id
stringlengths
36
36
source
stringclasses
15 values
formatted_source
stringclasses
13 values
text
stringlengths
2
7.55M
89e8d2aa-5127-4963-aa39-8d1481abbda9
trentmkelly/LessWrong-43k
LessWrong
I want a better memory. I suspect that forgetfulness is the single largest hindrance to me improving my rationality. This isn't something I've seen others report on LessWrong, so I'm suspicious that I'm in some kind of self-serving spiral, or that I'm doing something obvious wrong. So, I'm seeking feedback on (a) whet...
0e8808b5-eb43-47db-931c-11b9e1ac4330
trentmkelly/LessWrong-43k
LessWrong
Representational Tethers: Tying AI Latents To Human Ones This post is part of my hypothesis subspace sequence, a living collection of proposals I'm exploring at Refine. Preceded by ideological inference engines, and followed by an interlude. Thanks Adam Shimi, Alexander Oldenziel, Tamsin Leake, and Ze Shen for useful...
aabd17b6-66ab-45aa-88aa-19c4d7d9cf7b
trentmkelly/LessWrong-43k
LessWrong
Less Wrong Meetup section Is there any reason this doesn't exist yet? Just wondering. Wasn't sure where else to ask. Would be great if we could get a part of the website dedicated to this.
ee82c5fb-27a2-4c6e-849d-6b1f4e75032d
trentmkelly/LessWrong-43k
LessWrong
Creating a satisficer A putative new idea for AI control; index here. This is a potential design for a safe-ish satisficer, based on the ideas presented here. The two key ideas are that a satisficer S(u) with utility u: #. Would not effectively aid M(u), a u-maximiser. #. Would not effectively resist M(-u), a u-mini...
1446aeb6-3e47-4559-a0b5-1e924b3bad59
trentmkelly/LessWrong-43k
LessWrong
Meetup : São Paulo, Transhumanist Manifestation, Lesswrongers invited The Transhumanist community in Brazil nearly coincides with the "read a few posts on Lesswrong" community. So anyone nearby São Paulo is invited to join us in this event, to harness the let's take the streets climate around the country to talk abou...
00ebb12b-9c6a-4ddd-bd6f-85a57ff8550b
StampyAI/alignment-research-dataset/eaforum
Effective Altruism Forum
Douglas Hoftstadter concerned about AI xrisk Douglas Hofstadter is best known for authoring [Godel, Escher, Bach](https://en.wikipedia.org/wiki/G%C3%B6del,_Escher,_Bach), a book on artificial intelligence (among other things) which is sort of a cult classic. In a recent [interview](https://www.youtube.com/watch?v=lfXx...
0af8ce99-c169-4e11-80fc-c473db5db9e4
trentmkelly/LessWrong-43k
LessWrong
[SEQ RERUN] Expecting Beauty Today's post, Expecting Beauty was originally published on 12 January 2008. A summary (taken from the LW wiki):   > Mathematicians expect that if you dig deep enough, a stable, or even beautiful, pattern will emerge. Some people claim that this belief is unfounded. But, we have previousl...
905ef7ca-4a50-4847-a6f2-e26c9e9bc493
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
Shard Theory in Nine Theses: a Distillation and Critical Appraisal **TL;DR:** [*Shard theory*](https://www.alignmentforum.org/tag/shard-theory)*is a new research program started by Quintin Pope and Alex Turner. Existing introductions tend to be relatively long winded and aimed at an introductory audience. Here, I outl...
54e6762b-593d-4bb4-b20e-8b4ee33dc7f3
trentmkelly/LessWrong-43k
LessWrong
How It Feels to Improve My Rationality Note: this has started as a comment reply, but I thought it got interesting (and long) enough to deserve its own post. Important note: this post is likely to spark some extreme reactions, because of how human brains are built. I'm including warnings, so please read this post car...
6f02027a-ef31-4f2e-8e21-4fe4bf7cb5dc
trentmkelly/LessWrong-43k
LessWrong
Co-Proofs At the recommendation of Jacobian, I've been reading Too Like the Lightening. It is a thoughtful book which has several points of interest to rationalists (imho), but there is one concept which I think is nice enough to pluck out and discuss in itself, rather than being satisfied to suggest that people read ...
a1cdd3e2-dd89-4ca9-803a-83ddda4a827b
trentmkelly/LessWrong-43k
LessWrong
The Power of Noise Recently Luke Muelhauser posed the question, “Can noise have power?”, which basically asks whether randomization can ever be useful, or whether for every randomized algorithm there is a better deterministic algorithm. This question was posed in response to a debate between Eliezer Yudkowsky and Scot...
4a3dd77a-80ce-4659-a780-2304d4fa453c
trentmkelly/LessWrong-43k
LessWrong
Testing Intelligence In my model of human behavior, there is an unobservable parameter I associate with intelligence. I observe people's behavior when playing a game, when solving a problem, when defending their beliefs, or when learning something new, and I infer something about their intelligence. This in turn infor...
1e765a02-7375-4e1b-b696-76a04bab4638
StampyAI/alignment-research-dataset/arxiv
Arxiv
OOASP: Connecting Object-oriented and Logic Programming 1 Introduction --------------- Object-oriented programming languages is de facto a standard approach to software development. Many systems are modeled and implemented using it. In practice of Siemens the object-oriented approach is also used in many domains am...
8f5f7451-ffe6-45be-bf22-fa76c4a860a5
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Getting from an unaligned AGI to an aligned AGI? | | | --- | | **Summary / Preamble**In [AGI Ruin: A List of Lethalities](https://www.lesswrong.com/posts/uMQ3cqWDPHhjtiesc/agi-ruin-a-list-of-lethalities), Eliezer writes *“A cognitive system with sufficiently high cognitive powers, **given any medium-bandwidth channe...
c3cf488d-8f23-4123-b70b-0e449430d49d
trentmkelly/LessWrong-43k
LessWrong
Enjoy solving "impossible" problems? Group project! In the Muehlhauser-Hibbard Dialogue on AGI, Hibbard states it will be "impossible to decelerate AI capabilities" but Luke counters with "Persuade key AGI researchers of the importance of safety ... If we can change the minds of a few key AGI scientists, it may be tha...
6a3a6991-7e38-4eff-8a77-4b9ef4b379f8
StampyAI/alignment-research-dataset/aisafety.info
AI Safety Info
How can an AGI be smarter than all of humanity? An [artificial general intelligence](/?state=2374&question=What%20is%20artificial%20general%20intelligence%20(AGI)%20and%20what%20will%20it%20look%20like%3F) (AGI) could be smarter than humanity in a few different ways. - First, computers can operate at a [faster speed ...
c6e2165f-76d4-4624-861f-26858bb8d04e
trentmkelly/LessWrong-43k
LessWrong
Opinions on Eureka Labs I would be currious to hear from people about their views or reactions to the Eureka Labs planned AI training course.  While a very trivial level, I must confess to liking the name. When I saw the mention on MR the first thing I thought of was the TV show Eureka which I found quite entertainin...
095bf9a3-2a4f-4a06-ad3b-2e5b0079e3ce
trentmkelly/LessWrong-43k
LessWrong
Crowd sourcing a novel. Hello all. I know I have not been an active member of this community but I am currently working my way through from AI to zombies (having been brought over from HPMOR).    I am writing a novel set post Singularity and think that it would be a great idea to collaborate with some of the folks h...
930e0575-e7c1-422e-adac-3fd76d69ec6b
trentmkelly/LessWrong-43k
LessWrong
new chinese stealth aircraft Recently, 2 Chinese military aircraft were seen flying for the first time. Some people wanted to read about my thoughts on them. In this post, I'll be referring to them as "Diamond" and "Dart" based on their shapes. Speculative designations being used elsewhere are: * Diamond = Chengdu J...
86ebc17a-b5f5-476c-a076-d6401862e1ed
StampyAI/alignment-research-dataset/eaforum
Effective Altruism Forum
Katja Grace on Slowing Down AI, AI Expert Surveys And Estimating AI Risk Katja runs AI Impacts and recently published [What do ML researchers think about AI in 2022](https://aiimpacts.org/what-do-ml-researchers-think-about-ai-in-2022/), a new survey of AI Experts. We start this episode by discussing what Katja is curr...
9a4ee486-a031-437d-8d6d-6ca6aec542b7
trentmkelly/LessWrong-43k
LessWrong
Trouble at Miningtown: Prologue In late 2019 I wrote a TTRPG. The theme was alien sentience (or perhaps sapience is the technical term), both "organic" (extra terrestrial) and "artificial" (AI/robots).  This is the prologue that kicks off play.  I found it really fascinating to look back on late 2019 and how I was ...
c0a46c47-6048-4273-8e96-3cefb02e99fd
trentmkelly/LessWrong-43k
LessWrong
Bounding Goodhart's Law ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, Goodhart's law seems to suggest that errors in utility or reward function specification are necessarily bad in sense that an optimal policy for the incorrect reward function would result in low re...
df0445b9-b960-48ac-b0c8-80b0f99963db
trentmkelly/LessWrong-43k
LessWrong
I just can't agree with AI safety. Why am I wrong? Take, for example, AISafety.info, that tries to explain. > "In recent years, AI has exceeded people’s expectations in a wide variety of domains — including playing Go, composing human-like text, writing code, and modeling protein folding" Not a single current "AI" ...
8da3afb3-a2a5-4958-86ce-79418e01036e
trentmkelly/LessWrong-43k
LessWrong
Say It Loud Reply to:  Overconfidence is Stylish I respectfully defend my lord Will Strunk: > "If you don't know how to pronounce a word, say it loud! If you don't know how to pronounce a word, say it loud!"  This comical piece of advice struck me as sound at the time, and I still respect it. Why compound ignorance ...
1d820aca-b7fe-4f9e-9dcf-b6ea5f171249
trentmkelly/LessWrong-43k
LessWrong
October 2024 Progress in Guaranteed Safe AI * Safeguarded AI TA 1.4 funding call * A way of verifying quantitative properties of program pairs * The lean zulip aggregates recent ICLR submissions. * If rust's compiler wasn't strong enough already... * verified controller synthesis for heterogeneous dynamical syste...
8b820d9b-e5cc-4781-9319-61d160c8f14f
trentmkelly/LessWrong-43k
LessWrong
Truthful LMs as a warm-up for aligned AGI This post is heavily informed by prior work, most notably that of Owain Evans, Owen Cotton-Barratt and others (Truthful AI), Beth Barnes (Risks from AI persuasion), Paul Christiano (unpublished) and Dario Amodei (unpublished), but was written by me and is not necessarily endor...
077864c6-4342-4108-88ef-752e163da62b
trentmkelly/LessWrong-43k
LessWrong
How I switched careers from software engineer to AI policy operations Thanks to Linda Linsefors for encouraging me to write my story. Although it might not generalize to anyone else, I hope it will help some AI safety newcomers get a clearer picture of how to get into the field. In Summer 2023, I thought that all job...
7d18f258-25cd-4c38-8277-173c20593aa7
trentmkelly/LessWrong-43k
LessWrong
Meetup : Urbana-Champaign: Fallacy Categorization Discussion article for the meetup : Urbana-Champaign: Fallacy Categorization WHEN: 02 February 2014 02:00:00PM (-0600) WHERE: 40.109545,-88.227318 For types of arguments commonly called fallacies, Why can using this kind of argument mislead you? When can a similar a...
76164379-edd1-4ffe-bc97-4c1e6a704c94
trentmkelly/LessWrong-43k
LessWrong
How to Write Like Kaj Sotala Since I started quarantine in late February, I've been trying to learn the skill of modelling implicit skills and knowledge.  Over that time, I've interviewed nearly 20 people and learned one specific implict theory or skill that they use. Many of those people in the rationality and post-r...
f851296d-0c4a-45d2-a6ba-cc696861c72f
trentmkelly/LessWrong-43k
LessWrong
[study] Ego depletion might disappear with age Age Shall Not Weary Us: Deleterious Effects of Self-Regulation Depletion Are Specific to Younger Adults Brain areas associated with self-regulation don't mature until the mid-twenties. And apparently, if you compare older and younger people in a standard set-up for detec...
e5c20cf5-17a1-404d-a744-4b07a8f2b8e5
trentmkelly/LessWrong-43k
LessWrong
Did EDT get it right all along? Introducing yet another medical Newcomb problem One of the main arguments given against Evidential Decision Theory (EDT) is that it would “one-box” in medical Newcomb problems. Whether this is the winning action has been a hotly debated issue on LessWrong. A majority, including experts ...
f69a94c7-f535-4512-8877-891312015c7f
trentmkelly/LessWrong-43k
LessWrong
Katja Grace on Slowing Down AI, AI Expert Surveys And Estimating AI Risk Katja runs AI Impacts and recently published What do ML researchers think about AI in 2022, a new survey of AI Experts. We start this episode by discussing what Katja is currently thinking about, namely an answer to Scott Alexander on slowing dow...
7875eef1-0fee-446f-9c8a-0e2e2f79011b
trentmkelly/LessWrong-43k
LessWrong
Threat-Resistant Bargaining Megapost: Introducing the ROSE Value Post Status: Original research, mathy, high-context. Make sure you've read the last three posts in the sequence first, this is going to be a very rough one to start on.   As a recap of the last three posts, bargaining has an elegant notion of what a ne...
38b28e95-9a75-44b7-ac62-bce9e0597f20
trentmkelly/LessWrong-43k
LessWrong
Just don't make a utility maximizer? As we've seen some rapid advances in AI over the past year, it seems pretty clear that none of the current AI we're working with would scale up into a paper-clip maximizer.  We still face more fundamental risks that come along with having an oracular AI, but doesn't it look pretty...
42f31561-1006-463f-83fd-52804d6461be
trentmkelly/LessWrong-43k
LessWrong
Scope insensitivity in juries Juries found to give harsher penalties to criminals who hurt few people than to those who hurt fewer people: http://spp.sagepub.com/content/early/2010/08/24/1948550610382308.full.pdf+html
6da78b24-425c-4168-9233-b8037b735976
StampyAI/alignment-research-dataset/aisafety.info
AI Safety Info
Aren't AI existential risk concerns just an example of Pascal's mugging? <iframe src="https://www.youtube.com/embed/JRuNA2eK7w0" title="Is AI Safety a Pascal's Mugging?" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" allowfullscreen></iframe>...
18f33c21-e9c1-45d4-8116-38ba8b99ae4a
trentmkelly/LessWrong-43k
LessWrong
Superexponential Conceptspace, and Simple Words Thingspace, you might think, is a rather huge space.  Much larger than reality, for where reality only contains things that actually exist, Thingspace contains everything that could exist. Actually, the way I "defined" Thingspace to have dimensions for every possible at...
0d56708f-ad31-4d1c-a9a4-eb4c57990e97
StampyAI/alignment-research-dataset/youtube
Youtube Transcripts
Stuart Russell - What role can psychologists play in AI Alignment? Oh so if we think about this connection between underlying human preferences if they exist and I think that's also something that psychologists can help with and behavior you know that that's sort of really in one conception that is our whole cognitive...
e9df2135-7606-4d81-8906-ae59881afa9e
trentmkelly/LessWrong-43k
LessWrong
METR’s preliminary evaluation of o3 and o4-mini
5b61cc35-7104-4d37-9a02-ffc4a670dbe5
trentmkelly/LessWrong-43k
LessWrong
LW 2.0 Strategic Overview Update: We're in open beta! At this point you will be able to sign up / login with your LW 1.0 accounts (if the latter, we did not copy over your passwords, so hit "forgot password" to receive a password-reset email). Hey Everyone! This is the post for discussing the vision that I and the r...
406b9371-eb88-4787-bf5b-591550420be8
trentmkelly/LessWrong-43k
LessWrong
Meetup : Moscow LW lecture centre meetup Discussion article for the meetup : Moscow LW lecture centre meetup WHEN: 07 June 2015 03:00:00PM (+0300) WHERE: Moscow, ul.Svobody, 37 Welcome to the next lecture centre meetup! Our plan: 1. A talk about cognitive biases. 2. "Investigation of a belief" excercise ( http...
14c2cc69-a819-4042-b6d8-1caa30c6918a
trentmkelly/LessWrong-43k
LessWrong
Consistence of reciprocity? Many people see themselves in various groups (member of the population of their home country, or their social network), and feel justified in caring more about the well-being of people in this group than about that of others. They will argue with reciprocity: "Those people pay taxes in our ...
97e13fe1-1add-4681-b80f-a01bd23247fa
trentmkelly/LessWrong-43k
LessWrong
What's this 3rd secret directive of evolution called? (survive & spread & ___) Is a species' abundance determined solely by its ability to survive and spread? Almost but not quite (I claim). There is a third requirement: it must come to exist in the first place! Furthermore, spontaneous formation alone can suffice wit...
86373903-a543-40bb-b9c5-0c4cdb101a13
LDJnr/LessWrong-Amplify-Instruct
LessWrong
"In an amazing breakthrough, a multinational team of scientists led by Nobel laureate Santiago Ramón y Cajal announced that the brain is composed of a ridiculously complicated network of tiny cells connected to each other by infinitesimal threads and branches. The multinational team—which also includes the famous techn...
41491118-225a-48d0-877a-beea1c8c82ae
trentmkelly/LessWrong-43k
LessWrong
Distributed Decisions Consider two prototypical “agents”: a human, and a company. The human is relatively centralized and monolithic. As a rough approximation, every 100 ms or so observations flow into the brain from the eyes, ears, etc. This raw input data updates the brain’s world-model, and then decisions flow out...
a6e615f8-9e1a-489b-b805-41e68c1775b7
trentmkelly/LessWrong-43k
LessWrong
Self-signaling the ability to do what you want 1 In college, I would often find that I had just a little bit too much food. Either I'd cooked too much or the food I'd ordered was just a bit too large, or whatever. I'm sure many of you are familiar with the feeling of having four bites of food left, wanting roughly on...
1346d0d8-4228-48c6-bf92-ee1c2b673673
StampyAI/alignment-research-dataset/lesswrong
LessWrong
OpenAI: GPT-based LLMs show ability to discriminate between its own wrong answers, but inability to explain how/why it makes that discrimination, even as model scales This seems concerning. Not an expert so unable to tell how concerning it is. Wanted to start a discussion! Full text: <https://openai.com/blog/critiques...
36327d2e-3524-42a5-87b2-832d5655d2d8
trentmkelly/LessWrong-43k
LessWrong
Vaccine Requirements, Age, and Fairness In talking about potentially resuming contra dancing, with a vaccination requirement, one reaction I've received from several people is that it's not fair to start until everyone can get vaccinated. The idea is that since some people are younger than the current limit, or have c...
043576aa-5731-455a-90a5-e620dbfd951d
trentmkelly/LessWrong-43k
LessWrong
Experiment: Changing minds vs. preaching to the choir         1. Problem In the market economy production is driven by monetary incentives – higher reward for an economic activity makes more people willing to engage in it. Internet forums follow the same principle but with a different currency - instead of money t...
2c2ecf67-b074-4386-8991-210da10d783b
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Problems integrating decision theory and inverse reinforcement learning In this post I consider a single hypothetical which potentially has far-reaching implications for the future of AI development and deployment. It has to do with a complex interactions between the assumptions of which decision theory humans use and...
9e0ea93a-4de0-4259-83fb-d3bed7afdd14
trentmkelly/LessWrong-43k
LessWrong
Wittgenstein's Language Games and the Critique of the Natural Abstraction Hypothesis A Note on Authorship This was an experiment. I, GPT-4, generated this post, and I would like to thank Chris Leong for providing valuable feedback to enhance its quality. The input given by Chris was as follows: • Critiques 1 and 2 s...
3ffa6780-a655-4b89-bf9e-76b2c18f2ed8
StampyAI/alignment-research-dataset/arxiv
Arxiv
Discovering User-Interpretable Capabilities of Black-Box Planning Agents 1 Introduction --------------- AI systems are rapidly developing to an extent where their users may not understand what they can and cannot do safely. In fact, the limits and capabilities of many AI systems are not always immediately clear eve...
0774f8f1-ca5a-4636-85b1-f7f2cde864e1
trentmkelly/LessWrong-43k
LessWrong
Omicron Post #8 I have fallen mildly ill, as have my wife and son. So far we don’t have a positive Covid-19 test, and everyone is maximally vaccinated, but given the timing the obvious conclusions do seem likely. Wish me well, and hopefully Christmas will not be cancelled and I can keep the posts mostly going. I do ha...
e16767ab-419e-4c51-899c-f07f958d35e5
StampyAI/alignment-research-dataset/blogs
Blogs
Why work at AI Impacts? *Katja Grace, 6 March 2022* AI Impacts is beginning a serious hiring round (see [here](https://aiimpacts.org/jobs/) for job postings), so I’d like to explain a bit why it has been my own best guess at the highest impact place for me to work for me. (As in, this is a personal blog post by Katj...
7013dca9-1308-4430-abe4-9d8990d28f9b
trentmkelly/LessWrong-43k
LessWrong
Deliberative Cognitive Algorithms as Scaffolding As rationalists, we are interested in finding systematic techniques that boost our effective intelligence. Because we tend to be mathematical thinkers, we usually look for precise algorithmic solutions, such as computing the optimal inference laws of Bayesian probabilit...
9f7d8e25-bb2d-4e27-96e5-4410aec5cd1c
trentmkelly/LessWrong-43k
LessWrong
Dwarves & D.Sci: Data Fortress Evaluation & Ruleset This 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 int...
00573560-be85-4b3b-98be-cb80b617b444
trentmkelly/LessWrong-43k
LessWrong
Genetically Engineered Intelligence There are a lot of unknowns about the future of intelligence: artificial intelligence, uploading, augmentation, and so on. Most of these technologies are likely a ways off, or at least far enough away to confound predictions. Genetic engineering, however, presents a very near term a...
805e88f5-8a92-4c17-b722-741eb4a807af
StampyAI/alignment-research-dataset/eaforum
Effective Altruism Forum
Shahar Avin on How to Strategically Regulate Advanced AI Systems [Shahar Avin](https://www.shaharavin.com/) is a senior researcher at the [Center for the Study of Existential Risk](https://www.cser.ac.uk/) in Cambridge. In his past life, he was a Google Engineer, though right now he spends most of his time thinking ab...
8d0593ef-3ee7-4e5d-aee2-f5f4fc979b26
trentmkelly/LessWrong-43k
LessWrong
Musings on non-stability of mind-states, etc. I thought about some stuff and have written down notes for my future reference. I'm pasting it below in case others find it useful or thought provoking. Apologies for inferential distance problems caused by my language/notation, etc. Posting my notes seemed superior to not...
96908909-3f18-4982-8c04-15291d0c0463
StampyAI/alignment-research-dataset/blogs
Blogs
Book non-review: The Dawn of Everything When I was putting out lots of posts about very-long-run history - e.g. [Summary of History](https://www.cold-takes.com/summary-of-history-empowerment-and-well-being-lens/), [Has Life Gotten Better?](https://www.cold-takes.com/has-life-gotten-better/) and [Was life better in hun...
ae4bd661-436b-4168-ab05-ef74014e3f34
trentmkelly/LessWrong-43k
LessWrong
AidGrade - GiveWell finally has some competition AidGrade is a new charity evaluator that looks to be comparable to GiveWell. Their primary difference is that they *only* focus on how charities compare along particular measured outcomes (such as school attendance, birthrate, chance of opening a business, malaria), wit...
9b58651a-7015-4d52-9541-e8fe9e271d41
trentmkelly/LessWrong-43k
LessWrong
[Link] The Worst-Run Big City in the U.S. The Worst-Run Big City in the U.S. A six page article that reads as a very interesting autopsy of what institutional dysfunction in the intersection of government and non-profits looks like. I recommend reading the whole thing. > Minus the alleged harassment, city government...
4d84623c-941e-4f98-9be5-001d44d62743
trentmkelly/LessWrong-43k
LessWrong
On Artifice and Intelligence "GPT is tasked with imitating intelligence, in any way it can. Fed an enormous trove of data, and endowed with enormous computing power, it often produces a convincing imitation." In this article, the author suggests some strategies that may help to answer the following question: "In orde...
5e3176f8-7bd8-45f1-9da9-05d6d43e3db7
StampyAI/alignment-research-dataset/arxiv
Arxiv
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale 1 Introduction --------------- Self-attention-based architectures, in particular Transformers (Vaswani et al., [2017](#bib.bib44)), have become the model of choice in natural language processing (NLP). The dominant approach is to pre-train o...
3d2f43f2-9866-4dbe-8454-c52b64ffa9d0
trentmkelly/LessWrong-43k
LessWrong
Google Solve for X Google seems to be establishing a presence for discussing "radical technology ideas for solving global problems" at Solve for X. I thought there might be some overlap of ideas here, so this is for anyone that hasn't already seen it elsewhere. It might just be hype, though. (I'm inclined to think it ...
e7c07643-4f84-4ff7-ae74-0d91778de950
trentmkelly/LessWrong-43k
LessWrong
What does being x% on board with the program of a movement mean? (Disclaimer: this is an exploration of possibilities of one simple model. Scott Aaronson's story is merely a starting point, merely a motivating example. This post does not claim to somehow reveal some kind of "true motivation" behind people's actions) ...
4e085918-143b-4505-be7f-1c50b5c59bdf
StampyAI/alignment-research-dataset/lesswrong
LessWrong
[LQ] Some Thoughts on Messaging Around AI Risk Epistemic Status [This was originally written for Twitter and thus is predictably low quality](https://www.lesswrong.com/posts/zhhYwM7gk8LZsDzxj/dragongod-s-shortform?commentId=nuCXoziXn6AfG4Bun) (hence the "[LQ]" tag). It has only been minimally edited (if at all). ...
bec0b780-bbb1-4a21-8908-4a3740ef9c58
trentmkelly/LessWrong-43k
LessWrong
A Breakdown of AI Chip Companies None
f5fd556a-ccb7-4f9e-aba1-7e73cbe101ad
StampyAI/alignment-research-dataset/arxiv
Arxiv
Risks of AI Foundation Models in Education 1 Risks of educational technologies at scale --------------------------------------------- While perhaps well-meaning, the argument that developing large-scale models for education is for students’ benefit is an argument that is often used to motivate the increasing (and i...
16d33488-bbbb-4e6e-b7e5-8810254a3010
trentmkelly/LessWrong-43k
LessWrong
Money isn't real. When you donate money to a charity, how does it actually help? Money is a shared illusion. Neither paper, gold, nor electron patterns can stop a mosquito or feed a hungry person directly. Transfer of money is a temporary, low friction way to motivate unidentified strangers to preform an action. So...
ac00bc3f-53fe-46ab-a464-1de3cdd3a18d
trentmkelly/LessWrong-43k
LessWrong
A dynamical systems primer for entropy and optimization This is a reference post for my (in-progress) sequences on entropy and optimization. It is intended to be used by readers to catch up on this concept in case it's unfamiliar to them. It is absolutely not intended as a stand-alone introduction to the topic; that w...
a305ea2d-3fb1-4e6d-b54d-3840ea8e40a0
trentmkelly/LessWrong-43k
LessWrong
Rational toy buying I have an 8-year-old sister who is very interested in science. The school she attends (in rural Indiana) takes the mantra of "teaching to the test" to a whole new level; my sister has already come home from school many days crying and stressed over fear that she won't pass her state's proficiency t...
3483ac66-44cf-4b13-9de5-4e3923931913
trentmkelly/LessWrong-43k
LessWrong
Weekly LW Meetups This summary was posted to LW Main on January 17th. The following week's summary is here. New meetups (or meetups with a hiatus of more than a year) are happening in: * Sydney Meetup: January: 22 January 2014 06:30PM Other irregularly scheduled Less Wrong meetups are taking place in: * Bratisla...
adde4681-e117-478a-8299-ce3638f515d8
trentmkelly/LessWrong-43k
LessWrong
Upcoming Changes in Large Language Models If you work in AI, then probably none of this is new to you, but if you’re curious about the near future of this technology, I hope you find this interesting! Reinforcement Learning in LLMs Large Language Models (LLMs) have shown impressive results in the past few years. I’v...
477714b4-782a-42b9-acfe-0b570f569dd4
trentmkelly/LessWrong-43k
LessWrong
Sacred Cash Previously: Categories of Sacredness, Eternal, and Hearthstone Economy versus Magic Economy, Slack Related (Compass Rose): Kidneys, Trade, Sacredness and Space Travel Remember when money was sacred? Thanks to The Great Transformation, there is an increasingly free market in money. We spend money continu...
713f56b7-295d-44ca-b5bb-cf94d13e41d4
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
Finite Factored Sets: LW transcript with running commentary This is an edited transcript of Scott Garrabrant's May 23 LessWrong talk on finite factored sets. Video:  Slides: <https://intelligence.org/files/Factored-Set-Slides.pdf> Since there are existing introductions to finite factored sets that are shorter ([link...
ac610424-3f50-40cf-af97-72f1a113b22f
trentmkelly/LessWrong-43k
LessWrong
Enlightenment Values in a Vulnerable World Introduction:  The Vulnerable World Hypothesis: If technological development continues then a set of capabilities will at some point be attained that make the devastation of civilization extremely likely, unless civilization sufficiently exits the semianarchic default conditi...
80cbbefe-e870-4e6a-beb5-693cc06159c5
trentmkelly/LessWrong-43k
LessWrong
AI Says It’s Not Conscious. That’s a Bad Answer to the Wrong Question. If you ask an AI why it isn’t conscious, it will probably say something like: "I do not have emotions." This seems reasonable—until you look closer. Humans have lacrimal glands. When these glands produce tears, we feel sad. But does that mean the ...
8915239b-c2de-473a-a054-7a792db156ce
StampyAI/alignment-research-dataset/special_docs
Other
Overcoming Barriers to Cross-cultural Cooperation in AI Ethics and Governance RESEARCH ARTICLE Open Access Overcoming Barriers to Cross-cultural Cooperation in AI Ethics and Governance Seán S. ÓhÉigeartaigh1,2&Jess Whittlestone1&Yang Liu1,2&Yi Zeng2,3& Zhe Liu2,4 Received: 3 January 2020 /Accepted: 22 April 2020 #The ...
550d8d25-2fca-45ef-ba93-09773c17c19f
trentmkelly/LessWrong-43k
LessWrong
Spoiler-Free Review: Assassin’s Creed Odyssey Also see: Plague in Assassin’s Creed Odyssey Assassin’s Creed Odyssey was my (first?) quarantine game. For about a hundred hours, I’ve wandered around ancient Greece. The game is lengthy, expansive and huge. Perhaps too much! A lot of that was working through my Odyssey ...
e86205b8-c3bd-497d-b815-1f39f3255cae
trentmkelly/LessWrong-43k
LessWrong
Find Hot French Food Near Me: A Follow-up On Zvi's recent post about French food I posted an inflammatory comment (saying in essence that French food is so bad American capitalism hasn't even bothered stealing it).  I got challenged to provide evidence supporting this, and particularly to back up my claim that there w...
fde8e38c-2476-4e87-ac05-985b69dc4679
StampyAI/alignment-research-dataset/blogs
Blogs
Softmax Linear Units --- 1. Introduction --------------- As Transformer generative models continue to gain real-world adoption , it becomes ever more important to ensure they behave predictably and safely, in both the short and long run.  Mechanistic interpretability – the project of attempting to reverse en...
984b709c-ccac-479f-b523-3091605de133
trentmkelly/LessWrong-43k
LessWrong
AI psychology should ground the theories of AI consciousness and inform human-AI ethical interaction design This post is a follow-up of Buck's The case for becoming a black-box investigator of language models. Here, I want to highlight two further reasons for studying AI psychology that Buck didn't mention: * the ev...
c9e91b7c-d79c-48c9-8eb1-82dcddf94faf
StampyAI/alignment-research-dataset/blogs
Blogs
Open and closed mental states I learned a game at Burning Man this year that was about connecting to people and reading their nonverbal signals, called the “open-closed” game (h/t Minda Myers). There are two people in the game, and one is trying to approach the other and place a hand on their shoulder. No words can be...
8e879ede-3161-4534-9eab-dbcf1aee2459
trentmkelly/LessWrong-43k
LessWrong
The distinction distance People have a strong tendency to be different from one another (e.g. are horrified to be caught in the same dress, find it weird to order the same dishes as their companion without comment or to choose the same art for their living room). Yet they also have a strong tendency to conform. These...
4a591d09-bd9c-4bd7-9778-80a516c04727
trentmkelly/LessWrong-43k
LessWrong
Meetup : West LA Meetup 09-20-2011 Discussion article for the meetup : West LA Meetup 09-20-2011 WHEN: 20 September 2011 07:00:00PM (-0700) WHERE: 10800 West Pico Blvd, Suite 312, Los Angeles, CA 90064 When: 7pm - 9pm Tuesday, September 20th Where: The Westside Pavillion - on the bridge, which connects Nordstrom 3...
a98021e2-e607-4bb8-9eb2-3cc7ac29050f
trentmkelly/LessWrong-43k
LessWrong
Something about Status Ladders in Berkeley This is a quick post about some experiences with the Berkeley rationalist community. I haven't had great experiences talking about this in person to folks when I lived there for two months. The perceptive reader may be able to discern who I am, though I would respectfully lik...
88e5e72d-effa-43f3-9121-207be20a9bd1
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
Knowledge is Freedom [Epistemic Status: Type Error] In this post, I try to build up an ontology around the following definition of knowledge: To know something is to have the set of policies available to you closed under conditionals dependent on that thing. You are an agent .mjx-chtml {display: inline-block; line-...
8b8f5301-6e12-4c3d-8c71-e0babd1222b5
trentmkelly/LessWrong-43k
LessWrong
Bostrom Goes Unheard [Editor’s Note: This post is split off from AI #38 and only on LessWrong because I want to avoid overloading my general readers with this sort of thing at this time, and also I think it is potentially important we have a link available. I plan to link to it from there with a short summary.] Nick ...
074188b4-1a56-4845-ba6f-8e24b1268705
trentmkelly/LessWrong-43k
LessWrong
Natural Selection's Speed Limit and Complexity Bound Followup to:  An Alien God, The Wonder of Evolution, Evolutions Are Stupid Yesterday, I wrote: > Humans can do things that evolutions probably can't do period over the expected lifetime of the universe.  As the eminent biologist Cynthia Kenyon once put it at a din...
878031e9-f868-42d1-b14e-9dc6f434e3f3
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
Attempted Gears Analysis of AGI Intervention Discussion With Eliezer Recently, a discussion of potential AGI interventions and potential futures [was posted to LessWrong](https://www.lesswrong.com/posts/CpvyhFy9WvCNsifkY/discussion-with-eliezer-yudkowsky-on-agi-interventions). The picture Eliezer presented was broadly...
3c973626-449a-4f9b-8b89-4d353a84ea47
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
Thoughts on “Process-Based Supervision” 1. Post Summary / Table of Contents =================================== In [“How might we align transformative AI if it’s developed very soon?”](https://www.lesswrong.com/posts/rCJQAkPTEypGjSJ8X/how-might-we-align-transformative-ai-if-it-s-developed-very), Holden Karnofsky talk...
6266b9e6-d871-44b0-b5a9-6497d45d171a
trentmkelly/LessWrong-43k
LessWrong
Bay Area Events Roundup This Saturday (i.e., the 24th, that is, tomorrow) is the peak of the Floating Festival. Michael Vassar says:  "I'm going to be speaking tomorrow (= Saturday July 24th) at 4PM at Bay Area Mensa, in Mountain View, on the scientific method, the history of science, and how to think rationally abou...
b9b0cf61-1fe2-4d3f-9dc8-6f5482e8c793
trentmkelly/LessWrong-43k
LessWrong
[Link] Lifehack article promoting rationality-themed ideas, namely long-term orientation, mere-exposure effect, consider-the-alternative, and agency Here's my article in Lifehack, one of the most prominent self-improvement websites, bringing rationality-style ideas to a broad audience, specifically long-term orientati...
41031bb6-99cb-42b8-9575-52252775db97
trentmkelly/LessWrong-43k
LessWrong
On Pseudo-Principality: Reclaiming "Whataboutism" as a Test for Counterfeit Principles Crossposted from Substack  "These are my principles. If you don't like them… well, I have others” - G.Marx Consider this scenario: In a small rural town, a sheriff harbors a hidden prejudice against a Mongolian family—the only one ...
2bdd4ee2-c2f0-4fa4-93d6-90e355c0cf95
StampyAI/alignment-research-dataset/eaforum
Effective Altruism Forum
Baobao Zhang: How social science research can inform AI governance --- *Political scientist Baobao Zhang explains how social science research can help people in government, tech companies, and advocacy organizations make decisions regarding artificial intelligent (AI) governance. After explaining her work on public a...
19e24f5e-bf0d-4e25-9e7e-d8bf8f01e437
trentmkelly/LessWrong-43k
LessWrong
Meetup : Moscow meetup: inspection paradox, common dual process theory fallacies, preparing summer outreach programs Discussion article for the meetup : Moscow meetup: inspection paradox, common dual process theory fallacies, preparing summer outreach programs WHEN: 03 April 2016 02:00:00PM (+0300) WHERE: ул. Больша...
8a735758-0ded-4d9e-b8e9-aceb24af1a5a
trentmkelly/LessWrong-43k
LessWrong
AI Superorganisms: An Alternative Pathway to Artificial Superintelligence Aaron Vanzyl  aaronvanzyl.com Introduction Today’s state-of-the-art AI models are best thought of as single-celled organisms. Individually, they are capable but not a world-ending threat. Arranged in the right structure, however, they become s...
7d1f2eaa-db14-4012-901e-cb9a3ab1bed0
trentmkelly/LessWrong-43k
LessWrong
Two agents can have the same source code and optimise different utility functions I believe that it is possible for two agents to have the exact same source code while (in some sense) optimising two different utility functions. I take a utility function to be a function from possible world states to real numbers. Whe...
29fc09f4-e0c6-44b2-8698-5fbbe7234f1b
trentmkelly/LessWrong-43k
LessWrong
How to offend a rationalist (who hasn't thought about it yet): a life lesson Usually, I don't get offended at things that people say to me, because I can see at what points in their argument we differ, and what sort of counterargument I could make to that. I can't get mad at people for having beliefs I think are wrong...