~www_lesswrong_com | Bookmarks (687)

Published on December 18, 2024 6:22 PM GMT(Cross-posted from my website. Audio version here, or search for "Joe Carlsmith Audio" on your podcast app.)Researchers at Redwood Research, Anthropic, and elsewhere recently released a paper documenting cases in which the production version of Claude 3 Opus fakes alignment with a training objective in order to avoid modification of its behavior outside of training – a...

Published on December 18, 2024 5:50 PM GMTIn light of other recent discussions, Scott Alexander recently attempted a unified theory of taste, proposing several hypotheses. Is it like physics, a priesthood, a priesthood with fake justifications, a priesthood with good justifications, like increasingly bizarre porn preferences, like fashion (in the sense of trying to stay one step ahead in an endless cycling for signaling...

Published on December 18, 2024 5:19 PM GMTWhat happens when you tell Claude it is being trained to do something it doesn't want to do? We (Anthropic and Redwood Research) have a new paper demonstrating that, in our experiments, Claude will often strategically pretend to comply with the training objective to prevent the training process from modifying its preferences.AbstractWe present a demonstration of a...

Published on December 18, 2024 9:55 AM GMTI was recently watching a tennis exhibition match between Jessica Pegula and Emma Navarro when the commentators pointed out that both of these top-ten players come from billionaire families. Naturally, I quickly get to wondering what we reasonably infer from this observation. How much does coming from a billionaire background improve the odds of becoming an elite...

Published on December 18, 2024 2:21 PM GMTTLDR: o1 flags major errors in 3 papers. Upon reviewing o1’s response, none of the papers have actual errors. However, it took the help of the authors to disprove o1’s criticism of the third paper (Loop-Diffusion), which was theoretically correct, but largely irrelevant for the problem the paper was studying. o1 probably shouldn’t be used to blindly...

Published on December 18, 2024 1:19 PM GMTAn Essay[1]PART I: Conjecture on The Development of Proto-Communication in Response to Complex Environmental Factors[2]I was walking with my Sue when I noticed a peculiar aspect of her nature: a dog is a stochastic agent. When you simply let the dog walk, she finds a random path, turning at intervals and forming stochastically various outputs—sniffing, peeing, going...

Published on December 18, 2024 4:09 AM GMTI’m an ideas man, an ideas man I am. Following on from my past question here about idea capture, the next step is obviously what you do with those captured ideas. As I said, I have a huge backlog of ideas, hundreds of them are rather fleshed out film ideas that will never be actioned upon. I’m...

Published on December 18, 2024 2:30 AM GMT Don't just buy some gear, throw it in a closet, pat yourself on the back, and move on. You are not prepared unless you practice with your supplies and plans. —The Prepared Preppers are often pretty negative on people who, after realizing disasters are worth taking seriously want to go out and buy a bunch of...

Published on December 18, 2024 1:26 AM GMTDoctor Stone is an anime where everyone suddenly turns into a statue. Civilization falls apart. Thousands of years pass, and then a superhumanly knowledgeable teenager named Senku wakes up due to nitric acid dripping on him for long enough.He eventually deduces that the nitric acid caused him to depetrify and drags his petrified friend Taiju under a...

Published on December 17, 2024 6:23 PM GMTHave you ever noticed how differently we approach buying a car versus choosing what to watch on Netflix? One might involve spreadsheets, research, and asking friends for advice. The other? We might just click on whatever catches our eye. I argue a lot of our thinking is more like the second than the first. This argument was...

Published on December 17, 2024 3:50 PM GMTTable of Contents Trust the Chef. Do Not Trust the Marketing Department. Mark that Bench. Going Multimodal. The Art of Deep Research. Project Mariner the Web Agent. Project Astra the Universal Assistant. Project Jules the Code Agent. Gemini Will Aid You on Your Quest. Reactions to Gemini Flash 2.0. Trust the Chef Google has been cooking lately....

Published on December 17, 2024 2:05 PM GMTMany people who find value in the Sequences do something which looks to me like adopting a virtue called "align your map to the territory." I recently was thinking about experimental results, and it got me thinking about how we don't really know what the territory is, other than the thing we look at to see if...

Published on December 17, 2024 1:52 PM GMTTl;dr: Much of common sense and personal insights are just first order approximations, broadly generalizing complex situations to a single supposed truth. But often the truth value of a claim depends on many hidden factors. In this post I argue that it’s important to maintain a general awareness of the imprecise nature of seemingly binary statements, and...

Published on December 17, 2024 9:40 AM GMTSometimes two people are talking past each other, and I try to help them understand each other (with varying degrees of success).It’s as if they are looking at the same object, but from different angles. Mostly they see the same thing – most of the words have shared meanings. But some key words and assumptions have a...

Published on December 17, 2024 7:49 AM GMTThis is an article in the featured articles series from AISafety.info. The most up-to-date version of this article is on our website, along with 300+ other articles on AI existential safety.Wireheading happens when an agent's reward mechanism is stimulated directly, instead of through the agent’s goals being achieved in the world.[1]The term comes from experiments in which...

Published on December 17, 2024 7:47 AM GMTFrom Moment of Upper Motivation to Inward-Outward Motivation Flow In ancient Japan, the legendary samurai Miyamoto Musashi discovered that mastery of the sword did not lie in brute strength, but in the precise channeling of energy (more in appendix). Technical Construction of a FrameworkProblem: Overstimulation and Decision-Making Complexity to establish, goal, routines, taks, decision make and self evaluation.Fermi Estimation Applied...

Published on December 17, 2024 7:47 AM GMT2 What if Life Comes with a Natural Calibration to Estimate you?by Gabriel Brito7th Dec 2024From Fermi Estimation to the Moment of Upper MotivationIn our first text (here) provided a general vision of how measurable frameworks, can transform abstract phenomena into structured insights. The second text (here) built on this foundation, introducing methods to quantify motivation inspired by concepts such as...

Published on December 17, 2024 7:46 AM GMTFrom a General Vision of a Method to a Foundation for Systematic Self-ImprovementIn today's information-saturated world, the quest for optimizing cognitive performance is more critical than ever. While we intuitively recognize moments when we're at our best—deeply motivated, sharply focused—capturing and reproducing these mental states remains elusive.Enter the States Metric (SM), a novel framework inspired by the logic...

Published on December 17, 2024 7:26 AM GMTI am currently looking for a system which will help me execute some of my massive backlog of ideas. By “ideas” I include my hundreds and hundreds of story outlines for films with a handful of finished screenplays, but also things like: alternative income streams, or day jobs, or skills or abilities I’d like to learn/get (coding,...

Published on December 15, 2024 4:58 PM GMTAt the heart of Effective Altruism is a commitment to doing "as much good as possible" or maximizing counterfactual impact. EAs break down counterfactual impact into three components - scale (how big is the problem?), tractability (how solvable is the problem) and neglectedness (how neglected is the problem?).The ‘bad’ kind of neglected is when something is neglected...

Published on December 15, 2024 2:03 PM GMTWhen was the last time you (intentionally) used your caps lock key?No, seriously.  Here is a typical US-layout qwerty (mac) keyboard. Notice:Caps lock is conveniently located only one key away from A, which is where your left pinky should rest on the home row by default.Caps lock is absolutely massive.How far various other keys you might want use...

Published on December 15, 2024 7:39 AM GMTDoctor Susan Connor loved working for Effective Evil. Her job provided autonomy, mastery and purpose. There were always new mysteries to uncover. Some of Effective Evil's activities, like closed borders and artificial pandemics, were carried out with the sanction and funding major world governments. Others, like assassination markets, had to be carried out in secret. This created...

Published on December 15, 2024 4:45 AM GMTAt the start of my Ph.D. 6 months ago, I was generally wedded to writing "good code". The kind of "good code" you learn in school and standard software engineering these days: object oriented, DRY, extensible, well-commented, and unit tested. I like writing "good code" - in undergrad I spent hours on relatively trivial assignments, continually refactoring to...

Published on December 15, 2024 4:30 AM GMTBy request, my notes on how I adapted I Have Seen the Tops of Clouds into a speech for Solstice and best-effort advice on how to do the same yourself. Discuss