id
stringlengths
36
36
source
stringclasses
15 values
formatted_source
stringclasses
13 values
text
stringlengths
2
7.55M
29747cb3-08ce-4ec5-b2ad-d03dc9972ec1
trentmkelly/LessWrong-43k
LessWrong
You Be the Jury: Survey on a Current Event As many of you probably know, in an Italian court early last weekend, two young students, Amanda Knox and Raffaele Sollecito, were convicted of killing another young student, Meredith Kercher, in a horrific way in November of 2007. (A third person, Rudy Guede, was convicted e...
c01d2a21-9107-4243-b52b-9659ce93713a
StampyAI/alignment-research-dataset/blogs
Blogs
October 2018 Newsletter The AI Alignment Forum [has left beta](https://intelligence.org/2018/10/29/announcing-the-ai-alignment-forum/)! Dovetailing with the launch, MIRI researchers Scott Garrabrant and Abram Demski will be releasing a new sequence introducing our research over the coming week, beginning here: [Em...
59e44296-7de7-4d8c-aff8-a4c64a227bf0
StampyAI/alignment-research-dataset/blogs
Blogs
Energy efficiency of Boeing 747-400 *Updated Nov 5, 2020* The Boeing 747-400: * covers around 0.0021m/kJ. * and moves mass at around 0.39 – 0.83 kg.m/J Details ------- The *Boeing 747-400* is a 1987 passenger plane.[1](https://aiimpacts.org/energy-efficiency-of-boeing-747-400/#easy-footnote-bottom-1-2745 "&#82...
24104841-aa63-43db-81d2-2612bc9ab886
trentmkelly/LessWrong-43k
LessWrong
ChatGPT challenges the case for human irrationality TLDR: Humans make systematic errors: apparent overconfidence, the conjunction fallacy, the gambler’s fallacy, and belief inertia. For decades, this has been taken to show that we reason in an irrational, non-probabilistic way. But GPT4—a probabilistic inference machi...
ee3562b9-156c-4e4a-9083-709ea662e758
trentmkelly/LessWrong-43k
LessWrong
[Intuitive self-models] 4. Trance 4.1 Post summary / Table of contents Part of the Intuitive Self-Models series. “Trance” is an umbrella term for various states of consciousness in which “you lose yourself”, somehow. The first kind that I learned about was hypnotic trance, as depicted in the media: Source: tvtropes ...
2f8c1984-27d4-4571-9adc-eec54438d2de
StampyAI/alignment-research-dataset/lesswrong
LessWrong
EIS XII: Summary Part 12 of 12 in the [Engineer’s Interpretability Sequence](https://www.alignmentforum.org/s/a6ne2ve5uturEEQK7). **TAISIC = “the AI safety interpretability community”** **MI = “mechanistic interpretability”** There might be some addenda later, but for now, this is the final post in The Engineer’s ...
a0883d21-482f-450a-ae23-f9056c62e494
trentmkelly/LessWrong-43k
LessWrong
Beyond fire alarms: freeing the groupstruck Crossposted from AI Impacts [Content warning: death in fires, death in machine apocalypse] ‘No fire alarms for AGI’ Eliezer Yudkowsky wrote that ‘there’s no fire alarm for Artificial General Intelligence’, by which I think he meant: ‘there will be no future AI development...
8916c418-14c0-4497-8574-0d3eb3b505c7
trentmkelly/LessWrong-43k
LessWrong
D&D.Sci: Whom Shall You Call? [Evaluation and Ruleset] This is a followup to the D&D.Sci post I made twelve days ago; if you haven’t already read it, you should do so now before spoiling yourself. Here is the web interactive I built to let you evaluate your solution; below is an explanation of the rules used to gener...
c572a213-727d-498e-b7c8-7d3baa6e562b
StampyAI/alignment-research-dataset/special_docs
Other
Superintelligence As a Cause or Cure For Risks of Astronomical Suffering Introduction Work discussing the possible consequences of creating superintelligent AI (Yudkowsky 2008 , Bostrom 2014, Sotala & Yampolskiy 2015) has discussed superintelligence as a possible existential risk: a risk "where an adverse outcome wou...
9aec1906-2061-4320-9cb3-b4b81fa5c9f4
trentmkelly/LessWrong-43k
LessWrong
One-layer transformers aren’t equivalent to a set of skip-trigrams (thanks to Tao Lin and Ryan Greenblatt for pointing this out, and to Arthur Conmy, Jenny Nitishinskaya, Thomas Huck, Neel Nanda, and Lawrence Chan, Ben Toner, and Chris Olah for comments, and many others for useful discussion.) In “A Mathematical Fram...
0c01d743-169f-4287-b8fe-4664a26ecd29
trentmkelly/LessWrong-43k
LessWrong
Will Darkcoin Have Thiel's Last Mover Advantage Thiel plays with the concept of last mover advantage when talking about what is valuable about Google and Facebook. His claim is that being the first social network (who knows what that was) is not, contra intuition, what matters, instead, if you are the last search engi...
16c7725c-f184-443a-b580-f2d544ce61d5
trentmkelly/LessWrong-43k
LessWrong
Meetup : Salt Lake City, UT: Summer Solstice 2014 Discussion article for the meetup : Salt Lake City, UT: Summer Solstice 2014 WHEN: 22 June 2014 02:00:13PM (-0600) WHERE: 9771 S. 170 E. Summer Solstice 2014 Sunday June 22, 2014 2pm 9771 S. 170 E. Sandy This is an event to bring all the friends and family to. Ther...
a419297d-d3da-49e7-91cc-d5c320d9aa32
trentmkelly/LessWrong-43k
LessWrong
Beware the Delmore Effect This is a linkpost for https://lydianottingham.substack.com/beware-the-delmore-effect  You should be aware of the Delmore Effect: > “The tendency to provide more articulate and explicit goals for lower priority areas of our lives.” —Cognitive Bias Foundation It seemed responsible when I fi...
3e41ce68-1562-4553-8898-e8d822e50554
trentmkelly/LessWrong-43k
LessWrong
Turing Test Tournament For Funding It's always troubled me that the standard Turing test provides only a single-bit output, and that the human being questioned could throw the game to make their AI counterpart look good. Also, research and development gets entirely too much funding based on what sounds cool rather tha...
ac892bf5-2678-4f86-b56c-f1b5702307ce
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Questions about AI that bother me *Crossposted from the EA Forum:* [*https://forum.effectivealtruism.org/posts/4TcaBNu7EmEukjGoc/questions-about-ai-that-have-bothered-me*](https://forum.effectivealtruism.org/posts/4TcaBNu7EmEukjGoc/questions-about-ai-that-have-bothered-me)   As 2022 comes to an end, I thought it'd ...
233dfb61-9ed8-42c7-9c3b-4c7e93e85349
trentmkelly/LessWrong-43k
LessWrong
Summer vs Winter Strategies Abstract: I have a hypothesis that there are two different general strategies for life that humans might switch between predicated on the way general resource availability change in the society. If it is constant or increasing one strategy pays off, if predictably increasing then decreasing...
eb1ac595-4d06-43b1-9a98-a381da026cca
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Quantum AI Box With quantum computing, an AI box needs not guard against hardware exploits. Epistemic status: *armchair computer science* Here's a simple machine model no more powerful than a quantum computer. It's a computer with extra instructions. It can spend x\*t time to, within the machine, split time into x²...
09fe1421-6323-48d3-8031-740577209a66
StampyAI/alignment-research-dataset/lesswrong
LessWrong
"Misbehaving Machines: The Emulated Brains of Transhumanist Dreams", Corry Shores [“Misbehaving Machines: The Emulated Brains of Transhumanist Dreams”](http://jetpress.org/v22/shores.htm), by Corry Shores (grad student; [Twitter](https://twitter.com/#!/corryshores), [blog](http://piratesandrevolutionaries.blogspot.com...
a38b2ba1-2aa1-4ac1-997a-086b1fa01092
trentmkelly/LessWrong-43k
LessWrong
Becoming a Better Community So I've been following Project Hufflepuff, the efforts of the rationalist community to become, rather than better rationalists (per se), but a better community.  I recently read the summary of the recent Project Hufflepuff Unconference, and I had a thought.   The Problem LessWrong And ...
29b850ae-897f-4d4e-ad77-88ed036de2a3
trentmkelly/LessWrong-43k
LessWrong
Discussing how to align Transformative AI if it’s developed very soon Coauthored. Thanks to Holden Karnofsky, Misha Yagudin, Adam Bales, Michael Aird, and Sam Clarke for feedback. All views expressed are our own. Introduction Background Holden Karnofsky recently published a series on AI strategy "nearcasting”: Wh...
e129df8e-1e89-45b4-b831-f95a1f64915d
trentmkelly/LessWrong-43k
LessWrong
Luna Lovegood and the Chamber of Secrets - Part 8 Luna had too many Calls to Adventure. If there was an author writing Luna's life, Luna would file a complaint. Luna had three magical artifacts of incredible power and at least two of them were maps. What kind of author gives her hero two magical maps? It's redundant. ...
442c0dc9-a28e-4001-ba32-c60b7ac19064
trentmkelly/LessWrong-43k
LessWrong
Really Strong Features Found in Residual Stream [I'm writing this quickly because the results are really strong. I still need to do due diligence & compare to baselines, but it's really exciting!] Last post I found 600+ features in MLP layer-2 of Pythia-70M, but ablating the direction didn't always make sense nor the...
73b412e5-2524-41ee-9bef-2fb2f880700d
trentmkelly/LessWrong-43k
LessWrong
GPT-4.5 Can Play Losing Chess After recently playing some chess against GPT-4.5 (it is pretty good, a lot stronger at the game than GPT-4 was - I would say strong club player level), I decided to try losing chess. To my surprise, it was able to play a complete game, which is something I have not seen any other model b...
20b490b1-e563-4568-9155-8c1ac445aca0
StampyAI/alignment-research-dataset/arxiv
Arxiv
A Game-Based Approximate Verification of Deep Neural Networks with Provable Guarantees 1 Introduction --------------- Deep neural networks (DNNs or networks, for simplicity) have been developed for a variety of tasks, including malware detection [malware](#bib.bib1) , abnormal network activity detection [ryan:nips1...
414d25f2-ca4c-4792-8f3f-39ce1cd5c8f2
trentmkelly/LessWrong-43k
LessWrong
How do you read the news critically? Epistemic status: My own theories that developed over years based on reading and giving mainstream media interviews about Quantified Self. When the phrase 'fake news' became popular in 2016, many people called for critical news consumption. The American left used the term to talk ...
c44e3a7e-58ed-44ad-9250-90648c4ef171
trentmkelly/LessWrong-43k
LessWrong
Deregulating Distraction, Moving Towards the Goal, and Level Hopping This is the third post in a series discussing my recent bout of productivity. Within, I discuss two techniques I use to avoid akrasia and one technique I use to be especially productive. Deregulating Distraction I like to pretend that I have higher...
7b6447de-6b46-4903-939c-06f5208a90da
trentmkelly/LessWrong-43k
LessWrong
Deceptive Agents are a Good Way to Do Things A brief, accessible summary of the inner alignment problem. The safety problem with powerful ML training algorithms is that deceptive agents are a good way to do things. Meaning, when we search over a big space of models until we find a model that performs well on some los...
1229d90d-d6b0-4cc4-8629-0e2e3a3f4822
trentmkelly/LessWrong-43k
LessWrong
Fiction recommendation: "Spur" by Phil Geusz Our protagonist has accepted being cursed into the shape of a horse, taking the place of a wealthy man in exchange for luxuries and lots of monthly cash. The wealthy man dies - and to his horror, our hero finds he is still a horse. Events escalate, secrets are revealed, per...
24a37c5d-5b38-4ef3-b50b-df9e6002fb2b
trentmkelly/LessWrong-43k
LessWrong
Sleep Quality: Strategies that work for me TL/DR: Here are some strategies that helped me improve my sleep. Some of these are a bit unusual, and some I haven't seen discussed anywhere else. Interesting findings include: the "shoulder temperature test", using earplugs to program your mind for sleep, my confusion about ...
fe1ec66b-3510-4159-ad81-fe9c72756b69
trentmkelly/LessWrong-43k
LessWrong
Applying utility functions to humans considered harmful There's a lot of discussion on this site that seems to be assuming (implicitly or explicitly) that it's meaningful to talk about the utility functions of individual humans. I would like to question this assumption. To clarify: I don't question that you couldn't,...
0bd3cd30-3eb5-4c08-89f0-a7b6584f4990
trentmkelly/LessWrong-43k
LessWrong
Humans, chimpanzees and other animals [Epistemic status: underinformed musings; I am posting this not because I am sure of anything in it but because the point seems important and I don't recall seeing anyone else make it. Maybe that's because it's wrong.] A common analogy for the relationship between postulated supe...
95a8fa78-a9e9-44e9-8460-a46d9ce3a8cc
StampyAI/alignment-research-dataset/eaforum
Effective Altruism Forum
Anthropic's Responsible Scaling Policy & Long-Term Benefit Trust Anthropic's RSP is big news. Yay Anthropic for making good commitments about model development, deployment, and security. [Anthropic's Responsible Scaling Policy](https://www.anthropic.com/index/anthropics-responsible-scaling-policy) ===================...
e4c4d4af-1b1a-4b80-9135-21ed61c4fc77
trentmkelly/LessWrong-43k
LessWrong
Deconfusing Regret There are lots of different things, all of which are called regret. I treat them differently, so I need different names for the different classes of regret. Agentic regret. The sort of regret that you feel when you made a choice that was wrong and you knew it was the wrong choice and you failed as ...
afe60d52-0d92-4dc6-bd72-b412a2b28a3e
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
Shard theory alignment has important, often-overlooked free parameters. *A delayed hot take. This is pretty similar to* [*previous comments from Rohin*](https://www.lesswrong.com/posts/fopZesxLCGAXqqaPv/don-t-align-agents-to-evaluations-of-plans?commentId=hXFMkRexJN4weKGdF). Shard theory alignment requires magic - no...
c1b1ff32-c580-453b-8258-54781d96e0a4
trentmkelly/LessWrong-43k
LessWrong
What projects and efforts are there to promote AI safety research? As far as I can tell, the recognition of the existential danger of AI is at an all time. It is going mainstream! Unfortunately, most of the discourse seems very pessimistic. The bulk of the messaging seems to imply that the only thing we can do is wait...
7d918a7f-5ec2-4238-a015-a79b0195bcad
StampyAI/alignment-research-dataset/blogs
Blogs
David J. Atkinson on autonomous systems ![David J. Atkinson portrait](http://intelligence.org/wp-content/uploads/2014/04/Atkinson_w384.jpg) David J. Atkinson, Ph.D, is a Senior Research Scientist at the [Florida Institute for Human and Machine Cognition](http://www.ihmc.us/) (IHMC). His current area of research envisi...
9ed976e4-8c10-482b-86ad-9538416c2717
trentmkelly/LessWrong-43k
LessWrong
The allegory of the hospital Sylvanus, a professional programmer, has gotten into a car crash, with disastrous results: he’s paralyzed from the wrists down, and won’t be able to use a keyboard or mouse for six weeks while his hands heal. He has to take time off from his job, and worse, he has to take a break from his ...
90b521de-b542-4b76-a9b8-d4343e89d072
LDJnr/LessWrong-Amplify-Instruct
LessWrong
"It was a pleasure to see all major strands of the effective altruism movement gathered in one place at last week's Effective Altruism Summit. Representatives from GiveWell, The Life You Can Save, 80,000 Hours, Giving What We Can, Effective Animal Altruism, Leverage Research, the Center for Applied Rationality, and the...
0feabda2-cfa2-4d0f-b41a-0226b0fefec4
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
My Objections to "We’re All Gonna Die with Eliezer Yudkowsky" Introduction ============ I recently watched Eliezer Yudkowsky's [appearance on the Bankless podcast](https://www.youtube.com/watch?v=gA1sNLL6yg4), where he argued that AI was nigh-certain to end humanity. Since the podcast, [some](https://twitter.com/Noah...
8510df90-f20b-4858-9729-3234887c0c20
trentmkelly/LessWrong-43k
LessWrong
Fallibilism and Bayesianism Aren't fallibilism (a "natural science" principle) and Bayesianism pretty much the same thing?
ad9846ec-2c10-4870-a9db-d511a426c686
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
[AN #77]: Double descent: a unification of statistical theory and modern ML practice Find all Alignment Newsletter resources [here](http://rohinshah.com/alignment-newsletter/). In particular, you can [sign up](http://eepurl.com/dqMSZj), or look through this [spreadsheet](https://docs.google.com/spreadsheets/d/1PwWbWZ6...
2fdd316d-c7e1-466f-822f-47ed953ed1ec
trentmkelly/LessWrong-43k
LessWrong
General Cooperative Inverse RL Convergence This is brief technical note on how to get convergence in the cooperative inverse reinforcement learning framework. We extend cooperative inverse RL to partially observable domains and use a recent result on the grain of truth problem to establish (arguably very strong) condi...
d2a11bca-bc39-49d5-812d-f778478d6e6e
trentmkelly/LessWrong-43k
LessWrong
Things Probably Matter A while back when I wrote about how China’s economic development might not have increased happiness there much, Scott Sumner wrote a really interesting response, Does Anything Matter? He points out that it’s too easy to make this about exotic far-off Chinese. Much the same phenomenon occurs clo...
74c97244-255e-4e28-af60-f102b57f2716
trentmkelly/LessWrong-43k
LessWrong
[Poll] Less Wrong and Mainstream Philosophy: How Different are We? Despite being (IMO) a philosophy blog, many Less Wrongers tend to disparage mainstream philosophy and emphasize the divergence between our beliefs and theirs. But, how different are we really? My intention with this post is to quantify this difference....
c1b78f61-8c16-49e2-bc78-1a701ba1f3df
trentmkelly/LessWrong-43k
LessWrong
Option control Introduction and summary (This is the third in a series of four posts about how we might solve the alignment problem. See the first post for an introduction to this project and a summary of the posts that have been released thus far.) In the first post in this series, I distinguished between “motivatio...
42fb1474-7d66-4c79-a90e-f28b61aa15a4
StampyAI/alignment-research-dataset/arxiv
Arxiv
Interpretable Explanations of Black Boxes by Meaningful Perturbation 1 Introduction --------------- Given the powerful but often opaque nature of modern black box predictors such as deep neural networks [[4](#bib.bib4), [5](#bib.bib5)], there is a considerable interest in *explaining and understanding* predictors *...
afbfa400-31da-4597-96ac-1d3694869542
trentmkelly/LessWrong-43k
LessWrong
The shoot-the-moon strategy Sometimes you can solve a problem by intentionally making it "worse" to such an extreme degree that the problem goes away. Real-world examples: * I accidentally spilled cooking oil on my shoe, forming an unsightly stain. When soap and scrubbing failed to remove it, I instead smeared oil a...
a3ccf7a6-1700-4a82-9121-e8256fd25ae0
trentmkelly/LessWrong-43k
LessWrong
Meetup : Second Purdue Meetup Discussion article for the meetup : Second Purdue Meetup WHEN: 18 January 2013 07:00:00PM (-0500) WHERE: Purdue- Hicks Library The first meetup went really well. We had 5 people show up in total. We're hoping more people will show up for the next one. EDIT: We had 8 people show up for ...
532ccd80-33db-4e70-a815-ac0c819ac18f
trentmkelly/LessWrong-43k
LessWrong
Be your conformist, approval seeking, self People recommend that one another ‘be themselves’ rather than being influenced by outside expectations and norms. Nobody suggests others should try harder to follow the crowd. They needn’t anyway; we seem fairly motivated by impressing others and fitting in. Few seem interest...
c6f595bd-5966-49c6-bae8-b685fdccb8c5
trentmkelly/LessWrong-43k
LessWrong
Book Review: Existential Risk and Growth Many questions about existential risk are speculative and hard to answer. Even specific ones, like "How have the odds of nuclear war changed after the United States withdrew from the Iran deal?", are still multi-layered and rather chaotic. Because of this, there haven't been ma...
170ea945-1ff3-4b7f-9ab4-95ef9dcdc9d0
StampyAI/alignment-research-dataset/youtube
Youtube Transcripts
Stuart Russel on AI and the Midas Touch problem #shorts so ai in the standard model which has been pretty much enforced since the beginning of the field is that machines are intelligent to the extent that their actions can be expected to achieve their objectives so those objectives could be logical goals for problem s...
e6c19080-6b49-4b3f-b27e-908903bf29e5
trentmkelly/LessWrong-43k
LessWrong
Most empirical questions are unresolveable; The good, the bad, and the appropriately under-powered
4ca96332-298f-4ce5-9bcf-c700f0a99e22
StampyAI/alignment-research-dataset/eaforum
Effective Altruism Forum
Scale, schlep, and systems *Note: This post was crossposted from* [*Planned Obsolescence*](https://www.planned-obsolescence.org/scale-schlep-and-systems/?ref=planned-obsolescence-newsletter) *by the Forum team, with the author's permission. The author may not see or respond to comments on this post.* > This startlin...
73fc2764-dec0-40e0-91f5-5b65f377957f
trentmkelly/LessWrong-43k
LessWrong
Speculation on Current Opportunities for Unusually High Impact in Global Health Epistemic Status: armchair speculation from a non-expert. Short version: I expect things to get pretty bad in the Sahel region over the next year in particular. The area is an obvious target for global health interventions even in good ti...
17207735-23f4-40c3-834f-9a37d77a3b66
StampyAI/alignment-research-dataset/eaforum
Effective Altruism Forum
Working at EA organizations series: Machine Intelligence Research Institute [Cross-posted on [Lesswrong](http://lesswrong.com/r/discussion/lw/mxj/working_at_miri_an_interview_with_malo_bourgon/)] This is the fourth post in the Working At EA Organizations series. The posts so far: * [Why work at an EA organization...
48920653-2288-4907-a8fe-68df016e850a
StampyAI/alignment-research-dataset/blogs
Blogs
your terminal values are complex and not objective your terminal values are complex and not objective -------------------------------------------------- a lot of people seem to want [terminal](https://en.wikipedia.org/wiki/Instrumental_and_intrinsic_value) (aka intrinsic aka axiomatic) [values](what-is-value.html) (...
37c74319-80e9-4c2e-bbdd-da9d6fea3be3
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Hofstadter's Superrationality Possibly the main and original inspiration for Yudkowsky's various musings on what advanced game theories should do (eg. cooperate in [the Prisoner's Dilemma](/lw/tn/the_true_prisoners_dilemma/)) is a set of essays penned by Douglas Hofstadter (of *Godel, Escher, Bach*) 1983. Unfortunatel...
1fb275ae-5693-49c4-84fc-d839b5f80030
trentmkelly/LessWrong-43k
LessWrong
[Link] Eric Schmidt's new AI2050 Fund This is a linkpost for https://www.schmidtfutures.com/schmidt-futures-launches-ai2050-to-protect-our-human-future-in-the-age-of-artificial-intelligence/ I am posting this here as it may be of interest to some members. > Schmidt Futures Launches AI2050 to Protect Our Human Futur...
599ee714-65e2-467b-a6da-cc4906cd8fb4
trentmkelly/LessWrong-43k
LessWrong
Soulless morality Follow-up to: So you say you're an altruist The responses to So you say you're an altruist indicate that people have split their values into two categories: 1. values they use to decide what they want 2. values that are admissible for moral reasoning (where 2 is probably a subset of 1 for atheis...
a9f32482-6747-4f5b-b2a2-13b57ca401a9
StampyAI/alignment-research-dataset/lesswrong
LessWrong
Human preferences as RL critic values - implications for alignment TLDR: Human preferences might be largely the result of a critic network head, much like that used in SOTA agentic RL systems. The term "values" in humans might mean almost exactly what it does in RL: an estimate of discounted sum of future rewards. In ...
f54ff7d8-1703-4e29-805d-9b0977fe9306
trentmkelly/LessWrong-43k
LessWrong
The hundred-room problem This thought-experiment has been on my mind for a couple of days, and no doubt it's a special case of a more general problem identified somewhere by some philosopher that I haven't heard of yet. It goes like this: You are blindfolded, and then scanned, and ninety-nine atom-for-atom copies of ...
3a58158c-8b35-40fc-8ed5-bc0c5f7ebb75
trentmkelly/LessWrong-43k
LessWrong
Death - an essay This essay may not hold a position by the end. See [the original meaning of writing essays if you're confused](http://www.paulgraham.com/essay.html). [A cursory search for discussion articles on death](http://lesswrong.com/search/results?cx=015839050583929870010%3A-802ptn4igi&cof=FORID%3A11&ie=UTF-8&...
94a00298-018a-4ca3-a939-40de0b5474c4
trentmkelly/LessWrong-43k
LessWrong
Less Wrong Study Hall - Year Two The Less Wrong Study Hall is still going, two years after its creation. I'm trying to make last year's survey an annual thing, so I did it again. We've been growing slowly but steadily. According to this year's census, we have 60+ users. By comparison to Yvain's Less Wrong survey, that...
a07785de-9afb-49fe-bae0-e10f01ecb5ba
trentmkelly/LessWrong-43k
LessWrong
Podcast interview series featuring Dr. Peter Park Check out the Into AI Safety podcast on Spotify, Apple Podcasts, Amazon Music, YouTube Podcasts, and many other podcast listening platforms! As always, the best things come in 3s: dimensions, musketeers, pyramids, and... 3 installments of my interview with Dr. Peter P...
32cf7884-2e7d-408c-a31a-c05c6bc880c6
trentmkelly/LessWrong-43k
LessWrong
Fixed Point: a love story > Ring the bells that still can ring > Forget your perfect offering > There is a crack, a crack in everything > That's how the light gets in. God is making excuses again, which is a real drag. “I want to be authentic with you, I really do. But there’s a part of my mind that knows exactly ho...
8bd1002a-dca7-42e4-9913-40da87c5acb0
StampyAI/alignment-research-dataset/eaforum
Effective Altruism Forum
Criticism Thread: What things should OpenPhil improve on? This post was prompted by three other posts on the EA forum. A recent post raises the alarm about [abuse of power and relationships](https://forum.effectivealtruism.org/posts/JCyX29F77Jak5gbwq/ea-sexual-harassment-and-abuse) in the EA community. An [earlier pos...
6bb87c50-883a-4281-8285-3b1f99fa1ede
trentmkelly/LessWrong-43k
LessWrong
Forecasting Newsletter: August 2020. Highlights 538 releases model of the US elections; Trump predicted to win ~30% of the time. Study offers instructive comparison of New York covid models, finds that for the IHME model, reported death counts fell inside the 95% prediction intervals only 53% of the time. Biggest de...
6549f143-eafa-4382-a755-459d62c2ee00
trentmkelly/LessWrong-43k
LessWrong
Tips on how to promote effective altruism effectively? Less talk, more action. Hi! We are two 20 years old Finnish students from the University of Helsinki, who are very concerned about global issues. We have a lot of energy and have met a couple of times for organizing our thoughts. We found out about effective altr...
129e25ec-ddd3-4882-a1cd-7aeaa3a3372b
trentmkelly/LessWrong-43k
LessWrong
Why we want unbiased learning processes ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, tl;dr: if an agent has a biased learning process, it may choose actions that are worse (with certainty) for every possible reward function it could be learning. An agent learns its own reward function if there is a ...
e37fbd30-f1d2-46f7-90e2-dac0f620b348
trentmkelly/LessWrong-43k
LessWrong
Intertheoretic utility comparison: outcomes, strategies and utilities This post seeks to clarify the question of what utility functions should be defined over, in the situation where an agent has uncertainty over their own utility function. ---------------------------------------- Outcomes or worlds The most intuit...
7906e05d-fde6-4d29-9ac5-90d6c8b2b26b
trentmkelly/LessWrong-43k
LessWrong
Religious/Worldview Techniques This is really weird, but I find myself strongly drawn towards religion (specifically, Christianity), which for some reason feels intuitively right to me. I *know* or at least seem to know that I'm just infected with a meme, I know all the standard arguments, and the majority of my frien...
675bb3fe-61e3-4f65-bed6-d8b650384e66
trentmkelly/LessWrong-43k
LessWrong
Meetup : Tucson LW Meetup 0 Discussion article for the meetup : Tucson LW Meetup 0 WHEN: 30 March 2013 03:00:00PM (-0700) WHERE: 943 E University Blvd Tucson We will meet at Caffe Luce(943 E University Blvd #191), a coffee shop off of University Ave., near the U of A. I'll have a sign. This is a first meetup, so we...
cbf092be-358a-464d-af87-4cd722a37085
trentmkelly/LessWrong-43k
LessWrong
How to eat potato chips while typing Chopsticks. Sometimes I want to eat snacks while I work, honestly because I have a weirdly moving dread of stretches of work-time containing nothing in any way gratifying to me. Possibly because I realistically anticipate that I will not remain working long under those conditions....
d86f11c8-3ce4-4d5b-abda-17a9a726f66e
trentmkelly/LessWrong-43k
LessWrong
Embodiment is Indispensable for AGI There are many paths on the road to AGI. Nando de Freitas recently wrote that “Scale is all you need” following the release of Deepmind’s Gato paper. Larger and larger transformers, with language interfaces doing supervised learning on large datasets is one such touted path. While G...
2f3c6a3d-8b5c-4fac-aa86-3a1fee4096bb
trentmkelly/LessWrong-43k
LessWrong
Meetup : Pittsburgh Meetup: Expert Presentation on Motivation Discussion article for the meetup : Pittsburgh Meetup: Expert Presentation on Motivation WHEN: 18 October 2011 05:45:00PM (-0400) WHERE: 151 N Craig St, Apt #7C, Pittsburgh, PA 15213 Welcome to the second Pittsburgh meetup of the semester! Nick will be...
6290cec7-5a2c-47fd-a698-5dc15c5de0c5
LDJnr/LessWrong-Amplify-Instruct
LessWrong
"Click here to begin voting. Click here for a general overview of the 2019 review.We are now in the final stretches of the LessWrong 2019 Review. We’ve spent 2 weeks nominating posts, and a month reviewing them. And today, we begin the final vote!A recap on the major goals of the Review: First, it is an experiment in i...
6dd88065-13bd-4170-adfd-f10f36ae3668
trentmkelly/LessWrong-43k
LessWrong
Against Arguments For Exploitation [Kind of a rambling collection of thoughts that I tried to organize a bit. Hopefully it's interesting to read] Introduction I've recently had several conversations in which a member of the rationalist community made the case that exploiting people from less economically developed c...
db7aa85c-f130-43ab-b5eb-f98838358389
trentmkelly/LessWrong-43k
LessWrong
Enhancing Mathematical Modeling with LLMs: Goals, Challenges, and Evaluations   Introduction Motivation At QURI, we've been investing in developing and analyzing mathematical models, recently focusing on Squiggle AI, a system that leverages LLMs to generate models using the Squiggle language. We believe LLMs hold ...
466c1ab7-c241-4811-9af0-74681213d3b4
StampyAI/alignment-research-dataset/lesswrong
LessWrong
The Aliens have Landed! "General Thud! General Thud! Wake up! The aliens have landed. We must surrender!" General Thud's assistant Fred turned on the lights and opened the curtains to help Thud wake up and confront the situation. Thud was groggy because he had stayed up late supervising an ultimately successful missio...
98aa2df6-f700-4c61-9867-28036fd3e96e
StampyAI/alignment-research-dataset/arxiv
Arxiv
Stabilizing the Lottery Ticket Hypothesis 1 Introduction --------------- Neural network compression techniques like pruning (e.g., Han et al. ([2015](#bib.bib9))) and distillation (e.g., Hinton et al. ([2015](#bib.bib12))) are known to dramatically reduce the number of parameters necessary to represent the function...
0c0bc9a5-66d7-405d-ab51-f19804c12baf
trentmkelly/LessWrong-43k
LessWrong
Meetup : Rationality Meetup Vienna Discussion article for the meetup : Rationality Meetup Vienna WHEN: 09 May 2015 02:00:00PM (+0100) WHERE: Kaisermühlenstraße 24/2 directions: http://web.student.tuwien.ac.at/~e0326238/rationality_meetup/directions.html FB-Event for more details and discussions: https://www.facebo...
daf478f1-95e4-46a4-bd87-3c6a6c463e6d
StampyAI/alignment-research-dataset/youtube
Youtube Transcripts
Morality, uncertainty, and autonomous systems (Luciano Siebert) - 1st AiTech Symposium thank you note thank everybody for coming here to the high tech symposium I'm gonna be talking to you about morality uncertainty and autonomous systems and the message I wanted to give you that uncertainty can be a driving force to ...
77ad45b4-9576-47bd-bacc-91d3600c78cc
StampyAI/alignment-research-dataset/alignmentforum
Alignment Forum
Are there cognitive realms? *[Metadata: crossposted from <https://tsvibt.blogspot.com/2022/11/are-there-cognitive-realms.html>. First completed November 16, 2022. This essay is more like research notes than exposition, so context may be missing, the use of terms may change across essays, and the text may be revised la...
a059bb1a-37dd-4cd3-831e-4719a0938cef
trentmkelly/LessWrong-43k
LessWrong
Will autonomous cars be more economical/efficient as shared urban transit than busses or trains, and by how much? What's some good research on this? There's reason to think they would be. When people get into a shared public transit vehicle, usually, they're not all going to the same place. The vehicle has to take an ...
33f7b620-a72a-41ef-8b36-d8ee4e326ce4
trentmkelly/LessWrong-43k
LessWrong
Preference conditional on circumstances and past preference satisfaction I've mentioned conditional preferences before. These are preferences that are dependent on facts about the world, for example "I'd want to believe X if there are strong argument for X". But there is another type of preference that is conditional...
a785fff2-78b9-4a65-b4ec-697d8e33dd5b
trentmkelly/LessWrong-43k
LessWrong
Why Ranked Choice Voting Isn't Great Several friends are collecting signatures to put Instant-runoff Voting, branded as Ranked Choice Voting, on the ballot in Massachusetts ( Ballotpedia, full text). I'm glad that an attempt to try a different voting method is getting traction, but I'm frustrated that they've chosen I...
1aba05b6-0f8e-430c-a50a-6cfe479c7563
trentmkelly/LessWrong-43k
LessWrong
Metaculus Is Open Source We completely rewrote the platform to make it easier for open-source contributors to build on our code. Maybe that's you? If it is, you can find the Metaculus code repo and can learn more here.  We've included instructions for setting up a local development environment and on creating an issu...
d0769a1a-7e5d-4729-95e6-fa4034e3e97b
StampyAI/alignment-research-dataset/arxiv
Arxiv
Adaptive Online Planning for Continual Lifelong Learning 1 Introduction --------------- We consider agents in a human-like setting, where agents must simultaneously act and learn in the world continuously with limited computational resources. All decisions are made online; there are no discrete episodes. Furthermor...
5d2d6006-0d0b-4a9f-8a56-950a50e3b898
trentmkelly/LessWrong-43k
LessWrong
Open thread, Mar. 9 - Mar. 15, 2015 If it's worth saying, but not worth its own post (even in Discussion), 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 ...
f894efbe-22bd-42e9-b7ee-54d9c87ffcbe
trentmkelly/LessWrong-43k
LessWrong
Approaches to Thump In playing for contra dances there are logistical and economic pressures that push for smaller bands. While the older larger bands would often include drumkits, if you're thinking about how to allocate roles among three—or especially two—people you're usually (but not always) going to do without a ...
75e507f4-9527-479d-ac95-b1e3f4aaf58a
trentmkelly/LessWrong-43k
LessWrong
Pop Culture Alignment Research and Taxes TL;DR: A quick recap of all the AI progress published recently, a shortcoming of the alignment tax definition, and a dynamical systems model of AI progress. Previously in this series: Cognitive Biases in Large Language Models, Drug addicts and deceptively aligned agents - a co...
ea2e26e9-0b46-4bd4-ae2d-de0f0021d9bd
StampyAI/alignment-research-dataset/eaforum
Effective Altruism Forum
Counterarguments to the basic AI risk case *This is cross-posted from the* [*AI Impacts blog*](https://aiimpacts.org/counterarguments-to-the-basic-ai-x-risk-case/) This is going to be a list of holes I see in the basic argument for existential risk from superhuman AI systems[1](https://aiimpacts.org/counterarguments-...
56058588-ff7a-40d0-8fda-4dffd668be3a
trentmkelly/LessWrong-43k
LessWrong
You Can’t Predict a Game of Pinball Introduction When thinking about a new idea, it helps to have a particular example to use to gain intuition and to clarify your thoughts. Games are particularly helpful for this, because they have well defined rules and goals. Many of the most impressive abilities of current AI syst...
271ea53b-b19a-4f54-ab80-1bf4104203d6
trentmkelly/LessWrong-43k
LessWrong
[brainstorm] - What should the AGIrisk community look like? I've been thinking for a bit what I would like the AGI risk community to look like. I'm curious what all your thoughts are. I'll be posting all my ideas, but I encourage other people to post their own ideas.
daa222e5-cb9c-4b1c-96cb-24d5a231dd44
trentmkelly/LessWrong-43k
LessWrong
Blind Empiricism Follow-up to: Living in an Inadequate World ----------------------------------------   The thesis that needs to be contrasted with modesty is not the assertion that everyone can beat their civilization all the time. It’s not that we should be the sort of person who sees the world as mad and pursues...
6a1b9f8f-90b1-4a49-86dd-03b8bea30517
trentmkelly/LessWrong-43k
LessWrong
Meetup : Melbourne, practical rationality Discussion article for the meetup : Melbourne, practical rationality WHEN: 07 December 2012 07:00:00PM (+1100) WHERE: 55 Walsh St West Melbourne 3003 Practical rationality. This meetup repeats on the 1st Friday of each month and is distinct from our social meetup on the 3rd...
e8344778-086c-4ac1-a30d-2222eb2dbeb2
trentmkelly/LessWrong-43k
LessWrong
Outside View(s) and MIRI's FAI Endgame On the subject of how an FAI team can avoid accidentally creating a UFAI, Carl Shulman wrote: > If we condition on having all other variables optimized, I'd expect a team to adopt very high standards of proof, and recognize limits to its own capabilities, biases, etc. One of the...
b88e41d0-3c21-409b-a052-b8e527374c06
trentmkelly/LessWrong-43k
LessWrong
The Last Days of the Singularity Challenge From Michael Anissimov on the Singularity Institute blog: Thanks to generous contributions by our donors, we are only $11,840 away from fulfilling our $100,000 goal for the 2010 Singularity Research Challenge. For every dollar you contribute to SIAI, another dollar is contri...
f5b4565d-2287-42f6-8a30-6194ff790d49
trentmkelly/LessWrong-43k
LessWrong
Important open problems in voting Strategy-resistance Identify, or prove impossibility, of a voting system which incentivizes— 1. A strictly sincere ranking of all candidates in the zero-information setting, where it implements a "good" social choice rule such as the relative (normalized) utilitarian rule, a Condor...
ecbbbdba-ec2c-4fbc-b35d-ca54c505482f
trentmkelly/LessWrong-43k
LessWrong
On Having Enough Socks None