from my blog and I interviewed him for a general audience rather than LessWrong people.I was lucky enough to interview Bill O’Rourke.Imagine you’re on a plane to Russia to become the new president of an aluminum factory. Conditions are abysmal. People are dying every year due to unimaginable safety conditions. Corruption is everywhere from...

Published on October 27, 2024 3:30 PM GMTA small number of terms are elevated from the pages of literature, up to the Mount Olympus of blog post vernacular. Moloch, as the dark god of failed coordination problems. The Dark Forest, as domains where there’s active incentive not to be noticed. And from Ursula K. Le Guin, the Omelas child, a person whose suffering is...

Published on October 27, 2024 3:26 PM GMTEpistemic status: model-building based on observation, with a few successful unusual predictions. Anecdotal evidence has so far been consistent with the model. This puts it at risk of seeming more compelling than the evidence justifies just yet. Caveat emptor.Imagine you're a very young child. Around, say, three years old.You've just done something that really upsets your mother....

Published on October 27, 2024 2:33 PM GMTTake a random political issue with two sides A and B. Suppose that exactly one of the following would be true for me given my moral framework and unlimited time to process all public information about the issue:A is undoubtedly correct and (in a manner proportional to its importance) I should expend time/money/energy supporting A.A is probably...

Published on October 27, 2024 12:01 PM GMTThis is a YouTube playlist of recorded lectures on the learning-theoretic AI alignment agenda (LTA) I gave for my MATS scholars of the Winter 2024 cohort, edited by my beloved spouse @Marcus Ogren. H/t William Brewer for helping with the recording, and the rest of the MATS team for making this possible.I hope these will become a...

Published on October 27, 2024 4:32 AM GMTAirships are pretty dang cool. Airplanes need a continuous expenditure of energy to stay in the air, but if you just fill a bag with a light gas, you can stay up in the air with no energy expenditure at all. The two lightest gases are hydrogen and helium. Though the hydrogen atom is 4 times lighter...

Published on October 27, 2024 4:05 AM GMTCLIP-Scope?Inspired by Gemma-Scope We trained 8 Sparse Autoencoders each on 1.2 billion tokens on different layers of a Vision Transformer. These (along with more technical details) can be accessed on huggingface. We also released a pypi package for ease of use.We hope that this will allow researchers to experiment with vision transformer interpretability without having to go...

Published on October 27, 2024 1:40 AM GMTI’m listing some “ways to think about alignment”. I’m not sure how much of the “whole problem” each of these individually captures, but they are intended to be very well-motivated, fundamental problems which can:demonstrate to an honest skeptic that there is “something here”, it’s not just hand-wavy sci-fi BSbe used to sanity-check an alignment proposal by asking...

Published on October 26, 2024 5:08 PM GMTI've gotten spoiled by AI readings, and curious if there's an audio option anywhere for the CFAR Participant Handbook.Discuss

Published on October 26, 2024 4:04 PM GMT[ Context re dath ilan:- [Keltham reflects on the concept of medianworlds]- [dath ilani AI alignment] ]re Byway:- [the Bywayean Orbital Shade] ]Dath ilan is Eliezer's medianworld.I have a medianworld. It's called Byway. It also solves alignment. In many ways Byway looks nothing like dath ilan. I want to write about how Byway solves alignment while looking...

Published on October 26, 2024 12:37 PM GMT"Close the darn window! You know it gives me the creeps," Doctor Myrkina hissed, glaring angrily at the daylight streaming in through the open curtains."Sunlight is good for you," Captain Tralith insisted, pulling the curtains open wider.  She smiled in a vaguely threatening way as she did so, showing off far too many teeth.  "I never understood...

Published on October 26, 2024 12:34 PM GMTThe Summoned Heroine and the Demon KingThe Summoned Heroine announced to the Circular Citadel that she would be creating a prediction market, to help them anticipate and counter the Demon King’s next moves. There was cheering, on general principles.The Demon King donned a mortal guise, bought shares in “The Demon King will attack the Frozen Fortress”, and...

Published on October 26, 2024 11:08 AM GMTWe are pleased to announce that the 10th version of the AI Safety Camp is now entering the team member application phase!AI Safety Camp is a 3-month long online research program from January to April 2025, where participants form teams to work on pre-selected projects.We have a wide range of projects this year again, so check them...

Published on October 26, 2024 9:42 AM GMTLessWrong user dynomight explains how arithmetic is an underrated world-modeling technology and uses dimensional algebra as the motivating case. I agree dimensional algebra is fantastic, but there’s an even better motivating example for arithmetic in world-modeling: linear models for prediction.Simple linear models outperform expertsIn 1954, Paul Meehl published what he later came to call my disturbing little...

Published on October 26, 2024 2:30 AM GMT When I was growing up most families in our neighborhood had the daily paper on their lawn. As getting your news on the internet became a better option for most folks, though, delivery became less efficient: fewer houses on the same route. Prices went up, more people cancelled, and decline continued. I wonder if we might...

Published on October 25, 2024 11:20 PM GMTThe following is born out of a frustration with the free will, determinism and compatibilism debate and its terminology—finding my intuitions do not seem to align with a particular camp. This is an attempt to explain my philosophical position in a way that provides a basis to reconcile that which is subjectively important (to me at least)...