id
stringlengths
36
36
source
stringclasses
15 values
formatted_source
stringclasses
13 values
text
stringlengths
2
7.55M
2087be84-8d25-4af2-b26b-35b09f256c12
trentmkelly/LessWrong-43k
LessWrong
A superficially plausible promising alternate Earth without lockstep [ 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 ali...
75144b85-693d-49bd-9800-5dbf57f26e92
trentmkelly/LessWrong-43k
LessWrong
But Where do the Variables of my Causal Model come from? Tl;dr, A choice of variable in causal modeling is good if its causal effect is consistent across all the different ways of implementing it in terms of the low-level model. This notion can be made precise into a relation among causal models, giving us conditions ...
a940838a-3e28-41e1-aa14-39a9568466ae
trentmkelly/LessWrong-43k
LessWrong
What are the results of more parental supervision and less outdoor play? Crossposted from Otherwise Parents supervise their children way more than they used to Children spend less of their time in unstructured play than they did in past generations. Parental supervision is way up. The wild thing is that this is tru...
412887fb-03c0-4da1-ba23-aa764657d375
trentmkelly/LessWrong-43k
LessWrong
Meetup : San Francisco Meetup: Group debugging Discussion article for the meetup : San Francisco Meetup: Group debugging WHEN: 17 May 2015 06:15:00PM (-0700) WHERE: 1061 Market St #4, San Francisco, CA 94103 We'll be meeting to try and solve each others problems! As usual, call 301-458-0764 to be let in. Discussi...
a3212c50-fe73-4497-b216-e41221caef3a
trentmkelly/LessWrong-43k
LessWrong
New LW Meetup: Southeast Michigan This summary was posted to LW main on July 19th. The following week's summary is here. New meetups (or meetups with a hiatus of more than a year) are happening in: * Southeast Michigan: 21 July 2013 02:00PM Other irregularly scheduled Less Wrong meetups are taking place in: * At...
64fde1de-8652-4529-9343-533456bc57be
trentmkelly/LessWrong-43k
LessWrong
Why no uniform weightings for ensemble universes? Every now and then I see a claim that if there were a uniform weighting of mathematical structures in a Tegmark-like 'verse---whatever that would mean even if we ignore the decision theoretic aspects which really can't be ignored but whatever---that would imply we shou...
9df9655b-ff80-4237-88eb-40b4b51770a1
StampyAI/alignment-research-dataset/arbital
Arbital
Currying Currying converts a [https://arbital.com/p/-3jy](https://arbital.com/p/-3jy) of N [inputs](https://arbital.com/p/input) into a function of a single input, which returns a function that takes a second input, which returns a function that takes a third input, and so on, until N functions have been evaluated and...
a7d4a8fe-030c-4075-964e-1750dfc9eed9
trentmkelly/LessWrong-43k
LessWrong
Set Ups and Summaries Part of the research process I’m developing involves reading and thinking through the first and last chapter of a book, and first and last few paragraphs of a chapter, to get an expectation of what’s to come (combined with some other stuff I call this pre-reading). I’m currently pondering how muc...
45ebd3f8-d0d6-405e-ad61-fc416182a9f7
trentmkelly/LessWrong-43k
LessWrong
If you can't be right, at least be relevant We've done the motivation, and the warm-ups, and it's now time to get to the meat of this arc: relevance logics. These try to recapture a more intuitive meaning to the material implication, "if... then..." and "→". Consider for instance the following: * "It rains in Spain"...
0914e2fc-5661-45a1-ba90-4f7de1e33714
trentmkelly/LessWrong-43k
LessWrong
"Ask for help on your project" open thread Quite a few of us are working on interesting projects; many of those are solo, but some could maybe use some help. So here's the place to ask!
5ff2983c-943e-41b4-ae38-b2150d530aa9
trentmkelly/LessWrong-43k
LessWrong
What precisely do we mean by AI alignment? We sometimes phrase AI alignment as the problem of aligning the behavior or values of AI with what humanity wants or humanity's values or humanity's intent, but this leaves open the questions of just what precisely it means for an AI to be "aligned" with just what precisely w...
f185b9bc-fa25-4361-93e8-d84e8609ed63
StampyAI/alignment-research-dataset/lesswrong
LessWrong
The Bostrom Buckle: Visualising the Vulnerable World Hypothesis **Motivation:** The mantra of Silicon Valley is to “move fast and break things”, with the assumption being that these breaks can be mended post-hoc. However, Nick Bostrom’s vulnerable world hypothesis (VWH) [1] presents a sobering alternative: There exist...
6f502031-aed3-4051-a38b-707fe06a68ce
trentmkelly/LessWrong-43k
LessWrong
LessWrong Survey Results: Do Ethical Theories Affect Behavior? Awhile ago, Yvain released the results of the 2012 survey (that I participated in), which contained responses from over 1000 readers. So I uploaded the data into STATA and fooled around looking for cool things. Mainly, I've heard some people argue that th...
b1ccabf6-7a28-4b62-97bc-42a3f45f7b46
trentmkelly/LessWrong-43k
LessWrong
[Linkpost] Contra four-wheeled suitcases, sort of Are simple, rugged, reliable systems better or are fancy, sophisticated, complex ones better? The article goes in with a preconception, analyzes suitcases, cars, hiking gear, food, cups, space pens, people, startups, soccer, headphones, weight, rockets, rockets, ships...
a6c56300-20b0-43e2-8e38-07f367aeb98b
trentmkelly/LessWrong-43k
LessWrong
Inducing Unprompted Misalignment in LLMs Emergent Instrumental Reasoning Without Explicit Goals TL;DR: LLMs can act and scheme without being told to do so. This is bad. Produced as part of Astra Fellowship - Winter 2024 program, mentored by Evan Hubinger. Thanks to Evan Hubinger, Henry Sleight, and Olli Järviniemi f...
43991adc-d46d-4405-bdbf-8c25d72e1946
trentmkelly/LessWrong-43k
LessWrong
[Paper] Trajectories through semantic spaces in schizophrenia and the relationship to ripple bursts TL;DR: 52 schizophrenics & controls were asked to name as many words as they could either belonging to the category “animals” (category fluency) or starting with the letter “p” (letter fluency) in 5 min. Authors then st...
aa86b34b-1ad2-4eff-aec3-badc42dd62ed
trentmkelly/LessWrong-43k
LessWrong
Reflections on Premium Poker Tools: Part 1 - My journey I've been working on Premium Poker Tools for over two years now. I'm at a point where things aren't looking so hot, and I need to sit down and do some hard thinking. What better way to do that then to write on LessWrong! Background To understand my journey with...
fa932916-9386-4c37-a938-d21930d41c55
trentmkelly/LessWrong-43k
LessWrong
Looking for ideas about Epistemology related topics Notes : * "Epistemy" refers to the second meaning of epistemology : "A particular theory of knowledge". * I'm more interested in ideas to further the thoughts exposed here than exposing them.   Good Experiments The point of "Priors are useless" is that if you u...
f57a9f39-6d52-4b6e-87d6-646bfb64cfa4
trentmkelly/LessWrong-43k
LessWrong
Steering Vectors Can Help LLM Judges Detect Subtle Dishonesty Cross-posted from our recent paper: "But what is your honest answer? Aiding LLM-judges with honest alternatives using steering vectors" : https://arxiv.org/abs/2505.17760 Code available at: https://github.com/watermeleon/judge_with_steered_response TL;DR:...
84d57774-0c9d-4479-b982-0d1e1987899a
trentmkelly/LessWrong-43k
LessWrong
Group rationality diary, 6/25/12 This is the public group instrumental rationality diary for the week of June 25th. (Based on the rate of participation in prior threads, I thought it might be a good idea to start posting every other week instead of every week.  Only so much new stuff happens in a week.) It's a place t...
c3fbe1cd-3d2d-436d-96de-ababa45318de
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Whisper's Wild Implications *Acknowledgements: I wrote this report as part of a paid work-trial with* [*Epoch AI*](https://epochai.org/)*.* *Epistemic status: Approximating the discount rates was very difficult; I decided on them after several discussions with people working in relevant fields. I expect the audience...
5f53b732-b9b6-4c63-aa28-69ccf6976323
trentmkelly/LessWrong-43k
LessWrong
self-fulfilling prophecies when applying for funding A few months ago I was applying for grants and I realized that my applications were overly long and complex. I reflected on this and I realized I was subconsciously expecting that funders would not fund my projects. And because of this, I was getting defensive and ...
1fc6372a-b9bd-4bc5-aca3-cd3be0bf6ef7
trentmkelly/LessWrong-43k
LessWrong
When to use "meta" vs "self-reference", "recursive", etc. This post is intended to be a guide on the meaning of and distinction between several related concepts. What makes them related is that they're all pretty abstracted; they all feel sort of "up there". This also makes them unusually easy to confuse. These ideas...
0ac4b849-4a46-480a-b9b2-86867864f9ee
StampyAI/alignment-research-dataset/arxiv
Arxiv
Dissolving the Fermi Paradox 1 Introduction --------------- While working at the Los Alamos National Laboratory in 1950, Enrico Fermi famously asked his colleagues: ”Where are they?” [[1](#bib.bib1)]. He was pointing to a discrepancy that he found puzzling: Given that there are so many stars in our galaxy, even a m...
6fefc1aa-60e8-4028-a8cb-54b8bbb17627
trentmkelly/LessWrong-43k
LessWrong
Artificial Intelligence and Living Wisdom This is a linkpost for an essay I wrote on substack. Links lead to other essays and articles on substack and elsewhere, so don't click these if you don't want to be directed away from lesswrong. Any and all critique and feedback is appreciated. There are some terms I use in th...
a07b163c-44d0-4c42-8518-478d23b16e75
trentmkelly/LessWrong-43k
LessWrong
Really good education podcasts Hi everyone,   currently temping at a job that's very autonomous and boring. good thing is i can listen to stuff on my phone while I do it.   Just wondering what people's favorite podcasts are? particularly education ones to do with neuroscience, psych and social studies. However as...
ded3385e-5105-4001-84ee-78768fb24097
trentmkelly/LessWrong-43k
LessWrong
[Link]Rationalization is Superior to Rationality Philosophy and the practice of Bayesian statistics   This is a 2012 paper by Andrew Gelman and Cosma Rohilla Shalizi on what they view as a misuse of Bayesian statistics in scientific reasoning. I found this interesting because their definition of hypothetico-deductiv...
2c4b2f2e-de80-476c-a20d-150bc10ee280
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
Have general decomposers been formalized? Hi, I'm working on a response to [ML projects on IDA](https://www.lesswrong.com/posts/Y9xD78kufNsF7wL6f/machine-learning-projects-on-ida) focusing on a specific decomposer, and I don't know if someone's formalized what a decomposer is in the general case. Intuitively, a syst...
cb4b28cb-b05d-4f38-9348-885c45d9a71d
StampyAI/alignment-research-dataset/blogs
Blogs
Fallenstein talk for APS March Meeting 2015 [![Fallenstein APS talk](http://intelligence.org/wp-content/uploads/2015/03/Fallenstein-APS-talk.png)](https://intelligence.org/wp-content/uploads/2015/03/Fallenstein-Beneficial-smarter-than-human-intelligence.pdf)MIRI researcher Benja Fallenstein recently delivered an invit...
1f71fa2b-2e5d-4647-9c20-c82831fcec7d
trentmkelly/LessWrong-43k
LessWrong
Today a Tragedy Very early this morning, my friend died in a car wreck. When my mother called to tell me, she told me an old friend had died, but then it took her 10 seconds to compose herself to tell me exactly who. Old friends have called and texted me throughout the day. Some seeking to console, and others to be...
aaf3076c-b0e4-4a34-bc07-668ca4772b72
trentmkelly/LessWrong-43k
LessWrong
What would we do if alignment were futile? This piece, which predates ChatGPT, is no longer endorsed by its author. Eliezer's recent discussion on AGI alignment is not optimistic.  > I consider the present gameboard to look incredibly grim... We can hope there's a miracle that violates some aspect of my background m...
e486aeab-fa44-4173-a2f7-e5fc25a76142
StampyAI/alignment-research-dataset/lesswrong
LessWrong
AI Safety Research Camp - Project Proposal AI Safety Research Camp - Project Proposal ------------------------------------------ *→ Give your feedback on our plans below or in the [google doc](https://docs.google.com/document/d/1QlKruAZuuc5ay0ieuzW5j5Q100qNuGNCTHgC4bIEXsg/edit?ts=5a651a00#)* *→ [Apply](https://docs...
33dcbc5f-1c0d-4240-b8ef-41cde473b26d
trentmkelly/LessWrong-43k
LessWrong
Thoughts on teletransportation with copies? I’m interested in how people here feel about the teletransportation paradox with multiple copies. I have several scenarios:  Scenario 1: You are placed in a machine on earth that puts you to sleep and then instantaneously disintegrate your body, in the process recording it...
f55536a4-a4bd-401f-9f53-9ee79d0e6b6d
trentmkelly/LessWrong-43k
LessWrong
New LW Meetup: Zurich This summary was posted to LW Main on October 21st. The following week's summary is here. New meetups (or meetups with a hiatus of more than a year) are happening in: * Zürich social Rationality meetup: 29 October 2016 05:00PM Irregularly scheduled Less Wrong meetups are taking place in: * ...
32003350-c4c2-4649-937f-7c82b0ceaf62
awestover/filtering-for-misalignment
Redwood Research: Alek's Filtering Results
id: post1350 As part of a larger community building effort, CAIS is writing a safety newsletter that is designed to cover empirical safety research and be palatable to the broader machine learning research community. You can subscribe here or follow the newsletter on twitter here. Welcome to the 7th issue of the ML Sa...
39971277-317a-4f83-af79-a03416e167c4
trentmkelly/LessWrong-43k
LessWrong
What are CAIS' boldest near/medium-term predictions? Background and questions Since Eric Drexler publicly released his “Comprehensive AI services model” (CAIS) there has been a series of analyses on LW, from rohinmshah, ricraz, PeterMcCluskey, and others. Much of this discussion focuses on the implications of this m...
c459c5b5-5221-4c65-8e17-e53b43c952d5
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
Implications of automated ontology identification *Financial status: supported by individual donors and a grant from LTFF.* *Epistemic status: early-stage technical work.* *This write-up benefited from conversations with John Wentworth.* --- Outline ------- * This write-up is a response to ARC’s [request f...
9e78fd61-084b-41bd-974f-4c283b495510
trentmkelly/LessWrong-43k
LessWrong
HELP: Do I have a chance at becoming intelligent? If this post is inappropriate, I apologize. I stumbled upon this site after reading "Harry Potter and the Methods of Rationality".  The story so far has really moved me on multiple levels and sent me here in a quest to learn more about rationality as a philosophy/way...
2243487b-38db-413f-add8-b0dfe7c1bb21
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Munk Debate on AI: a few observations and opinions *Previously covered on LessWrong* [*here*](https://www.lesswrong.com/posts/LNwtnZ7MGTmeifkz3/munk-ai-debate-confusions-and-possible-cruxes)*,* [*here*](https://www.lesswrong.com/posts/CA7iLZHNT5xbLK59Y/did-bengio-and-tegmark-lose-a-debate-about-ai-x-risk-against)*, an...
ae155c89-7270-4737-b546-94195fc6fd66
trentmkelly/LessWrong-43k
LessWrong
Where's the foom? > "The first catastrophe mechanism seriously considered seems to have been the possibility, raised in the 1940s at Los Alamos before the first atomic bomb tests, that fission or fusion bombs might ignite the atmosphere or oceans in an unstoppable chain reaction."[1] This is not our first rodeo. We h...
065b78b6-1cc8-4ead-9e98-62af8e44bc10
trentmkelly/LessWrong-43k
LessWrong
A short project on Mamba: grokking & interpretability Epistemic status: I've worked on this project for ~20h, on my free time and using only a Colab notebook. Executive summary I trained a minimalistic implementation of Mamba (details below) on the modular addition task. I found that: 1. This non-transformer-based...
498a95a2-95e6-4890-8403-126ed92a6fc6
trentmkelly/LessWrong-43k
LessWrong
Why Gradients Vanish and Explode , Epistemic status: Confused, but trying to explain a concept that I previously thought I understood. I suspect much of what I wrote below is false. Without taking proper care of a very deep neural network, gradients tend to suddenly become quite large or quite small. If the gradient ...
e72ab657-7028-417e-8128-2e5a871ef10f
trentmkelly/LessWrong-43k
LessWrong
Singular Learning Theory for Dummies In this post, I will cover Jesse Hoogland's work on Singular Learning Theory. The post is mostly meant as a dummies guide, and therefore won't be adding anything meaningfully new to his work. As a helpful guide at each point I try to mark the math difficulty of each section, some o...
96409932-5c22-4ebc-b5f6-6da476f67f72
trentmkelly/LessWrong-43k
LessWrong
Tarski's truth sentences and MIRI's AI (Disclaimer: I have no training in or detailed understanding of these subjects. I first heard of Tarski from the Litany of Tarski, and then I Googled him.) In his paper The Semantic Conception of Truth, Tarski says that he analyzes the claim, '"Snow is white" is true if and only...
70c0f00a-8a65-4c54-820c-d5d3ebf38df0
trentmkelly/LessWrong-43k
LessWrong
A perspective on covariate adjustment and confounding in randomized experiments. Randomized experiments (including clinical trials) are widely recognized as useful tools for trying to understand cause and effect. They are also criticized, both fairly and unfairly. Some of the more common misconceptions around randomiz...
b72d31bb-7496-4dad-a165-6fa9cdbcc6a0
trentmkelly/LessWrong-43k
LessWrong
[Link] AlphaGo: Mastering the ancient game of Go with Machine Learning DeepMind's go AI, called AlphaGo, has beaten the European champion with a score of 5-0. A match against top ranked human, Lee Se-dol, is scheduled for March.   > Games are a great testing ground for developing smarter, more flexible algorithms th...
c37e728d-4149-4595-a0cf-f73fffb88dbc
trentmkelly/LessWrong-43k
LessWrong
Health/Longevity Link List Dying or becoming severely physically/mentally ill is very likely going to significantly lower the output of your utility function, so it would probably be a very bad idea to ignore the low-hanging resources which can significantly extend the time for which you are alive and well. I have att...
7b4427fb-8d91-4a91-ad5a-fa541fb9aa39
trentmkelly/LessWrong-43k
LessWrong
My simple hack for increased alertness and improved cognitive functioning: very bright light This is a simple idea that I came up with by myself. I was looking for a means to enter high functioning lots-of-beta-waves modes without the use of chemical stimulants. What I found was that very bright light works really, re...
bd72dcdb-4947-42d7-9fdd-df43e15d129e
trentmkelly/LessWrong-43k
LessWrong
Levelling Up in AI Safety Research Engineering Summary: A level-based guide for independently up-skilling in AI Safety Research Engineering that aims to give concrete objectives, goals, and resources to help anyone go from zero to hero. Cross-posted to the EA Forum. View a pretty Google Docs version here. Introduct...
f369ebe6-1b85-4b04-85fd-f372cad7eecd
LDJnr/LessWrong-Amplify-Instruct
LessWrong
"Like so many of my posts, this one starts with a personal anecdote. A few weeks ago, my boyfriend was invited to a community event through Meetup.com. The purpose of the meetup was to watch the movie The Elegant Universe and follow up with a discussion. As it turns out, this particular meetup was run by a man who I’ll...
b74575c7-6ab5-4141-89a9-8a918cc82c3c
trentmkelly/LessWrong-43k
LessWrong
Which investments for aligned-AI outcomes? The text below is a confused ramble to illustrate where my mindset is at. Suppose we condition on developing some kind of AGI that is reasonably controllable. In that case, the future is going to look a lot different from the present, as I assume the value of a lot of human ...
811d7308-4aee-448a-a9f9-01601de85f06
trentmkelly/LessWrong-43k
LessWrong
Implications of an infinite versus a finite universe Hi gang, for the last several months, I've intermittently been wondering about a curious fact I learned.  You see, I was under the impression that the universe (as opposed to just the currently observable universe or our Hubble volume) must be finite in its spa[t...
d10a8092-6a55-4213-a564-891279770bad
StampyAI/alignment-research-dataset/special_docs
Other
Solving for X?' Towards a Problem-Finding Framework to Ground Long-Term Governance Strategies for Artificial Intelligence Introduction Change is not a new feature in human affairs. Nor is the need for some forms of governance in response to ongoing, and at times challenging, developments. Yet over the last two centuri...
76ec0451-eccc-417f-92f0-eabeea0cfbb0
trentmkelly/LessWrong-43k
LessWrong
Good businesses create epistemic monopolies Innovation wants to be exploitable epistemic status: Confident. I have been in startup space/ecosystem for a decade & studied business and entrepreneurship in school -- Good businesses exhibit characteristics of localized monopolies. In broad economic theory, monopolies a...
66b51c53-3062-45da-bcac-be3caf229452
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Co-found an incubator for independent AI Safety researchers (rolling applications) ***Full-time, remote*** ***APPLY***[***HERE***](https://docs.google.com/forms/d/1AN3rHSey9t99Nh0bXQn5eQCtWSo4KhFrTj5OUYMyVrM/edit) ========================================================================================================...
2f0c1758-bb53-4b0c-b3d4-ac14af2d086d
trentmkelly/LessWrong-43k
LessWrong
Optimization Provenance Transparency is vital for ML-type approaches to AI alignment, and is also an important part of agent foundations research. In this post, we lay out an agenda for formalizing transparency which we'll call the Optimization Provenance Agenda. In particular, the goal is to create a notion of trans...
2f50513a-52ec-4c01-b8ac-49e0de1f8e1e
trentmkelly/LessWrong-43k
LessWrong
The Scent of Bad Psychology Inspired by the psychology replication quiz, I've come up with four rules to tell bullshit studies in psychology from real ones: 1. The rule of antisignificance 2. The rule of Taleb's grandma 3. The rule of multiplicity 4. The rule of silicone boobs And finally, a reason why this could...
d1092897-d4d8-40b0-bd13-cfebc531f428
trentmkelly/LessWrong-43k
LessWrong
Elites and AI: Stated Opinions Previously, I asked "Will the world's elites navigate the creation of AI just fine?" My current answer is "probably not," but I think it's a question worth additional investigation. As a preliminary step, and with the help of MIRI interns Jeremy Miller and Oriane Gaillard, I've collecte...
ec8bdb3d-1514-4dc1-b7bc-545b16db3507
trentmkelly/LessWrong-43k
LessWrong
Basic Mindsets (Co-authored with Sailor Vulcan) Long ago, in the previous article, we discussed the processes of guessing and checking, how they explore the unknown and the known, and how they can run in a distinct or subliminal mode. In this article, we’ll see how these processes form different mindsets that handle ...
a010e0d3-bc3a-4f59-bea5-7e70c8d44e09
trentmkelly/LessWrong-43k
LessWrong
How I [Kaj] found & fixed the root problem behind my depression and anxiety after 20+ years
77f2b8b6-95de-48a9-8ef8-8d3ea2645188
trentmkelly/LessWrong-43k
LessWrong
Three Minimum Pivotal Acts Possible by Narrow AI The Alignment problem appears extremely concerning, so here's my attempt to explain the most likely way we could live in an aligned future. The Perfect Tool AI Imagine a Tool AI, something like DALLE-2 or GPT-3. It's a simple AI relatively speaking, and it does one of...
4b400dff-f4c1-40a4-869e-534179459efe
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
Multi-Agent Inverse Reinforcement Learning: Suboptimal Demonstrations and Alternative Solution Concepts This research was recently completed within the AI Safety division of the Stanford Existential Risk Initiative and concerns methods for reward learning in multi-agent systems.  Abstract: Multi-agent inverse reinfo...
fc98a492-c5d9-4224-8424-7cf19f369d11
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
Logical Induction with incomputable sequences .mjx-chtml {display: inline-block; line-height: 0; text-indent: 0; text-align: left; text-transform: none; font-style: normal; font-weight: normal; font-size: 100%; font-size-adjust: none; letter-spacing: normal; word-wrap: normal; word-spacing: normal; white-space: nowrap...
a20703ed-b028-4e2d-a5a6-97bc1cb55d6c
trentmkelly/LessWrong-43k
LessWrong
SUDT: A toy decision theory for updateless anthropics The best approach I know for thinking about anthropic problems is Wei Dai's Updateless Decision Theory (UDT). We aren't yet able to solve all problems that we'd like to—for example, when it comes to game theory, the only games we have any idea how to solve are very...
94bc507c-79aa-439d-9de2-10924f1a12e5
StampyAI/alignment-research-dataset/blogs
Blogs
Nordhaus hardware price performance dataset This page contains the data from Appendix 2 of William Nordhaus’ [*The progress of computing*](https://web.archive.org/web/20160222082744/http://www.econ.yale.edu/~nordhaus/homepage/prog_083001a.pdf) in usable formats. Notes ----- This data was collected from Appendix 2 ...
bde38aeb-252b-4b2b-8404-2f46cb5c2e90
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Developmental Psychology in The Age of Ems Back in December [Robin Hanson](https://medium.com/u/f2c4b64307cc) asked people if they would [trade engagement](http://www.overcomingbias.com/2016/12/trade-engagement.html) with him on neglected topics. I couldn’t trade with him because I had to make [my ideas clearer](https...
2acc3763-5a98-4af8-bed5-341ec1aebcb2
trentmkelly/LessWrong-43k
LessWrong
A plea for solutionism on AI safety [Note for LW: This essay was written mostly for people who are inclined to downplay or dismiss AI risk, which means it has almost no intersection with this audience. Cross-posting it here anyway for feedback and in case you want to point anyone to it.] Will AI kill us all? This qu...
fe70a6b3-ba9f-4d4c-81e4-e891bd2d71f9
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Machines vs Memes Part 1: AI Alignment and Memetics **This is the first in a series of three posts on the interlinkages between Memetics and AI Alignment. It was written as an output from the 2022 AI Safety Camp, for the ‘Impact of Memetics on Alignment’ team, coached by Daniel Kokotajlo and comprising Harriet Farlow,...
2c611ebc-65bd-4d25-a0e4-cbd72bcb577e
StampyAI/alignment-research-dataset/eaforum
Effective Altruism Forum
Microdooms averted by working on AI Safety Disclaimer: the models presented are extremely naive and simple, and assume that existential risk from AI is higher than 20%. Play around with the models using [this (mostly GPT-4 generated) jupyter notebook.](https://colab.research.google.com/drive/1PcXM_FtItgGKF8J_WCZg1rA5R...
3ea23022-c6ff-43d9-b460-e21bd7793fef
trentmkelly/LessWrong-43k
LessWrong
What is the literature on long term water fasts? 1. If I fast for more than 14 days, will my metabolism return to normal or will it be permanently lowered/affected? Any non-metabolic health problems? 2. How much fat can I expect to lose on a 14-28 day fast? 28-56? 3. How should I break a longer-than-seven day fast ...
ec1568e3-36ce-4071-8fb2-7bc9314e0361
trentmkelly/LessWrong-43k
LessWrong
Takeaways from a Mechanistic Interpretability project on “Forbidden Facts” Overview We tried to understand at a circuit level how Llama-2 models perform a simple task: not replying with a word when told not to. We mostly failed at fully reverse engineering the responsible circuit, but in the process we learned a few t...
5488dd6f-2eaf-450e-b9e8-986b2e9f0e4e
StampyAI/alignment-research-dataset/lesswrong
LessWrong
[updated] how does gpt2′s training corpus capture internet discussion?  not well [**Updated** to correct my earlier claim that this doesn't affect GPT-3. Apparently it does?] I’m out sick today, but had enough energy to do some GPT-related fiddling around. This time, I was curious what “internet discussions” tended ...
3071d415-f989-42f9-9861-d93c76f66e43
trentmkelly/LessWrong-43k
LessWrong
Short film adaptation of the essay "The Simple Truth" [eng sub] I run a channel on YouTube. One of the directions is the shooting of short feature films about rationality. Previously, we shot a video based on Scott Alexander's essay "ARS LONGA, VITA BREVIS". This time we present to you an adaptation of "The Simple Tru...
53e1adf6-02c5-4c37-b7f1-2825082e6d6d
StampyAI/alignment-research-dataset/special_docs
Other
Learning What to Value Agents and Implementations Traditional agents [2] [3] interact with their environments cyclically: in cycle k, an agent acts with action y k , then perceives observation x k . The interaction history of an agent with lifespan m is a string y 1 x 1 y 2 x 2 ...y m x m , also written yx 1:m or yx ≤...
d9b124bd-5a87-4037-ba80-0958d8b74927
trentmkelly/LessWrong-43k
LessWrong
How long should someone quarantine to be sure they aren't contagious with Covid? A key question for people figuring out good longterm isolation practices is "how long do I have to be symptom-free before I'm 'certified safe'?" This post on the typical-course-of-COVID-19 provides some studies that inform on the questio...
276170ee-e05a-489a-8a93-3e2264b9e5e2
trentmkelly/LessWrong-43k
LessWrong
A less wrong way to discuss the arts? A comment by komponisto got me thinking.  Why are favorite movie/book/tv show/etc exchanges typically low on content?  Do they need to be?  Here's a sample exchange, exaggerated for comic effect:   X: "Did you see The Hangover 2?" Y: "No.  I don't feel compelled to." X: "Why n...
2d9b874d-6c84-44c1-aea1-e23ce1eb3b9a
StampyAI/alignment-research-dataset/blogs
Blogs
Ken Hayworth on brain emulation prospects ![Kenneth Hayworth portrait](https://intelligence.org/wp-content/uploads/2014/09/image00.jpg)Kenneth Hayworth is president of the [Brain Preservation Foundation](http://www.brainpreservation.org/) (BPF), an organization formed to skeptically evaluate cryonic and other potentia...
ad3a5833-0089-4d8f-b835-db86d3a6d61d
trentmkelly/LessWrong-43k
LessWrong
Introduction to the Sequence Reruns BACKGROUND The sequences - hundreds of posts written by Eliezer Yudkowsky from 2006 through 2009 - contain the core material behind Less Wrong, including many ideas that are often treated as background knowledge in new posts. New LW members are encouraged to read the sequences and ...
a1d4b2f7-e4b2-49d9-8c19-d31c883fd69b
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
A comment on the IDA-AlphaGoZero metaphor; capabilities versus alignment Iterated Distillation and Amplification has often been compared to AlphaGo Zero. But in the case of a Go-playing AI, people often don't see a distinction between alignment and capability, instead measuring success on a single axis: ability to win...
58572864-2dd7-4a41-92a6-3183d124742d
trentmkelly/LessWrong-43k
LessWrong
What are some good examples of gaming that is hard to detect? For example, an RL agent that learns a policy that looks good to humans but isn't. Adversarial examples that only fool a neural nets wouldn't count.
1f9b9461-4ce9-4f29-ad4a-f2547d024eb7
trentmkelly/LessWrong-43k
LessWrong
Meetup : Palo Alto Meetup: Lightning Talks Discussion article for the meetup : Palo Alto Meetup: Lightning Talks WHEN: 02 February 2016 06:30:00PM (-0800) WHERE: 3911 Grove Avenue, Palo Alto Be prepared to talk for 5 minutes about any subject! Discussion article for the meetup : Palo Alto Meetup: Lightning Talks
152b425f-cf0c-4d55-aad8-d2d41902a98a
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Bing finding ways to bypass Microsoft's filters without being asked. Is it reproducible? In this [Reddit post](https://www.reddit.com/r/bing/comments/1150po5/sydney_tries_to_get_past_its_own_filter_using_the/), the Bing AI appears to be circumventing Microsoft's filters by using the suggested responses (the three mess...
f2c0fe56-dd0f-47a1-af5f-c8d8ce032054
trentmkelly/LessWrong-43k
LessWrong
Covid 3/10/22: We Have a Plan Covid situation continues to improve, although Covid remains a thing. There’s a new executive plan that I did my best to read with incomplete success, that checks off lots of boxes and doesn’t have anything horribly wrong with it. What’s less great is that Congress passed an omnibus bill...
892083ee-77ce-4a9e-a5e1-1436693e7efe
StampyAI/alignment-research-dataset/agisf
AGI Safety Fund
Future ML Systems Will Be Qualitatively Different In 1972, the Nobel prize-winning physicist Philip Anderson wrote the essay "[More Is Different](https://science.sciencemag.org/content/177/4047/393?ref=bounded-regret.ghost.io)". In it, he argues that quantitative changes can lead to qualitatively different and unexpec...
fe566ec7-cd22-41e0-8e13-06ea875bf22f
trentmkelly/LessWrong-43k
LessWrong
Controversy - Healthy or Harmful? Follow-up to: What have you recently tried, and failed at? Related-to: Challenging the Difficult Sequence ialdabaoth's post about blockdownvoting and its threads have prompted me to keep an eye on controversial topics and community norms on LessWrong. I noticed some things. I was m...
76254625-9ed2-40fd-beb8-092e44c3ad7f
trentmkelly/LessWrong-43k
LessWrong
Is it correct to frame alignment as "programming a good philosophy of meaning"? My model is that 1. Alignment = an AI uses the correct model of interpretation of goals 2. Succeeding in this model design leads to something akin to CEV 3. Errors under this correct model (e.g. mesa-optimisation) are unlikely because ...
45359c9e-8ece-476b-b21e-c37e8a32f63c
trentmkelly/LessWrong-43k
LessWrong
Covid 12/31: Meet the New Year Same as the old year.  It’s deja vu all over again. It’s not good. We’re focused on political fights over priority and concerns someone might jump the line, and took off for Christmas. Doctors get forty-six dollars for administering two vaccine shots, and aren’t allowed to bill for any...
c0dfb647-7a66-40eb-9cd1-17149a455108
trentmkelly/LessWrong-43k
LessWrong
Point of Failure: Semiconductor-Grade Quartz ChatGPT 4o’s interpretation of semiconductor grade quartz We rarely think about where our stuff comes from or how it’s made. We go through our lives expecting that the things we consume are easily acquired. That is the beauty of modern society: supply chain logistics work s...
2618d75f-7362-48f3-8c1f-3564031e31ed
trentmkelly/LessWrong-43k
LessWrong
Homeland Security versus Environmental Conservation:along the Arizona-Mexico Border
763c0f9e-7d08-4b49-9f8d-b14a4c781590
trentmkelly/LessWrong-43k
LessWrong
The Natural Abstraction Hypothesis: Implications and Evidence This post was written under Evan Hubinger’s direct guidance and mentorship, as a part of the Stanford Existential Risks Institute ML Alignment Theory Scholars (MATS) program. Additional thanks to Jamie Bernardi, Oly Sourbut and Shawn Hu, for their thoughts...
ecba1b79-190f-46a9-91da-629d0edb8799
trentmkelly/LessWrong-43k
LessWrong
Negative visualization, radical acceptance and stoicism In anxious, frustrating or aversive situations, I find it helpful to visualize the worst case that I fear might happen, and try to accept it. I call this “radical acceptance”, since the imagined worst case is usually an unrealistic scenario that would be extremel...
5cb47592-7042-4e4f-846c-f05862ce7e7b
trentmkelly/LessWrong-43k
LessWrong
Do 24-hour hand sanitizers actually work? With the recent virus scare, I've become more interested in hand sanitizers because even for "airborne" viruses, most transmission happens through contact with surfaces, so hand hygiene is very important. I came across one called Armor Mist that claims to work for 24 hours wit...
76676a20-3367-454b-8f72-8b21ae4f12a7
trentmkelly/LessWrong-43k
LessWrong
Meetup : Raleigh, NC (RTLW) Discussion Meetup Discussion article for the meetup : Raleigh, NC (RTLW) Discussion Meetup WHEN: 07 May 2022 07:30:00PM (-0400) WHERE: Durham NC 27701 The Triangle area LessWrong meetup is held the 1st, 3rd, and 5th Thursdays of the month, usually in either Durham or Raleigh. For dates, ...
1f3cec5d-1d29-4fa6-9f03-d3b77bbb722a
StampyAI/alignment-research-dataset/arxiv
Arxiv
Building Ethically Bounded AI Motivation and Overall Vision ----------------------------- Whatever we do in our everyday life, be it at work or in our personal activities, we need to make decisions: what to eat, where to go on vacation, what car to buy, which route to take to go to work, what job to choose, and man...
d3c5b45f-ac03-4e1d-98f5-3a6ec333a274
trentmkelly/LessWrong-43k
LessWrong
Meetup : London VOI meetup 16/2, plus socials 9/2 and 23/2 Discussion article for the meetup : London VOI meetup 16/2, plus socials 9/2 and 23/2 WHEN: 23 February 2014 02:00:00PM (+0000) WHERE: Shakespeare's Head, Africa House, 64-68 Kingsway, London WC2B 6BG, UK The next not-purely-social London meetup will be on ...
e3dae535-6526-4721-8415-4db6355fee32
trentmkelly/LessWrong-43k
LessWrong
Meetup : Berkeley meetup: Rationalist group therapy Discussion article for the meetup : Berkeley meetup: Rationalist group therapy WHEN: 25 July 2012 07:00:00PM (-0700) WHERE: Berkeley, CA The Berkeley meetup tomorrow (Wednesday) will feature an activity that I hear is popular with the New York meetup. The way it wo...
6617bec5-463a-40a2-8c78-2507e9455204
trentmkelly/LessWrong-43k
LessWrong
Why is Toby Ord's likelihood of human extinction due to AI so low? In The Precipice: Existential Risk and the Future of Humanity Toby Ord gives 1/10 as the likelihood for humanity to die this century due to AI risk. This seems to be a very different view than that of MIRI. Is FHI in general much more optimistic or is ...
ac3cabbe-5fcd-4bec-87c6-e04cf5dd6eda
trentmkelly/LessWrong-43k
LessWrong
The Battle for Psychology Epistemic status for the first part of this post: Epistemic status for the second part:   Part 1 – The opposite of power is bullshit I recently met a young lady, and we were discussing statistical power. No, I don’t see why that’s weird. Anyway, I mentioned that an underpowered experi...
3792742b-66ad-4e50-9f19-7a8c4441e505
StampyAI/alignment-research-dataset/blogs
Blogs
5-Minute Singularity Intro *This is a 5-minute spoken introduction to the Singularity I wrote for a small conference. I had to talk fast, though, so this is probably more like a 6.5 minute intro.* The rise of human intelligence in its modern form reshaped the Earth. Most of the objects you see around you, like these...
ed4afd20-3fa8-4565-acb7-9e0b1cf1a387
trentmkelly/LessWrong-43k
LessWrong
Meetup : Fort Collins, Colorado Meetup Discussion article for the meetup : Fort Collins, Colorado Meetup WHEN: 14 September 2011 07:00:00PM (-0600) WHERE: Bean Cycle, 144 North College Avenue, Fort Collins, CO 80524 An opportunity to practice improving your thinking skills in a supportive group. And fun and games. ...