id
stringlengths
36
36
source
stringclasses
15 values
formatted_source
stringclasses
13 values
text
stringlengths
2
7.55M
591f98ec-5289-4bb3-a126-dddde3230e78
trentmkelly/LessWrong-43k
LessWrong
Model Integrity: MAI on Value Alignment EVERYONE, CALM DOWN! Meaning Alignment Institute just dropped their first post in basically a year and it seems like they've been up to some cool stuff. Their perspective on value alignment really grabbed my attention because it reframes our usual technical alignment conversat...
49739364-5087-44b8-aae8-e60b71edf105
trentmkelly/LessWrong-43k
LessWrong
Meetup : [Boston] Using Causal Graphs to Understand Bias in the Medical Literature WHEN: 11 August 2013 02:00:00PM (-0400) WHERE: 25 Ames St, Cambridge, MA Meetup event page This talk is a very gentle introduction to modern causality theory, as developed by Jamie Robins and Judea Pearl. We will use as little mathema...
dddb0850-3747-4b19-b1e9-306f1ebb4657
trentmkelly/LessWrong-43k
LessWrong
Meetup : Montréal Meetup Discussion article for the meetup : Montréal Meetup WHEN: 10 October 2012 05:30:00PM (-0400) WHERE: Caffe Art Java - 645, av. du Président-Kennedy, Montréal, QC H3A 1K1 Montréal LessWrong Meetup The topic will be Bayesian vs Frequentist statistics, with a rundown of how Bayesian stats work...
df857f37-93ae-485b-869b-c4fce7af90ba
trentmkelly/LessWrong-43k
LessWrong
How do you turn an idea into a reality? (warning: involves politics) I am still relatively new to this community; if you do not want questions that imply or directly state political persuasions, I promise you won't get any more of them from me. Let's say that there's a movement to implement a specific idea that you b...
a7f6a992-1efd-4c83-904e-5942138cf94c
trentmkelly/LessWrong-43k
LessWrong
Did Christopher Hitchens change his mind about waterboarding? There's a popular story that goes like this: Christopher Hitchens used to be in favor of the US waterboarding terrorists because he thought it wasn't bad enough to be considered torture. Then he had it tried on himself, and changed his mind, coming to belie...
bdd0d700-cb7d-4db6-8ec2-6f37b2e57b04
trentmkelly/LessWrong-43k
LessWrong
You don't have to be "into EA" to attend EAG(x) Conferences I have recently shared a post about upcoming x/s-risk events, but I wanted to highlight this general part,  as I often come across this (understandable) misconception about the scope of EAG(x) conferences when I talk to rationalists/AIS people.  Here is a me...
8e4636ef-6f69-43f0-8587-d705a3524283
trentmkelly/LessWrong-43k
LessWrong
NeurIPS Safety & ChatGPT. MLAISU W48 Listen to this week’s update on YouTube or podcast. This week, we’re looking at the wild abilities of ChatGPT, exciting articles coming out of the NeurIPS conference and AGI regulation at the EU level.  My name is Esben and welcome to week 48 of updates for the field of ML & AI s...
4be78812-63c3-46a3-aa05-18ccf2c94da1
trentmkelly/LessWrong-43k
LessWrong
Personal story about benefits of Rationality Dojo and shutting up and multiplying My wife and I have been going to Ohio Rationality Dojo for a few months now, started by Raelifin, who has substantial expertise in probabilistic thinking and Bayesian reasoning, and I wanted to share about how the dojo helped us make a r...
6ef5c635-0905-4245-8093-a71fac663c6b
trentmkelly/LessWrong-43k
LessWrong
Meetup : Welcome Scott Aaronson to Texas Discussion article for the meetup : Welcome Scott Aaronson to Texas WHEN: 13 August 2016 06:00:00PM (-0500) WHERE: 4212 Hookbilled Kite Dr Austin, TX 78738 We're having another all-Texas party in Austin. We'll be welcoming Scott Aaronson, who's moved here to teach at UT Aust...
6c5bbb49-d7f4-44f0-b2a2-f4b5319e7857
StampyAI/alignment-research-dataset/blogs
Blogs
Sign up for DAGGRE to improve science & technology forecasting In [When Will AI Be Created?](http://intelligence.org/2013/05/15/when-will-ai-be-created/), I named four methods that might improve our forecasts of AI and other important technologies. Two of these methods were *explicit quantification* and *leveraging ag...
0557e66d-66f2-4aac-bf78-c0c9efeb7386
StampyAI/alignment-research-dataset/arxiv
Arxiv
The Off-Switch Game 1 Introduction --------------- From the 150-plus years of debate concerning potential risks from misbehaving AI systems, one thread has emerged that provides a potentially plausible source of problems: the inadvertent misalignment of objectives between machines and people. Alan Turing, in a 1951...
c5acefa7-d868-4029-acd4-5b69c099584f
trentmkelly/LessWrong-43k
LessWrong
Abandoned prototype video game teaching elementary circuit theory About a decade and several hobbies ago, I was inspired to make an educational video game that would teach the intuitions behind elementary circuit theory (batteries, switches, diodes, capacitors, inductors, etc.). After an inordinate amount of time,[1]...
788509f0-5796-43cd-9c4e-0a72022192bc
trentmkelly/LessWrong-43k
LessWrong
Thoughts on Ben Garfinkel's "How sure are we about this AI stuff?" I liked this talk by Ben. I think it raises some very important points. OTTMH, I think the most important one is: We have no good critics. There is nobody I'm aware of who is seriously invested in knocking down AI-Xrisk arguments and qualified to do s...
58990ad8-f7b9-4239-b362-189f8206ac33
StampyAI/alignment-research-dataset/arxiv
Arxiv
Evaluating Robustness of Neural Networks with Mixed Integer Programming 1 Introduction --------------- Neural networks have seen significant success on a range of problems. In particular, within computer vision, they represent the state-of-the-art for classification of images [[8](#bib.bib8), [17](#bib.bib17)] and ...
f94eafc7-d7bd-4200-81a5-948ebbd1c104
trentmkelly/LessWrong-43k
LessWrong
When do alignment researchers retire? At what point will it no longer be useful for humans to be involved in the process of alignment research? After the first slightly-superhuman AGI, well into superintelligence, or somewhere in between? Feel free to answer differently for different kinds of human involvement: * H...
c467786a-e2b1-430e-9686-f862efe414c1
trentmkelly/LessWrong-43k
LessWrong
Dancing to Positional Calling In a contra dance when only some of the dancers should take an action, the caller typically identifies them by role: "Robins start a Hey for Four" (or "Ladies" with gendered calling). In positional calling, instead, the caller doesn't use role terms: "pass right shoulders to start a Hey f...
3d6c605c-464a-4f77-8c76-71a4b2ab5ca4
trentmkelly/LessWrong-43k
LessWrong
Meditation course claims 65% enlightenment rate: my review TL;DR * I eliminated my impostor syndrome and dramatically reduced my work-related anxiety. I did this in a way that I think can be replicated.  * Different techniques work for different people. If you want to get the benefits of meditation, you should exper...
a042dd3d-5804-43b9-8be3-48f98e626160
trentmkelly/LessWrong-43k
LessWrong
“Dirty concepts” in AI alignment discourses, and some guesses for how to deal with them Meta: This is a short summary & discussion post of a talk on the same topic by Javier Gomez-Lavin, which he gave as part of the PIBBSS speaker series. The speaker series features researchers from both AI Alignment and adjacent fiel...
9a1c5fb1-9cb9-4006-ba39-dcb232930d6f
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Conjecture: A standing offer for public debates on AI **Tl;dr: If you want to publicly debate AI risk with us, send us an email at hello@conjecture.dev with information about you, suggested topics, and the suggested platform.** Public debates strengthen society and public discourse. They spread truth by testing ideas...
9b36ed01-add6-49ff-af68-0ecdeb3c7249
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
Deceptive Alignment .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; float: none; direction: ...
4bb4f2b9-c943-4573-a7cd-c756e85cda93
trentmkelly/LessWrong-43k
LessWrong
Rationalism and Dependent Origination? My understanding of the Buddhist concept of dependent origination or dependent arising is that everything arises from conditions required or conducive to its arising. Things are the way they are for reasons. Those reasons or causalities may be unfathomably complex, but everythin...
4e399606-0e45-4bb1-b37f-a2a18deb50d6
trentmkelly/LessWrong-43k
LessWrong
The Pavlov Strategy Epistemic Status: Common knowledge, just not to me The Evolution of Trust is a deceptively friendly little interactive game.  Near the end, there’s a “sandbox” evolutionary game theory simulator. It’s pretty flexible. You can do quick experiments in it without writing code. I highly recommend play...
96c0ebfa-57c5-493e-9cc0-0ab8483d8ecb
StampyAI/alignment-research-dataset/arxiv
Arxiv
The Best of Both Worlds: Combining Recent Advances in Neural Machine Translation 1 Introduction --------------- In recent years, the emergence of seq2seq models Kalchbrenner and Blunsom ([2013](#bib.bib24)); Sutskever et al. ([2014](#bib.bib34)); Cho et al. ([2014](#bib.bib7)) has revolutionized the field of MT by ...
47a34155-58d2-485c-b347-d7303dc904e4
trentmkelly/LessWrong-43k
LessWrong
An experiment Alexei's postmortem of Arbital got me thinking about incentivizing good online explanations, so I got an idea for a test that we can run here and now: What's your simplest, most vivid, most memorable explanation of correlation does not equal causation for someone who's unaware of the concept? In your ow...
0d7dfb3e-1f33-463f-a324-65fb4ae09bff
trentmkelly/LessWrong-43k
LessWrong
Stable Pointers to Value: An Agent Embedded in Its Own Utility Function (This post is largely a write-up of a conversation with Scott Garrabrant.) Stable Pointers to Value How do we build stable pointers to values? As a first example, consider the wireheading problem for AIXI-like agents in the case of a fixed util...
be52bdab-3cae-447d-b9f7-fef9cfef770b
trentmkelly/LessWrong-43k
LessWrong
Relative Configuration Space Previously in series:  Mach's Principle: Anti-Epiphenomenal Physics Followup to:  Classical Configuration Spaces > Warning:  The ideas in today's post are taken seriously by serious physicists, but they are not experimentally proven and are not taught as standard physics. > > Today's pos...
0d6fefa1-8418-437b-81e2-bce1670b2cb4
trentmkelly/LessWrong-43k
LessWrong
Exploring non-anthropocentric aspects of AI existential safety Epistemic status: These are some initial thoughts. Goal: Anthropocentric approaches dominate the field of AI existential safety. In particular, AI existential safety is usually formulated in terms of alignment to human goals and values. It might be useful...
863d077c-0088-449b-baf6-f6460bb4be82
trentmkelly/LessWrong-43k
LessWrong
Architecture-aware optimisation: train ImageNet and more without hyperparameters A deep learning system is composed of lots of interrelated components: architecture, data, loss function and gradients. There is a structure in the way these components interact - however, the most popular optimisers (e.g. Adam and SGD) d...
13f108a3-a256-4dbd-a89c-ca1384d7f151
trentmkelly/LessWrong-43k
LessWrong
A newcomer’s guide to the technical AI safety field This post was written during Refine. Thanks to Jonathan Low, Linda Linsefors, Koen Holtman, Aaron Scher, and Nicholas Kees Dupuis for helpful discussion and feedback. Disclaimer: This post reflects my current understanding of the field and may not be an accurate rep...
7e8ddbb8-ed1c-4847-bd0c-d8704758643a
trentmkelly/LessWrong-43k
LessWrong
Rationality Reading Group: Part G: Against Rationalization This is part of a semi-monthly reading group on Eliezer Yudkowsky's ebook, Rationality: From AI to Zombies. For more information about the group, see the announcement post. ---------------------------------------- Welcome to the Rationality reading group. Th...
71d93fbb-0774-4763-9982-5dbf2660de87
trentmkelly/LessWrong-43k
LessWrong
Meetup : Washington DC Meetup Discussion article for the meetup : Washington DC Meetup WHEN: 17 June 2012 03:00:00PM (-0400) WHERE: Portrait Gallery Topic is EY's response to the Givewell critique, although additional suggestions are solicited. (Also, I'll be there) Discussion article for the meetup : Washington D...
6ec488fe-c59e-4249-b2e4-655624d3525e
LDJnr/LessWrong-Amplify-Instruct
LessWrong
"TL;DR: It can be helpful to reframe arguments about tone, trigger warnings, and political correctness as concerns about false cognates/false friends. You may be saying something that sounds innocuous to you, but translates to something much stronger/more vicious to your audience. Cultivating a debating demeanor that i...
9ba16fd9-0668-4ed0-aa78-271221ab7f69
trentmkelly/LessWrong-43k
LessWrong
18 possible meanings of "I Like Red" A speaker says, "I like red". Here are a few of the many things they could mean[1] by that.  1. I like red. 2. I want you to believe that I like red. 3. I want myself to believe that I like red, and this can be assisted by a public statement. 4. I think that the community ...
07ac326a-35b4-44d4-ba4e-c44389ce06af
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
AI safety as featherless bipeds *with broad flat nails* There's a [famous story about Diogenes and Plato](https://en.wikipedia.org/wiki/Diogenes#In_Athens): > > [...] when Plato gave the tongue-in-cheek definition of man as "featherless bipeds," Diogenes plucked a chicken and brought it into Plato's Academy, sayin...
0b547803-d2e5-49be-aa68-36d879aa7caf
trentmkelly/LessWrong-43k
LessWrong
Minimum Viable Workout Routine is Dangerously Misinformative Edit: Dangerously misleading on one crucial point.   This started out as a short reply to the Less Wrong post Minimum Viable Workout Routine. Unfortunately I was unable to summarise my points sufficiently, so this reply grew into a post of it's own. I real...
e80b6a49-7c1b-47ea-a121-de384971e0fe
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Seeking Input to AI Safety Book for non-technical audience **TLDR:**  You are invited to make suggestions for ideas/information/framing for an upcoming AI safety book for a non-technical audience.  **Context:** I'm still [writing an accessible book about AI Safety/Risk](https://www.lesswrong.com/posts/E5Mwmuury...
231262c5-8df6-4c2d-a928-1ae6103941c3
trentmkelly/LessWrong-43k
LessWrong
Counterfactability This post will assume an understanding of the finite factored set ontology. It will be more speculative that the main FFS sequence, and I will leave out some proofs. It seems likely that I will later regret some of the definitions laid out here. I will be particularly sloppy in defining evidential a...
bd7355ee-ecf7-4ddc-8a36-bb35388492f7
StampyAI/alignment-research-dataset/blogs
Blogs
Generalizing Across Categories Humans are very good at correctly generalizing rules across categories (at least, compared to computers). In this post I will examine mechanisms that would allow us to do this in a reasonably rigorous manner. To this end I will present a probabilistic model such that conditional inferenc...
28929c2c-b24a-4189-a6ed-324981e3547e
trentmkelly/LessWrong-43k
LessWrong
Please don't throw your mind away Dialogue [Warning: the following dialogue contains an incidental spoiler for "Music in Human Evolution" by Kevin Simler. That post is short, good, and worth reading without spoilers, and this post will still be here if you come back later. It's also possible to get the point of this p...
b574aaec-aa73-4587-b6c6-374f2b228651
trentmkelly/LessWrong-43k
LessWrong
In Which I Make the Mistake of Fully Covering an Episode of the All-In Podcast I have been forced recently to cover many statements by US AI Czar David Sacks. Here I will do so again, for the third time in a month. I would much prefer to avoid this. In general, when people go on a binge of repeatedly making such inac...
605a1385-5d2c-448e-aab4-c5c97dfc8ea2
trentmkelly/LessWrong-43k
LessWrong
An Apprentice Experiment in Python Programming, Part 2 [Epistemic status: me trying to recall what happened in a 4.5-hour pair-programming session, during which I was focused on solving problems. This post was reconstructed from scratch code files, written messages (the session was over voice call) and my python shell...
ad2a68ff-8dcd-43f6-8b58-9a21cbbecfda
trentmkelly/LessWrong-43k
LessWrong
Apply to be a mentor in SPAR! Mentor applications are now open for the Spring 2025 round of the Supervised Program for Alignment Research (SPAR), running from February 10th to May 17th, 2025. Apply as a mentor SPAR is a remote-first part-time program that connects mentors and mentees for three-month AI safety and gov...
d5b3ed00-bea4-41a6-b974-5d3bf4cb8245
trentmkelly/LessWrong-43k
LessWrong
SIA is basically just Bayesian updating on existence I love the sleeping Beauty problem, but I think it leads people astray, by making them think that anthropic reasoning is about complicated situations with multiple duplicates. That leads to people saying erroneous things, like "SIA implies there are a lot of observe...
246db30e-3532-4a4e-8ce0-8f931bbc59c9
trentmkelly/LessWrong-43k
LessWrong
Uncategories and empty categories Savory What does "savory" mean when talking about food? Merriam-Webster says: * having a pleasant taste or smell * having a spicy or salty quality without being sweet * pleasing to the sense of taste especially by reason of effective seasoning * pungently flavorful without sweet...
225ae24c-cf1c-4f4f-97e9-188d3c4cc1c1
trentmkelly/LessWrong-43k
LessWrong
Book recommendations for the history of ML? Since I'm working in history and philosophy of science (HPS) and I'm trying to build my HPS model of alignment, in particular, I think it'd be good to read a book or two on the history of machine learning.  If you know of a good one, please share. Thanks! 
2fb205c4-f883-4fc2-a8a6-9fa760b1a0e7
trentmkelly/LessWrong-43k
LessWrong
Rant: the extreme wastefulness of high rent prices 09:46: Everyone wants to be close to everyone else to do good business effectively. Like Silicon Valley for example, smart people want to go there and be close to other smart people in order to start a successful business. 09:47: But as soon as they get close to each...
64251734-4146-4ae7-b9e0-4e6293850b64
trentmkelly/LessWrong-43k
LessWrong
What are the most common social insecurities? What do you think the most common social insecurities are? I'm doing rationality research on a grant, and answers will help my research.  I'm looking for a diversity of answers, please try not to look at the existing comments before responding. (Note: I'm deliberately n...
ccc1f504-fd7c-44ce-bb27-ed076cc691bd
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Dragon Ball's Hyperbolic Time Chamber > > A time dilation tool from an anime is discussed for its practical use on Earth; there seem surprisingly few uses and none that will change the world, due to the severe penalties humans would incur while using it, and basic constraints like Amdahl's law limit the scientific us...
ae3ae85d-632a-47d1-8c1e-5d243b7fe0b8
trentmkelly/LessWrong-43k
LessWrong
The 3 Books Technique for Learning a New Skilll When I'm learning a new skill, there's a technique I often use to quickly gain the basics of the new skill without getting drowned in the plethora of resources that exist. I've found that just 3 resources that cover the skill from 3 separate viewpoints(along with either ...
629cbd56-a384-42bc-b8f9-17af29224482
trentmkelly/LessWrong-43k
LessWrong
The Anti-Placebo Effect Just about everyone is familiar with the placebo effect at this point.  What I've discovered through my personal studies working with people suffering from anxiety and depression is that there is actually a significant related effect, which I have dubbed the anti-placebo effect.   Google says:...
18ccbb20-db75-4d8c-b508-576e102ff179
trentmkelly/LessWrong-43k
LessWrong
Looking for judges for critiques of Alignment Plans Hello!  AI-Plans.com recently held a "critique-a-thon," where participants submitted and refined 40+ critiques for AI Alignment plans. Here are the finalized critiques from the event: https://docs.google.com/document/d/1mW4SAxFN_aI6KyYXpl9qz5B9nVdeV9Xyc69GTNme5cA/ed...
e015de65-84e8-458f-ae62-d95cc655ebc5
StampyAI/alignment-research-dataset/arbital
Arbital
Abortable plans "Abortable" plans are those which can readily be switched to having very low net [impact](https://arbital.com/p/2pf) on the world. Suppose an AI is told to paint a car pink. The AI starts to do so by constructing replicating nanomachines that will paint the car pink. If the AI has successfully been ...
102cc3b5-c232-45e7-ab25-996837dd8ff4
trentmkelly/LessWrong-43k
LessWrong
The blue-minimising robot and model splintering A long time ago, Scott introduced the blue-minimising robot: > Imagine a robot with a turret-mounted camera and laser. Each moment, it is programmed to move forward a certain distance and perform a sweep with its camera. As it sweeps, the robot continuously analyzes the...
5ee5649c-c0bc-481f-b7fe-061d40113811
trentmkelly/LessWrong-43k
LessWrong
An Emergency Fund for Effective Altruists (second version) This is a rewrite of an old post of mine, incorporating ideas from Alicorn's earlier post on the same topic. One of the things stopping people from donating a substantial part of their income to charity is the risk of a future financial downfall, which can be...
d2f40cee-a3ed-4815-b90f-e7aa7da59bd7
StampyAI/alignment-research-dataset/arxiv
Arxiv
The Challenge of Crafting Intelligible Intelligence 1. Introduction ---------------- Artificial Intelligence (AI) systems have reached or exceeded human performance on many circumscribed tasks. As a result, they are increasingly deployed in mission critical roles — credit scoring, predicting if a bail candidate wil...
7be90ed5-2adc-4ce4-96a3-0aaeb40cc8c2
trentmkelly/LessWrong-43k
LessWrong
What's Up With Confusingly Pervasive Goal Directedness? Fictionalized/Paraphrased version of a real dialog between me and John Wentworth.   Fictionalized Me: So, in the Eliezer/Richard dialogs, Eliezer is trying to get across this idea that consequentialism deeply permeates optimization, and this is important, and t...
13685a82-cba2-466a-98f1-db2fc55464b7
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Superintelligence 7: Decisive strategic advantage *This is part of a weekly reading group on [Nick Bostrom](http://www.nickbostrom.com/)'s book, [Superintelligence](http://www.amazon.com/Superintelligence-Dangers-Strategies-Nick-Bostrom/dp/0199678111). For more information about the group, and an index of posts so far...
6468fa80-78fe-4b33-87ff-f31438004ef8
trentmkelly/LessWrong-43k
LessWrong
LessOnline (May 31—June 2, Berkeley, CA) A Festival of Writers Who are Wrong on the Internet[1] > LessOnline is a festival celebrating truth-seeking, optimization, and blogging. It's an opportunity to meet people you've only ever known by their LessWrong username or Substack handle. We're running a rationalist confer...
5ef9b48b-22af-40fc-b870-4fe1e954b757
trentmkelly/LessWrong-43k
LessWrong
An Introduction To Rationality This article is an attempt to summarize basic material, and thus probably won't have anything new for the hard core posting crowd. It'd be interesting to know whether you think there's anything essential I missed, though. Summary We have a mental model of the world that we call our bel...
fe8e896c-c2de-446a-9c48-c81bbd1dac1a
trentmkelly/LessWrong-43k
LessWrong
Forecasting Newsletter. June 2020. Highlights 1. Facebook launches Forecast, a community for crowdsourced predictions. 2. Foretell, a forecasting tournament by the Center for Security and Emerging Technology, is now open. 3. A Preliminary Look at Metaculus and Expert Forecasts: Metaculus forecasters do better. Sig...
5a5eb3e4-592a-4541-bea7-d898406f5b8f
trentmkelly/LessWrong-43k
LessWrong
Wolf Incident Postmortem Incident #210 Status Complete, one action item outstanding. Summary Sentinel consumed by wolf after repeated false alarms. Impact Loss of sentinel. No flock impact. Root causes Sentinel generated noisy alerts due to premature deployment, incomplete training, and overly monotonous task....
8438baa7-cbfd-48e4-b8a3-f5d992312738
trentmkelly/LessWrong-43k
LessWrong
Singularity Non-Fiction Compilation to be Written Call for Essays:<http://singularityhypothesis.blogspot.com/p/submit.html> The Singularity Hypothesis A Scientific and Philosophical Assessment Edited volume, to appear in The Frontiers Collection<http://www.springer.com/series/5342>, Springer Does an intelligence expl...
513214d5-b5eb-4773-a41b-3bde8b4f0653
trentmkelly/LessWrong-43k
LessWrong
Super intelligent AIs that don't require alignment An idealized computer program acts on a defined set of inputs, and produces a defined set of outputs. When we talk about AI misalignment there's two fears: 1. The AI will hack the system and produce outputs that it's not theoretically meant to be able to produce at...
32ff7c9a-660d-4515-8f5b-863c615b62b3
trentmkelly/LessWrong-43k
LessWrong
Formulating Reductive Agency in Causal Models The previous post talked about what agenty systems look like, in the context of causal models. The reductive agency problem asks: how are agenty systems built out of non-agenty pieces? In the context of causal models, we know that non-agenty models look like this: … and ...
48bfe337-78fe-4568-9e3c-a2607b068602
StampyAI/alignment-research-dataset/eaforum
Effective Altruism Forum
Has private AGI research made independent safety research ineffective already? What should we do about this? *This post is a variation on "*[*Private alignment research sharing and coordination*](https://www.lesswrong.com/posts/ZKNxbcACeHKEfgKvW/private-alignment-research-sharing-and-coordination)*" by porby. You can ...
8758187c-6ce4-4bc8-823c-5e2c65e9b93d
trentmkelly/LessWrong-43k
LessWrong
Meta: Meetup Section of the Site Recently there has been an explosion in the number of posts about meetups. I think this is a great thing, as it seems to me like one of the best uses of this site would be to facilitate the formation of rationality groups.  The problem with this is that they're starting to take over th...
32b9c5d8-75b1-4cbd-8e96-f93b3e24a0a6
trentmkelly/LessWrong-43k
LessWrong
Is there any policy for a fair treatment of AIs whose friendliness is in doubt? A game theory question. If there ever exists a roughly-human-level agenty AI who could grow to overpower humans but who humans have an opportunity to stop because takeoff is slow enough. Assume the AI could coexist with humanity but fears...
f45b06e9-c8a2-41b8-a950-559ba2db22cd
trentmkelly/LessWrong-43k
LessWrong
Picasso in the Gallery of Babel tl;dr When (not if) generative models get so good at faking human artefacts that no human can distinguish between human and synthetic artefacts, Strange Things will happen to the way humans consume and think about and make art. Claim 1: Models get better Generative AI is well on the w...
9ea7ecab-1a07-42ef-b8ab-1617014fccf2
trentmkelly/LessWrong-43k
LessWrong
Meetup : Meetup: Southwestern Ohio Discussion article for the meetup : Meetup: Southwestern Ohio WHEN: 09 September 2012 04:00:00PM (-0400) WHERE: 21 Greene Boulevard, Beavercreek, Ohio 45440, USA New location! Dinner at The Wine Loft near Dayton, Ohio. Reservation is under Dugan. After dinner activities may includ...
1e5541e7-7ff4-41af-9f6e-5dd91632f80b
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
AI Benefits Post 1: Introducing “AI Benefits” This is a post in a series on "AI Benefits." It is cross-posted from my [personal blog](https://cullenokeefe.com/blog/ai-benefits-1). For other entries in this series, navigate to the [AI Benefits Blog Series Index page](https://cullenokeefe.com/ai-benefits-index). This ...
c2ea2567-d0a0-4b00-9ac0-e2a0bf7f1106
trentmkelly/LessWrong-43k
LessWrong
Bad at Arithmetic, Promising at Math n-Cohesive Rings Definition: Let n be a positive integer. We define an n-cohesive ring to be a commutative ring S such that, for every prime p dividing the characteristic of S, pn divides the order of the multiplicative group S×. We define an n-cohesive ideal of a ring R to be an i...
eec2af49-7916-41fb-b723-5e1a85464d9e
trentmkelly/LessWrong-43k
LessWrong
The set of Logical Inductors is not Convex Sam Eisenstat asked the following interesting question: Given two logical inductors over the same deductive process, is every (rational) convex combination of them also a logical inductor? Surprisingly, the answer is no! Here is my counterexample. We construct two logical in...
dab987f5-a99e-4957-9f7b-d8bba7ad0df0
StampyAI/alignment-research-dataset/arbital
Arbital
Omnipotence test for AI safety Suppose your AI suddenly became omniscient and omnipotent - suddenly knew all facts and could directly ordain any outcome as a policy option. Would the executing AI code lead to bad outcomes in that case? If so, why did you write a program that in some sense 'wanted' to hurt you and wa...
d47af386-5788-49f1-bd44-7c5a04472b14
trentmkelly/LessWrong-43k
LessWrong
Against Modest Epistemology Follow-up to: Blind Empiricism ----------------------------------------   Modest epistemology doesn’t need to reflect a skepticism about causal models as such. It can manifest instead as a wariness about putting weight down on one’s own causal models, as opposed to others'. In 1976, Rob...
945fc211-2113-4ffe-90ff-b08027928a3b
trentmkelly/LessWrong-43k
LessWrong
What is the optimal frontier for due diligence? The title isn't quite right.  My current take is that: * There is a ~10k word post sharing some damaging information about someone else. * 3-4k words are describing accusations that are disputed. * However, a few have already had substantial evidence provided again...
5f240dd0-9a47-44b7-832e-ca3f2c370a3d
trentmkelly/LessWrong-43k
LessWrong
Trustworthy Computing 6. Trustworthy Computing Summary of entire Series:  An alternative approach to designing Friendly Artificial Intelligence computer systems. Summary of this Article: How to use non-open-ended task programs to permit a computing environment in which AIs could trust each other enough to cooperate o...
3fabcc55-e323-41a6-9e0b-66adcc188ff4
StampyAI/alignment-research-dataset/arxiv
Arxiv
The Algonauts Project: A Platform for Communication between the Sciences of Biological and Artificial Intelligence 1 Abstract ----------- In the last decade, artificial intelligence (AI) models inspired by the brain have made unprecedented progress in performing real-world perceptual tasks like object classificatio...
4bd3914e-d56a-4c7c-9253-537e6ebcddb5
trentmkelly/LessWrong-43k
LessWrong
Meetup : Boston - Aversion factoring and calibration Discussion article for the meetup : Boston - Aversion factoring and calibration WHEN: 15 January 2014 07:00:00PM (-0500) WHERE: Citadel, 98 Elm St Apt 1, Somerville, MA, 02144 Victoria Krakovna will be running a session on Aversion Factoring and Calibration (base...
0ec34657-986a-4a99-b5ae-712457d8498e
trentmkelly/LessWrong-43k
LessWrong
Can indifference methods redeem person-affecting views? My uninformed paraphrase/summary of "the person-affecting view" is: "classical utilitarianism + indifference to creating/destroying people". These views seem problematic (e.g. see Hillary Greeves interview on 80k), and difficult to support. Indifference methods...
bd207d72-ed35-4725-8e18-826b41132c98
trentmkelly/LessWrong-43k
LessWrong
Reply to Nate Soares on Dolphins > A similar definition of intelligence was expressed by Aquinas as "the ability to combine and separate"—the ability to see the difference between things that seem similar and to see the similarities between things which seem different. > > —A. R. Jensen In a June 2021 Twitter thread...
16992e1b-65a8-4c57-b0f4-fcb5856ecc65
trentmkelly/LessWrong-43k
LessWrong
New circumstances, new values? Crossposted at the Intelligent Agents Forum. Quick, is there anything wrong with a ten minute pleasant low-intensity conversation with someone we happen to disagree with? Our moral intuitions say no, as do our legal system and most philosophical or political ideals since the enlightenm...
e9b86d90-29e8-4c3d-ae39-e1fa5b002860
trentmkelly/LessWrong-43k
LessWrong
Why You Might Want a New Way to Visualize Biases A few days ago I made a post called A New Way to Visualize Biases, and the community seems to have been generally skeptical as to the value of the diagram. I don't think the original post made clear enough the types of use cases I was primarily interested in, and I know...
a31b42ea-a673-4303-b11b-f59e2ce44a6b
StampyAI/alignment-research-dataset/eaforum
Effective Altruism Forum
Is it crunch time yet? If so, who can help? Regarding AI safety, of course. The detailed version of this question follows: "Is AI safety sufficiently talent-constrained, and on an imminent enough timeline, that ~~me and any friends I can convince~~ ordinary non-Ivy-League non-top-tier computer scientists, mathemat...
745e823d-c7ab-4c08-b277-0d705a0593b5
trentmkelly/LessWrong-43k
LessWrong
Narcissistic Contrarianism The recent discussion on neo-reactionary-ism brought out some references to (intellectual hipsters and) meta-contrarianism linking to a 2010 posting by Yvain. For some time I've been thinking about "narcissistic contrarians" -- those who make an art form of their exotically counterintuitive...
07398c76-e469-4005-a51c-9a4cd9cc7e74
trentmkelly/LessWrong-43k
LessWrong
If ‘birth’ is worth nothing, births are worth anything It seems many people think creating a life has zero value. Some believe this because they think the average life contains about the same amount of suffering and satisfaction. Others have more conceptual objections, for instance to the notion that a person who does...
97d77a97-7126-4d10-8ca6-25fcd1c54c29
trentmkelly/LessWrong-43k
LessWrong
Matchmaking to Machine Alignment: Why Relational Design Can’t Be Left to Chance We say alignment is about control, safety, precision. But after a decade working as a matchmaker in India’s increasingly chaotic relationship market, I’ve learnt that what sustains a system isn’t control, it’s trust. And trust doesn’t live...
38dc0839-2a0f-4848-98bc-875d667a9559
trentmkelly/LessWrong-43k
LessWrong
The Stereotype of the Stereotype I want to define a new term that will be useful in the discourse. Imagine you landed in a foreign country with dogs, but with no word for dogs. You would see these things, and want to talk about them. You would like to be able to tell people whether they can bring dogs into your house...
e089595d-de32-4348-bd92-97fb9cb22e83
trentmkelly/LessWrong-43k
LessWrong
What books are for: a response to "Why books don't work." ...and a follow up to my recent post “Recommended Readings.” Introduction Andy Matuschak’s 2019 essay “Why books don’t work” has been on my mind for a while now. I highly recommend you give it a read. The piece’s central claim is that books fail at their fund...
a71c1c8b-9b6e-4fd5-afc2-ada5cc7d8985
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Making decisions using multiple worldviews *Tl;dr: the problem of how to make decisions using multiple (potentially incompatible) worldviews (which I'll call the problem of meta-rationality) comes up in a range of contexts, such as epistemic deference. Applying a policy-oriented approach to meta-rationality, and evalu...
51fa8853-8591-4a92-9349-fd2f201476de
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
Chris Olah’s views on AGI safety *Note: I am not Chris Olah. This post was the result of lots of back-and-forth with Chris, but everything here is my interpretation of what Chris believes, not necessarily what he actually believes. Chris also wanted me to emphasize that his thinking is informed by all of his colleague...
ec47f25c-e241-4566-83f0-9f0b033c9808
StampyAI/alignment-research-dataset/youtube
Youtube Transcripts
Opportunities for Cooperation on AI at Academic and Corporate Levels [Music] thank you for being here I'm really excited about this panel we're very proud that we got the largest panel at the conference and we also have a broad representation from people in China Russia Italy and the UK so the goal of this panel is to...
88d82cf6-7970-4e93-8ee0-601139ad236a
trentmkelly/LessWrong-43k
LessWrong
Reading group for Calculus by Spivak I am getting started reading Calculus by Spivak. (It is a rigorous textbook on Calculus that flirts with Real Analysis) I plan on moving fast, especially on the first few chapters which I read a few months back before refocusing on something else. Are there LWers who want to join i...
7bfae8cd-04bc-4904-a5fe-d11a99d7281f
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Deepmind has made a general inductor ("Making sense of sensory input") > Our system [the Apperception Engine] is able to produce interpretable human-readable causal theories from very small amounts of data, because of the strong inductive bias provided by the unity conditions. A causal theory produced by our system is...
87c98d02-63e2-49b3-ad5e-c3ddc2867308
trentmkelly/LessWrong-43k
LessWrong
What Does It Mean to Align AI With Human Values? The author has some weird misunderstandings about what AI-will-kill-everyone-ism advocates belive, but seems to have a weirdly[1] decent grasp of the problem, given their aforementioned misunderstandings. They argue IRL won't be enough[2]. Here's the interesting quote I...
c1d5e764-c159-432c-8e18-af1e1baa9b35
trentmkelly/LessWrong-43k
LessWrong
When I left Google About five years ago a friend had recently started at Wave, and encouraged me to join them. It sounded like a great startup doing really important work in East Africa, and potentially more valuable than continuing my "earning to give" approach, but because it was a fully remote company and I like be...
6e37f87d-71f0-4f80-b55b-215d41903a68
trentmkelly/LessWrong-43k
LessWrong
In Defense of Those Reclusive Authors Notable Authors of the 20th Century Who Were Introverted Writing  fiction is usually a solitary profession. Among those individuals who  end up producing art to express themselves, one can logically assume  that there will be many who like to keep their distance from society.  Whi...
9867b03d-becd-40b5-8f88-9bbc11514de9
trentmkelly/LessWrong-43k
LessWrong
Announcement: LessWrong Coronavirus Links Database 2.0 There is a massive amount of information and discussion about the coronavirus, and it can be challenging to keep track of all the things coming in from different sources. To combat that, 11 days ago we announced that the LessWrong team built and is maintaining a ...
505faa63-2cfd-4af9-bddc-7c719a9c8222
trentmkelly/LessWrong-43k
LessWrong
Meetup : Vancouver Minicamp Report-Back Discussion article for the meetup : Vancouver Minicamp Report-Back WHEN: 18 May 2012 05:00:00PM (-0700) WHERE: Vancouver Minicamp was totally awesome and I learned so much. I think I can put it best like this: I no longer have "a sense that more is possible" it is now "More i...
3211eaf5-caa4-4336-a865-eb786456f820
StampyAI/alignment-research-dataset/eaforum
Effective Altruism Forum
Machine Learning for Scientific Discovery - AI Safety Camp ### Applications for [AI Safety Camp](https://aisafety.camp/) 2023 have opened! I'll be Research Lead for the project [Machine learning for scientific discovery: the present and future of science-producing AI models](https://docs.google.com/document/d/1J6aad...
0e9ba4a6-6e4e-4985-a487-aac3014fb67e
trentmkelly/LessWrong-43k
LessWrong
HOWTO: Screw Up The LessWrong Survey and Bring Great Shame To Your Family