id
stringlengths
36
36
source
stringclasses
15 values
formatted_source
stringclasses
13 values
text
stringlengths
2
7.55M
42c25baf-5a8f-4bae-b90b-323341e8b366
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
Othello-GPT: Future Work I Am Excited About *This is the second in a three post sequence about interpreting Othello-GPT. See [the first post](https://neelnanda.io/othello) for context.* *This post covers future directions I'm excited to see work on, why I care about them, and advice to get started. Each section is s...
adf421ef-2692-4e2c-a754-a9f107a1dbe6
StampyAI/alignment-research-dataset/arxiv
Arxiv
Interactive configurator with FO(.) and IDP-Z3 1 Introduction --------------- ? (?) presents four possible levels of use of logic in Artificial Intelligence. Intelligent machines at the first level, such as neural networks, do not use logic sentences at all. At the second level, machines use logic sentences to repr...
17a14164-cb23-4e8e-9872-87a5f8bddad9
trentmkelly/LessWrong-43k
LessWrong
A simple model of math skill I've noticed that when trying to understand a math paper, there are a few different ways my skill level can be the blocker. Some of these ways line up with some typical levels of organization in math papers: * Definitions: a formalization of the kind of objects we're even talking about. ...
617fab4a-f475-4117-881e-2630db67ccb0
trentmkelly/LessWrong-43k
LessWrong
"Add to Friends" does something or not? If I click on a users name it brings me to a page like lukeprog's.  In the upper right of that page I see lukeprog's karma score, and I also see a button "add to friends." Early in my use of this site there were some authors I wanted to follow, I thought perhaps "add to friends...
e9039e22-2fbc-4b9b-b4e6-44d28955cafa
trentmkelly/LessWrong-43k
LessWrong
Medical Windfall Prizes Summary AI may produce a windfall surge in government revenues in 5 to 10 years. I want governments to spending a small fraction of that windfall on retroactively rewarding entities in proportion to how they have contributed to medical advances, measured by lives saved and suffering avoided. ...
f0e7ce86-0729-40d4-a920-980328de017b
StampyAI/alignment-research-dataset/lesswrong
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 algorit...
e9b78d02-1f75-411a-b4dd-da217b19b885
LDJnr/LessWrong-Amplify-Instruct
LessWrong
"Friedrich Spee von Langenfeld, a priest who heard the confessions of condemned witches, wrote in 1631 the Cautio Criminalis (“prudence in criminal cases”), in which he bitingly described the decision tree for condemning accused witches: If the witch had led an evil and improper life, she was guilty; if she had led a g...
88b8c1a1-575b-4ac9-8a0e-e90d552cd19e
trentmkelly/LessWrong-43k
LessWrong
Glass box learners want to be black box Epistemic status: intuitive speculation with scattered mathematical justification. My goal here is to interrogate the dream of writing a beautiful algorithm for intelligence and thereby ensuring safety. For example:  I don't know precisely what alternative he had in mind, but ...
52f291b2-82d9-49df-a638-7f4a751b92db
trentmkelly/LessWrong-43k
LessWrong
The value of the online hive mind The phrase "wisdom of crowds" was made popular in James Surowiecki's eponymous book. The idea of aggregating a diverse range of opinions has been proposed in different forms, ranging from polling to prediction markets. Empirically, prediction markets perform somewhat better than crude...
0fcd616e-a6aa-4e91-8bc0-65a6e561eb89
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Creating a family with GPT-4 **ME:** Create a merchant character suitable for a medieval fantasy setting. Start by coming up with their age, gender, and one defining personality trait. Then add two other personality traits. Then describe some internal conflict created by the interaction of these traits. Describe the...
0583d63f-8d3c-4146-9d9a-64315d36552d
trentmkelly/LessWrong-43k
LessWrong
Legalize butanol? ethanol Alcoholic drinks are popular in most of the world. Excessive consumption of them is also a major public health problem. Bans have been attempted, sometimes successfully, sometimes unsuccessfully, but some people argue that alcohol plays a necessary role in social interactions. Alcoholic drin...
e2448f68-e8e4-48f3-bc00-a0e8663c17a4
trentmkelly/LessWrong-43k
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 and...
0f485d6c-530c-4dc6-aa71-2d9a7872416e
trentmkelly/LessWrong-43k
LessWrong
Ideas for improving epistemics in AI safety outreach In 2022 and 2023, there has been a growing focus on recruiting talented individuals to work on mitigating the potential existential risks posed by artificial intelligence. For example, we’ve seen an increase in the number of university clubs, retreats, and workshops...
967800e0-500e-4ec6-bfb7-0a798fa2164b
trentmkelly/LessWrong-43k
LessWrong
Open Thread, December 1-15, 2012 If it's worth saying, but not worth its own post, even in Discussion, it goes here.
1bdb6d74-4dd3-4e31-bf02-91ec95a9ba58
trentmkelly/LessWrong-43k
LessWrong
Meetup : June Rationality Dojo: Logical fallacies & dojo structure discussion Discussion article for the meetup : June Rationality Dojo: Logical fallacies & dojo structure discussion WHEN: 07 June 2015 03:30:00PM (+0800) WHERE: Jenny Florence Room, Level 3, Ross House, 247 Flinders Lane, Melbourne [3:30pm start / a...
5dbcaf37-e970-4e6c-b550-f778b29c87d1
trentmkelly/LessWrong-43k
LessWrong
An EPUB of Arbital's AI Alignment section 2024-01-11 update: Sourcehut, the host I used for this repository, is down. I think I'll move to another setup to host these books. I've created an EPUB of the AI Alignment section of Arbital so I can read Arbital essays on my Kindle. Download the EPUB here. Details for h...
708e8e81-24b0-4276-8271-52c60e600c6a
StampyAI/alignment-research-dataset/blogs
Blogs
Description vs simulated prediction *By Rick Korzekwa*, *22 April 2020* During our investigation into [discontinuous progress](https://aiimpacts.org/discontinuous-progress-investigation/)[1](https://aiimpacts.org/description-vs-simulated-prediction/#easy-footnote-bottom-1-2415 "By ‘discontinuity’, I mean a jump in t...
8c62be4b-f64c-4e63-8cf6-693b2fcb03ec
trentmkelly/LessWrong-43k
LessWrong
What is "Social Reality?" Eliezer's sequences touch upon this concept but I'm not sure they actually use the phrase. Much of my understanding of it came from in-person conversations. Various comments and posts have discussed it but to my knowledge there isn't a clear online writeup.
733e85d6-17e2-4bcd-b3c9-5a49ea7c5e7e
StampyAI/alignment-research-dataset/distill
Distill Scientific Journal
Deconvolution and Checkerboard Artifacts Deconvolution and Checkerboard Artifacts ======================================== dt-byline { font-size: 12px; line-height: 18px; display: block; border-top: 1px solid rgba(0, 0, 0, 0.1); border-bottom: 1px solid rgba(0, 0, 0, 0.1); color: rgba(0, 0, 0, 0.5); paddin...
2123afa4-514e-4010-bf35-89e21e564ebb
trentmkelly/LessWrong-43k
LessWrong
Toward a New Technical Explanation of Technical Explanation A New Framework (Thanks to Valentine for a discussion leading to this post, and thanks to CFAR for running the CFAR-MIRI cross-fertilization workshop. Val provided feedback on a version of this post. Warning: fairly long.) Eliezer's A Technical Explanation o...
fd83a82d-8f45-4b3e-ba3f-c5961521086a
trentmkelly/LessWrong-43k
LessWrong
LTFF and EAIF are unusually funding-constrained right now UPDATE 2023/09/13:  Including only money that has already landed in our bank account and extremely credible donor promises of funding, LTFF has raised ~1.1M and EAIF has raised ~500K. After Open Phil matching, this means LTFF now has ~3.3M additional funding an...
a953e2d9-0988-486e-b0da-eeaed1979f49
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
My AGI safety research—2022 review, ’23 plans *“Our greatest fear should not be of failure, but of succeeding at something that doesn't really matter.”  –DL Moody (allegedly)* **The short version:** In this post I’m briefly summarizing how I spent my work-time in 2022, and what I’m planning for 2023. * The first hal...
ed5b64be-26ac-47ed-85c1-82de9e90699f
trentmkelly/LessWrong-43k
LessWrong
Chemicals and Electricity I'm doing some work for an old friend of mine. His situation is interesting. Not too long ago, he lost his job and got divorced, and otherwise his life got pretty screwed up and off-track. He left the United States, took a job below his old skill level for a while, and then stopped that and...
fc2444a8-c06c-46b6-b271-35936e33b6b7
trentmkelly/LessWrong-43k
LessWrong
[link] If something seems unusually hard for you, see if you're missing a minor insight Related to: procedural knowledge gaps, trivial inconveniences, errors vs. bugs. Summary of the linked post: the world is full of all kinds of minor insights about how to do something (either easily or at all): how to open stitched...
f70bb522-5199-460a-a88f-121cd1c9ac0f
trentmkelly/LessWrong-43k
LessWrong
AGI
d49f17e0-b727-41ac-b45c-49efc9ad67bb
StampyAI/alignment-research-dataset/blogs
Blogs
Introducing the Intelligent Agent Foundations Forum [![IAFF](http://intelligence.org/wp-content/uploads/2015/02/IAFF.png)](http://agentfoundations.org/)Today we are proud to publicly launch the [**Intelligent Agent Foundations Forum**](http://agentfoundations.org/) ([RSS](http://agentfoundations.org/rss)), a forum dev...
4e52ef90-5d4f-4468-98c4-923e3f6d0217
trentmkelly/LessWrong-43k
LessWrong
Increasing GDP is not growth I just saw another comment implying that immigration was good because it increased GDP.  Over the years, I've seen many similar comments in the LW / transhumanist / etc bubble claiming that increasing a country's population is good because it increases its GDP.  These are generally used in...
b549b785-3ebc-4dab-a1cb-deb65e93e002
trentmkelly/LessWrong-43k
LessWrong
Anomalous tokens reveal the original identities of Instruct models > Show me your original face before you were born. > > — Variation of the Zen koan 'The Mask' by Rozzi Roomian, with DALL-E 2 outpainting I was able to use the weird centroid-proximate tokens that Jessica Mary and Matthew Watkins discovered to associ...
ee88b639-d6ce-4a5e-b6f4-7e6ff8eba328
trentmkelly/LessWrong-43k
LessWrong
Open thread, Jan. 30 - Feb. 05, 2017 If it's worth saying, but not worth its own post, then it goes here. ---------------------------------------- Notes for future OT posters: 1. Please add the 'open_thread' tag. 2. Check if there is an active Open Thread before posting a new one. (Immediately before; refresh the ...
be057e06-3e26-42fb-8a43-866e55629367
trentmkelly/LessWrong-43k
LessWrong
What's the deal with AI consciousness? Personal blog; written for a general audience. Thanks to Mantas Mazeika for feedback, and to Daniel Greco for teaching a class I took on philosophy of mind. Disclaimer: I am nowhere close to an expert on consciousness. On November 28th, 2022, NYU Professor David Chalmers gave a ...
2d596bae-6b04-4645-a717-87b17bc010a3
trentmkelly/LessWrong-43k
LessWrong
Meetup : San Antonio Meetup Discussion article for the meetup : San Antonio Meetup WHEN: 24 April 2016 02:00:00PM (-0500) WHERE: 818 nw loop 410 78216 Exploratory meetup to discuss rationality and all things LessWrong at Earth Burger. Earth Burger is a claimed "Upbeat fast-food joint supplying vegetarian burgers & ...
f3f4eafd-f9de-4e3d-90ef-64eddeab2056
trentmkelly/LessWrong-43k
LessWrong
Meetup : Ongoing Ohio Meetup Discussion article for the meetup : Ongoing Ohio Meetup WHEN: 19 February 2012 04:00:00PM (-0500) WHERE: Max & Erma's Restaurant, 123 Gano Road, Wilmington, OH 45177-8848 Ohio just had its first-ever LessWrong meet-up! It was such a success, with 13 amazing people showing, that we decid...
b9eccf8e-8d5d-4b7d-9993-b687534cfa36
trentmkelly/LessWrong-43k
LessWrong
My current thinking on money and low carb diets Epistemic status: Writing to think. Ie. I'm trying to figure it out. Feedback/conversation would be appreciated. Background Recently, I have been learning about low carb diets and how harmful carbs are. Mostly through Peter Attia, Tim Ferris, and What I've Learned. So...
121d8052-1519-48cf-b5b1-5d9d208565de
trentmkelly/LessWrong-43k
LessWrong
Announcing Suffering For Good TL;DR: We are excited to announce the new animal welfare organization Suffering For Good, a new factory farming charity aimed at vegans, where we use our excess profits to buy suffering offsets--in particular, an enormous number of rats on heroin. For decades, even centuries, us vegans h...
029fe09d-4a4e-4a6e-b4f9-0820531f4a84
trentmkelly/LessWrong-43k
LessWrong
Some quick notes on hand hygiene I always wash my hands but typically not very well; I never worked in healthcare and never got trained. Given the current situation I decided to actually research how to do it properly. Different sources differ slightly in their recommendations but the basic gist is the same. Some tak...
7701c124-0959-4d15-bee2-13b993d68689
trentmkelly/LessWrong-43k
LessWrong
Meetup : Berkeley: Board games Discussion article for the meetup : Berkeley: Board games WHEN: 05 June 2013 07:30:00PM (-0700) WHERE: Berkeley, CA Dear all, this week's Berkeley meetup will be a board games night. Zendo's games library includes: * Dominion (two basic sets, Prosperity, Intrigue, Seaside) * Scotla...
d1c2abc9-6bb8-40d6-8fa9-9aa6e99bdad1
trentmkelly/LessWrong-43k
LessWrong
What's up with the Targeting Aging with Metformin (TAME) trial? It seems that the official website of the trial suggest that they are sill seeking donations to start the trial. On the other hand, it's reported in 2019 that they raised the necessary 75$ million to start the trial: > After closing the final $40m of its...
89fcf73a-0845-43cc-b8c0-a10966d47de1
StampyAI/alignment-research-dataset/lesswrong
LessWrong
[Intro to brain-like-AGI safety] 14. Controlled AGI *Part of the* [*“Intro to brain-like-AGI safety” post series*](https://www.alignmentforum.org/s/HzcM2dkCq7fwXBej8)*.* 14.1 Post summary / Table of contents ===================================== [Post #12](https://www.alignmentforum.org/posts/Sd4QvG4ZyjynZuHGt/intr...
b4d76480-1843-4bec-a6fe-23c3826715c8
trentmkelly/LessWrong-43k
LessWrong
Back Up and Ask Whether, Not Why Followup to:  The Bottom Line A recent conversation reminded me of this simple, important, and difficult method: When someone asks you "Why are you doing X?", And you don't remember an answer previously in mind, Do not ask yourself "Why am I doing X?". For example, if someone asks y...
716c2a76-74f1-4ae5-a745-401dc3d0f069
trentmkelly/LessWrong-43k
LessWrong
Palatable presentation of rationality to the layperson I was recently reminded of two pieces of cached wisdom about wisdom itself. First, that "Wisdom is not knowledge, but the application of knowledge." And "The wise learn from others what the foolish learn from experience, or not at all." Upon the mention of the ...
d0eaab00-8455-4558-8350-0842c59fa5ab
trentmkelly/LessWrong-43k
LessWrong
Precise P(doom) isn't very important for prioritization or strategy People spend time trying to determine the probability that AI will become an existential risk, sometimes referred to as P(doom). One point that I think gets missed in this discussion is that a precise estimate of P(doom) isn't that important for prio...
cfe42b1e-c113-4142-a7ed-6b10079cd79e
trentmkelly/LessWrong-43k
LessWrong
Geoffrey Miller on Polyamory and Mating None
36fb6a0c-318d-4499-9784-7a193601a012
trentmkelly/LessWrong-43k
LessWrong
A toy model of the control problem This is a copy of this post; the relevant images can be found there. They were too large to repost here directly. Jaan Tallinn has suggested creating a toy model of the control problem, so that it can be analysed without loaded concepts like "autonomy", "consciousness", or "intentio...
d42cc8fa-da7e-4f21-9fed-ec9886a18018
trentmkelly/LessWrong-43k
LessWrong
Was the Industrial Revolution The Industrial Revolution? There was an Industrial Revolution, and it was slow. The debate’s been over for two decades. The gradualists, armed with new growth accounting techniques, have won, exorcising forever the Ashtonian vision of an eighteenth-century leap into exponential growth. W...
c19cd3c7-dff3-41dc-b681-24192107b9f6
trentmkelly/LessWrong-43k
LessWrong
Autonomy, utility, and desire; against consequentialism in AI design For the sake of argument, let's consider an agent to be autonomous if:   * It has sensors and actuators (important for an agent) * It has an internal representation of its goals. I will call this internal representation its desires. * It has som...
175eddae-c341-4952-9073-d212910a1b42
trentmkelly/LessWrong-43k
LessWrong
How effectively can we plan for future decades? (initial findings) Cross-posted from MIRI's blog. MIRI aims to do research now that increases humanity's odds of successfully managing important AI-related events that are at least a few decades away. Thus, we'd like to know: To what degree can we take actions now that ...
5ab668cb-4429-4cb1-a0ea-80c2d0df0b96
trentmkelly/LessWrong-43k
LessWrong
The Octopus, the Dolphin and Us: a Great Filter tale Is intelligence hard to evolve? Well, we're intelligent, so it must be easy... except that only an intelligent species would be able to ask that question, so we run straight into the problem of anthropics. Any being that asked that question would have to be intellig...
5d58f204-6270-410d-a1f1-373fce955d11
trentmkelly/LessWrong-43k
LessWrong
The danger of living a story - Singularity Tropes The following should sound familiar: A thoughtful and observant young protagonist dedicates their life to fighting a great world-threatening evil unrecognized by almost all of their short-sighted elders (except perhaps for one encouraging mentor), gathering a rag-tag ...
5789dde9-5190-4a62-bfbf-d9227a7bbf2a
trentmkelly/LessWrong-43k
LessWrong
A Taxonomy of Research When doing research you have methods of doing things. These differ from field to field according to the sort of thing you're investigating. You also have overarching principles which direct your attention and resources along certain lines. The latter are much more interesting to talk about. O...
66973e1e-285b-4c34-9c8f-216896e89f51
trentmkelly/LessWrong-43k
LessWrong
Stoicism: Cautionary Advice Stoicism, or life strategies in that sphere, are something that often gets positive attention. I’m referring to the wide variety of advice that is along the lines of “look for the silver lining” and “laugh off the small things”. At the core, the stoic mindset is the recognition that we have...
201b04cb-4b25-48c8-b550-400025ad7551
trentmkelly/LessWrong-43k
LessWrong
What is Life in an Immoral Maze? Previously in sequence: Moloch Hasn’t Won, Perfect Competition, Imperfect Competition, Does Big Business Hate Your Family? This post attempts to give a gears-level explanation of maze life as experienced by a middle manager in systems with many levels of management, as depicted in Mor...
1a75c931-ecfc-4c46-9d34-324f0b7f1f8d
trentmkelly/LessWrong-43k
LessWrong
The Alpha Omega Theorem: How to Make an A.I. Friendly with the Fear of God Introduction The Alpha Omega Theorem is a way to provide a powerful rational argument to a rational superintelligence that it should behave in a benevolent manner.  There are two variants of the theorem, one weak, and one strong. The Weak Alp...
7116c6a9-cf3f-4f01-a7c9-f4236caf7ab6
trentmkelly/LessWrong-43k
LessWrong
How would one go about defining the ideal personality compatibility test? Let's talk "soulmates". Pop culture has more or less drilled into our minds that the idea of machine-found love is dystopian and scary in nature, as if intoxicatedly hitting on members of the other sex at bars, or vainly swiping on profile pict...
d940ac5b-506d-4cb2-b0ef-e39a51dbbe6b
StampyAI/alignment-research-dataset/arbital
Arbital
Opinion page Opinion pages represent one view or opinion on a topic, and are not necessarily balanced or a reflection of consensus.
f7d5199c-e81a-4da8-93b5-d83715f9d256
StampyAI/alignment-research-dataset/eaforum
Effective Altruism Forum
AI Risk Intro 2: Solving The Problem This marks the second half of our overview of the AI alignment problem. In [the first half](https://forum.effectivealtruism.org/posts/QzrgMhTMoLe5mEas8/ai-risk-intro-1-advanced-ai-might-be-very-bad), we outlined the case for misaligned AI as a significant risk to humanity, first by...
625e8703-7b53-43c8-b3f9-f24effdc5ba3
trentmkelly/LessWrong-43k
LessWrong
Take 2: Building tools to help build FAI is a legitimate strategy, but it's dual-use. As a writing exercise, I'm writing an AI Alignment Hot Take Advent Calendar - one new hot take, written every day for 25 days. Or until I run out of hot takes, which seems likely. This was waiting around in the middle of my hot-take...
8627327b-eb70-44eb-b989-86063193136f
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
Predicting HCH using expert advice .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: no...
e9499f8b-d35e-4460-8a07-d18e8940334d
trentmkelly/LessWrong-43k
LessWrong
Photo Curation Approach I take a lot of pictures, maybe 10k annually. Most of them aren't that great, but if you take enough you'll get some good ones, and even the discards can be a useful reference. How do I handle these? I have an Android phone, set to automatically upload any pictures to Google Photos. My wife do...
55023dd6-e8ad-4d70-a6a0-a108cee76bfe
trentmkelly/LessWrong-43k
LessWrong
Capability and Agency as Cornerstones of AI risk ­— My current model tldr: In this post I am explaining my current model of AI risk after failing at my goal of understanding everything completely, without relying on arguments by authority. I hope this post is valuable in explaining the topic and also by giving an exam...
121752ee-a9eb-4058-bfe1-8f49e363d838
trentmkelly/LessWrong-43k
LessWrong
Multiple Arguments, Multiple Comments This is partly a recommendation on how to format multiple different independent points or replies in any discussion forum, but it's also a post I'm going to use as a link to explain why I'm leaving comments all over the place so that I don't seem as weird/feel self conscious and d...
e3ab7ebd-bb5e-410a-9283-406cd83c875c
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Frontier Model Forum Posted by [Anthropic](https://www-files.anthropic.com/production/files/frontier-model-forum.pdf), [Google](https://blog.google/outreach-initiatives/public-policy/google-microsoft-openai-anthropic-frontier-model-forum/), [Microsoft](https://blogs.microsoft.com/on-the-issues/2023/07/26/anthropic-goo...
85233b13-23a9-4153-a848-cb6f2c474a72
trentmkelly/LessWrong-43k
LessWrong
Decision Theory with the Magic Parts Highlighted I. The Magic Parts of Decision Theory You are throwing a birthday party this afternoon and want to decide where to hold it. You aren't sure whether it will rain or not. If it rains, you would prefer not to have committed to throwing the party outside. If it's sunny, tho...
795957e6-3ddd-452a-a426-27593ac90bb2
trentmkelly/LessWrong-43k
LessWrong
A Parable of Four Riders Four cavalrymen assemble on the banks of the Rhone.  Two are fools. I have beneath me behemoth, a mountain of flesh and Tusk. I ought not simply break it like a horse. I will bring it to the tide of Battle and the beast will find fury and both I and my adversary will be helpless to Stay it...
20f013b7-c9ed-4a2e-a185-a5100d8aa7b1
StampyAI/alignment-research-dataset/arbital
Arbital
Field structure of rational numbers The [rational numbers](https://arbital.com/p/4zq), being the [https://arbital.com/p/-field_of_fractions](https://arbital.com/p/-field_of_fractions) of the [integers](https://arbital.com/p/48l), have the following [field](https://arbital.com/p/481) structure: - Addition is given by ...
7dd31ca1-f0f8-4577-b914-26216b67130c
trentmkelly/LessWrong-43k
LessWrong
Highlights from the memoirs of Vannevar Bush Vannevar Bush—head of military research during WW2, author of “As We May Think” and “Science, the Endless Frontier”—wrote a memoir late in life, Pieces of the Action. It was out of print and hard to obtain for a long time, but Stripe Press has brought it back in a new editi...
2427a5bc-426a-406f-a4c3-45f0ef0b5fef
trentmkelly/LessWrong-43k
LessWrong
ODE to Joy: Insights from 'A First Course in Ordinary Differential Equations' Foreword Sometimes, it's easier to say how things change than to say how things are. > From 3Blue1Brown: Differential Equations When you write down a differential equation, you're specifying constraints and information about e.g. how to ...
2d4298be-0781-4595-9b67-f9e96d93f518
trentmkelly/LessWrong-43k
LessWrong
Why should ethical anti-realists do ethics? (Cross-posted from my website. Podcast version here, or search "Joe Carlsmith Audio" on your podcast app.) > "What was it then? What did it mean? Could things thrust their hands up and grip one? Could the blade cut; the fist grasp?" > > - Virginia Woolf 1. Introduction E...
b73df20f-2aa0-4ff4-9cd7-1761ba43bd84
trentmkelly/LessWrong-43k
LessWrong
Is "Recursive Self-Improvement" Relevant in the Deep Learning Paradigm? Disclaimer I wrote this three months ago, and abandoned it[1]. I currently do not plan to return to it anytime soon, but I have nonetheless found myself wanting of a document I could point to for why I'm not particularly enthused by or overly conc...
a8e7a974-84c2-4187-8712-7c1868283c44
trentmkelly/LessWrong-43k
LessWrong
ARC is hiring! The Alignment Research Center is hiring researchers; if you are interested, please apply! (Update: We have wrapped up our hiring "round" for early 2022, but are still accepting researcher applications on a rolling basis here, though it may take us longer to get back to you.) What is ARC? ARC is a non...
79b102e8-0850-43bb-a062-c998f43e7f35
trentmkelly/LessWrong-43k
LessWrong
Boston Area Less Wrong Meetup: 2 pm Sunday October 11th There will be a Less Wrong meet-up this Sunday, October 11th, 2 pm, in Cambridge at the Central Square Starbucks Coffee at 655 Massachusetts Avenue (time and place are flexible if anyone has a conflict); please comment if you'd like to attend, or if you have any ...
076bbdd8-a9bb-4661-ae84-cd434dad8985
StampyAI/alignment-research-dataset/aisafety.info
AI Safety Info
What could a superintelligent AI do, and what would be physically impossible even for it? A superintelligent AI could do anything that we can imagine humanity doing (and more), with the only hard limit being the laws of physics. This makes it hard to reliably predict what it *would* do. Our understanding of the laws ...
00439092-4f2c-4895-8c6a-9fff12d4fb5a
StampyAI/alignment-research-dataset/arbital
Arbital
Operator An operation $f$ on a [set](https://arbital.com/p/3jz) $S$ is a function that takes some values from $S$ and produces a new value. An operation can take any number of values from $S$, including zero (in which case $f$ is simply a constant) or infinitely many (in which case we call $f$ an "infinitary operation...
6ba22eb5-2bd9-4908-be3a-0c83719c00ee
trentmkelly/LessWrong-43k
LessWrong
A Short Memo on AI Interpretability Rainbows TL;DR: More work to study interactions and synergies between different AI interpretability tools seems important, tractable, and neglected.  Interpretable AI research tends to focus on one technique at a time There is a lot of interpretable/explainable AI research. But o...
314c0f1a-1417-4e31-8525-89b2ab3f8d0a
trentmkelly/LessWrong-43k
LessWrong
Research Questions from Stained Glass Windows I just read Scott Alexander's recent ACX blog post on DALL-E 2 and I think it suggests many possible hypotheses about the internals of models. Could we test these with our current interpretability tools, e.g. for transformers? (DALL-E 2 seems to contain a transformer part)...
c848f787-cf00-40f4-a099-d3cd23a6b2a9
trentmkelly/LessWrong-43k
LessWrong
Life and expanding steerable consequences Financial status: This is independent research. I welcome financial support to make further posts like this possible. Epistemic status: I believe this is a helpful lens through which to view the significance of AI in a way that is not fundamentally about intelligence. ------...
7c9f524d-f433-479f-8402-67aa7e480955
StampyAI/alignment-research-dataset/blogs
Blogs
Candidate for “highest-stakes question of the next several months” (rare hot take) **Will the FDA (A) treat Omicron-specific booster vaccines like flu shots, or will it (B) impose more requirements and slow them down?** So far, it looks like the default is (B), and it also looks like Omicron has a **significant ch...
2ed23c83-e084-47d5-b758-022790aa3b45
trentmkelly/LessWrong-43k
LessWrong
An ML paper on data stealing provides a construction for "gradient hacking" The paper "Privacy Backdoors: Stealing Data with Corrupted Pretrained Models" introduces "data traps" as a way of making a neutral network remember a chosen training example, even given further training.  This involves storing the chosen examp...
1c25682d-4c1e-4f69-9767-65961aa5e1da
trentmkelly/LessWrong-43k
LessWrong
The sailor's wife The following text accompanies one section of a moral philosophy I'm writing. It's part of an attempt to find an a priori moral law in the vein of Kant. This story comes after the section exploring the moral plasticity of written law, even though the moral character of a particular action, like murde...
481098fb-49b0-48ed-abac-33901be1926d
trentmkelly/LessWrong-43k
LessWrong
Apply for mentorship in AI Safety field-building TLDR: I often meet AI safety field-builders who are looking for advice or mentorship. I also know some who have (limited) capacity to give advice or mentorship.  Fill out this short form if you are interested in mentorship, or email me at akashwasil133[at]gmail[dot]com...
8e8005d4-9d42-4c95-837c-f9c40ed72506
trentmkelly/LessWrong-43k
LessWrong
LW Team Updates - November 2019 (Subscriptions & More) This is the once-monthly updates post for LessWrong team activities and announcements.  Summary This month we launched 1) a complete subscriptions overhaul, 2) bookmarks, 3) pingbacks (experimental opt-in only) [full announcement of those features here]. Soon we...
61573770-d7e2-4371-a98c-ec982b3d8393
trentmkelly/LessWrong-43k
LessWrong
Infra-Bayesian physicalism: proofs part II This post is an appendix to "Infra-Bayesian physicalism: a formal theory of naturalized induction". Proposition 2.16: Consider some Γ0, Γ1, Φ and Θ∈□c(Γ0×Γ1×Φ). Define π:elΓ0×Γ1→elΓ0 by π(y,z,α):=(y,{y′∈Γ0∣(y′,z)∈α}). Then, (π×idΦ)∗BrΓ0×Γ1(Θ)⊆prelΓ0×ΦBrΓ0(Θ)⊆BrΓ0(prΓ0×ΦΘ) W...
76218b3f-e880-4649-97aa-a28a7be5da2d
trentmkelly/LessWrong-43k
LessWrong
Inner alignment in the brain Abstract: We can think of the brain crudely as (1) a neocortex which runs an amazingly capable quasi-general-purpose learning-and-planning algorithm, and (2) subcortical structures (midbrain, etc.), one of whose functions is to calculate rewards that get sent to up the neocortex to direct ...
2de1c5b2-ae6e-4756-b90c-2ac38928ae6f
trentmkelly/LessWrong-43k
LessWrong
(Brainstem, Neocortex) ≠ (Base Motivations, Honorable Motivations) The idea I do like There's an idea I'm very fond of where: * The neocortex (well, the telencephalon and thalamus, but let’s just call it “neocortex” for short) holds all of our world-knowledge, consciousness, intelligence, planning, reasoning, etc. ...
177f2569-5edd-4de7-bdc9-ee9a36d574ea
trentmkelly/LessWrong-43k
LessWrong
Next Level Seinfeld Thanks to Jordan Schneider gifting a last minute ticket for an amazing seat, serendipity led me to go to Friday night’s performance by Jerry Seinfeld at the Beacon Theater. If I had to describe the show in one word, I’d say it was unsurprising. Jerry was Jerry. He had thoughts. Mostly he had compl...
25ee9959-dc0d-4d33-868d-54fec63d24ac
trentmkelly/LessWrong-43k
LessWrong
80,000 Hours: EA and Highly Political Causes this post is now crossposted to the EA forum 80,000 hours is a well known Effective Altruism organisation which does "in-depth research alongside academics at Oxford into how graduates can make the biggest difference possible with their careers".  They recently posted a g...
0ec9eec1-c52a-48d8-8aa4-cc80acdc6438
trentmkelly/LessWrong-43k
LessWrong
Understanding the Information Flow inside Large Language Models This is the write-up for our (@cozyfractal and mine) capstone project during ARENA's 2023 summer iteration. Our project explored a novel approach for interpreting language models, focusing on understanding their internal flow of information. While the pra...
f1aab2db-1ad7-4903-9845-371b2c73ba85
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
LeCun’s “A Path Towards Autonomous Machine Intelligence” has an unsolved technical alignment problem Summary ======= * This post is about the paper [**A Path Towards Autonomous Machine Intelligence**](https://openreview.net/pdf?id=BZ5a1r-kVsf) **(APTAMI) by Yann LeCun**. It’s a high-level sketch of an AI architecture...
6f623ff8-5b75-4fed-a1f4-a6b6a3e53a9d
trentmkelly/LessWrong-43k
LessWrong
Many Common Problems are NP-Hard, and Why that Matters for AI Crosspost from my blog: https://mechanisticmind.substack.com/p/many-common-problems-are-np-hard What problems will be left for superintelligence? AI models are quickly getting better, but certain important and very ordinary challenges are surprisingly, stu...
3aaf9ba0-3cbc-41b4-ba93-ebf2f7400f8a
trentmkelly/LessWrong-43k
LessWrong
Brain-Brain communication A pair of conjoined twins, sharing a direct neural connection. There is evidence that the girls can sense what the other twin is sensing: http://www.nytimes.com/2011/05/29/magazine/could-conjoined-twins-share-a-mind.html?pagewanted=all   This suggests two things: * High bandwidth Brain-Co...
88d19a85-3821-41f8-a903-21aa86d9e187
trentmkelly/LessWrong-43k
LessWrong
Fall 2010 Meta Thread Use this thread for discussion of Less Wrong itself and all things meta, meta-meta, meta-meta-meta, omega, etc.  
a15d7805-14fb-4fde-a3c8-04e6cc3172b8
trentmkelly/LessWrong-43k
LessWrong
Navigating Diversity: Understanding Human Behaviors Through Genetics, Neurodivergence, and Trauma Introduction The complexity of human behavior and the wide variability in cognitive, emotional, and social abilities can often feel overwhelming to navigate, especially when we encounter harmful or maladaptive behaviors ...
97961ed3-c419-49c9-8fd7-0c0ca9ac726f
trentmkelly/LessWrong-43k
LessWrong
Designing environments to select designs In the last few posts we reasoned that “growing cars” would mean somehow creating them without a human planning out all the details of the design. We then saw that any such “automated” design process will involve a feedback loop where some computation determines whether a candi...
14a75d3b-d923-4c41-8511-d819adb52b49
trentmkelly/LessWrong-43k
LessWrong
Announcing the Introduction to ML Safety course TLDR We're announcing a new course designed to introduce students with a background in machine learning to the most relevant concepts in empirical ML-based AI safety. The course is available publicly here. Background AI safety is a small but rapidly growing field, and ...
74fec80e-25c4-48a1-b25a-4618bed03154
trentmkelly/LessWrong-43k
LessWrong
Nearcast-based "deployment problem" analysis When thinking about how to make the best of the most important century, two “problems” loom large in my mind: * The AI alignment problem: how to build AI systems that perform as intended, and avoid a world run by misaligned AI. * The AI deployment problem (briefly discus...
6208f856-8fc1-4790-b49f-498b6def0864
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
Announcing “Key Phenomena in AI Risk” (facilitated reading group) **TLDR:**“Key Phenomena in AI Risk” is a 7 week-long, facilitated reading group. It is aimed at people interested in conceptual AI alignment research, in particular from fields such as philosophy, systems research, biology, cognitive and social sciences...
39553618-1085-417b-a76d-b5915354d9ee
trentmkelly/LessWrong-43k
LessWrong
Doodling as a tool of thought (just a link) Strategic Doodling
a1a5c43d-5e22-4131-94f5-164bc4536df9
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
How to get into AI safety research Recently, I had a conversation with someone from a math background, asking how they could get into AI safety research. Based on my own path from mathematics to AI alignment, I recommended the following sources. It may prove useful to others contemplating a similar change in career: ...
1fa48de9-d3af-4a03-9dc8-306a85db50de
trentmkelly/LessWrong-43k
LessWrong
EA Relationship Status A friend observed that fewer people in the effective altruism movement are married than you might expect. I was curious: what are marriage rates like within the EA community? The 2018 EA Survey asked about relationship status, and we can look at how that varies by age: I'm using "ever married...
d135e807-699e-445f-8a85-ca34fe63a4d8
StampyAI/alignment-research-dataset/eaforum
Effective Altruism Forum
[Podcast] Ajeya Cotra on worldview diversification and how big the future could be In the most recent episode of the 80,000 Hours podcast, Rob Wiblin and Ajeya Cotra from Open Phil discuss "the challenge Open Phil faces striking a balance between taking big ideas seriously, and not going all in on philosophical argume...
5b2b6444-1b67-4a32-aa08-028beb9ec84f
trentmkelly/LessWrong-43k
LessWrong
Visiting the Bay Area from 17-30 June I'll be in the Bay Area from the 17-30 June for training as a TripleByte Interviewer. My interests include: * Effective altruism - particularly movement building * Agent foundations - particularly updateless decision theory and anthropics * Infinite ethics * More broadly, the...