id stringlengths 36 36 | source stringclasses 15
values | formatted_source stringclasses 13
values | text stringlengths 2 7.55M |
|---|---|---|---|
426a2e4c-9410-4814-a7c8-14ff4ce8c4c1 | trentmkelly/LessWrong-43k | LessWrong | How Smart Are Humans?
Epistemic status: free speculation
How intelligent should we expect AI to be, compared to humans, and how quickly should we expect it to reach this level of intelligence? This is of course an important strategic question, which has a large impact on our AI threat models. At the moment, the most ... |
22694b71-3688-4732-b496-40c398148aa6 | StampyAI/alignment-research-dataset/eaforum | Effective Altruism Forum | A Viral License for AI Safety
A Viral License for AI Safety: The GPL as a Model for Cooperation
=================================================================
*Written by Ivan Vendrov & Nat Kozak*
Introduction
------------
The GPL or *General Public License* was originally written by Richard Stallman for the ... |
87a374ea-8ecb-45c3-a386-a6344fd2966b | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | Parsing Abram on Gradations of Inner Alignment Obstacles
*This is independent research. To make further posts like this possible, please consider* [*supporting me*](https://www.alexflint.io/donate.html)*.*
*Epistemic status: This is my highly imperfect understanding of a lot of detailed thinking by Abram, Evan, John... |
67c1ebeb-eb22-46dc-a80d-13ee93826844 | awestover/filtering-for-misalignment | Redwood Research: Alek's Filtering Results | id: post3893
These are problems both for utility indifference and the causal version of factoring out variables . Thanks to Eliezer for pointing out the second problem. Suppose there are two utility functions, U a and U b . There is some button; let B be the event that a button is pressed. The button can only be pr... |
972f9a97-ba79-4b9e-977b-e12e25943f8c | trentmkelly/LessWrong-43k | LessWrong | Six-Door Cars
While at this point my family has decided what we want to do about cars, I'm still thinking about why the market for moderately larger cars is so awkward.
One issue with three-row vehicles is that the third row is hard to get into. If that row is rear-facing, like in a traditional station wagon, you get... |
917962ee-6bc3-4afb-81d9-e6b42882de83 | trentmkelly/LessWrong-43k | LessWrong | Advice I found helpful in 2022
Here are 10 of helpful pieces of advice I received in 2022:
1. Understand things for yourself. Figure out what you actually think about this AI stuff. What do you think will happen? What concrete risks and scenarios are you worried about? (Credit to Oliver Habryka. Last year at a winte... |
983ab654-1a38-40aa-890b-458cdde17405 | trentmkelly/LessWrong-43k | LessWrong | Near-Term Risks of an Obedient Artificial Intelligence
[Hi everyone, Yassine here: long-time orbiter first-time poster. I figured this piece I published on 1/23/23 would be as good as any of an introduction.]
I’ll be honest: I used to think talk of AI risk was so boring that I literally banned the topic at every part... |
e84c8347-a2d1-48c4-beca-019a6d486068 | trentmkelly/LessWrong-43k | LessWrong | Meetup : Monthly Bay Area meetup
Discussion article for the meetup : Monthly Bay Area meetup
WHEN: 26 November 2011 07:00:00PM (-0800)
WHERE: Berkeley
The next monthly Less Wrong meetup will be Saturday, November 26, at 7pm, in Berkeley. We'll gather in the Starbucks at 2128 Oxford Street, and then walk to the Free... |
fb80c975-212e-49ae-bde5-af2ba9391285 | trentmkelly/LessWrong-43k | LessWrong | All About Concave and Convex Agents
An entry-level characterization of some types of guy in decision theory, and in real life, interspersed with short stories about them
A concave function bends down. A convex function bends up. A linear function does neither.
Decisionmaking agents have utility functions. A utility ... |
8dc926a7-b702-48b8-9426-18d89c82f021 | trentmkelly/LessWrong-43k | LessWrong | Judgment Under Uncertainty summaries, Part 1: Representativeness
Judgment Under Uncertainty: Heuristics and Biases is one of the foundational works on the flaws of human reasoning, and as such gets cited a lot on Less Wrong — but it's also rather long and esoteric, which makes it inaccessible to most Less Wrong users.... |
109d778b-25af-4e4e-af71-65329f1e8893 | trentmkelly/LessWrong-43k | LessWrong | Being a Realist (even if you believe in God)
I have a small prediction. Some time before the end of the next paragraph you are going to form a strong opinion as to whether this article has any value to you. I'm going to be using the G word, appealing to emotion, and generally flinging around rhetoric not backed by any... |
066fed0d-2c6c-4def-a5ab-1a0e82a30daa | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | REPL's and ELK
In my previous [post](https://www.lesswrong.com/posts/kN2cFPaLQhExEzgeZ/repl-s-a-type-signature-for-agents).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-s... |
50e8eec4-a276-4ad7-9b0e-2b0353393605 | trentmkelly/LessWrong-43k | LessWrong | Cohabitive Games so Far
A cohabitive game[1] is a partially cooperative, partially competitive multiplayer game that provides an anarchic dojo for development in applied cooperative bargaining, or negotiation.
Applied cooperative bargaining isn't currently taught, despite being an infrastructural literacy for peace, ... |
9f1f0938-9e8d-4b41-863e-785f4bc9475f | StampyAI/alignment-research-dataset/eaforum | Effective Altruism Forum | My summary of “Pragmatic AI Safety”
This post is my summary of [“Pragmatic AI Safety”](https://www.lesswrong.com/s/FaEBwhhe3otzYKGQt) on [complex systems](https://www.lesswrong.com/s/FaEBwhhe3otzYKGQt/p/n767Q8HqbrteaPA25) and [capabilities externalities](https://www.lesswrong.com/s/FaEBwhhe3otzYKGQt/p/dfRtxWcFDupfWpL... |
3df4f6eb-bc4d-4c97-a7d3-5e5acd8ba370 | StampyAI/alignment-research-dataset/arxiv | Arxiv | Generalizing meanings from partners to populations: Hierarchical inference supports convention formation on networks.
A hierarchical Bayesian model of
convention
-------------------------------------------
Figure 1: Schematic of hierachical Bayesian mo... |
d0f6ed28-d4ab-48ae-86de-0b2e2c5b7d99 | trentmkelly/LessWrong-43k | LessWrong | The Long Long Covid Post
A while back I mentioned I’d aim to write a longer post on Long Covid and Katja Grace’s post on it. This is that post. First I deal with Katja’s post, then Scott Alexander’s Long Covid: Much More Than You Wanted to Know.
My core model of Long Covid after writing this post:
1. Long Covid is... |
88bc360d-9e49-4b2e-9b22-3cafbce91126 | trentmkelly/LessWrong-43k | LessWrong | Learning from Human Preferences - from OpenAI (including Christiano, Amodei & Legg)
|
6cf064bc-8c8f-4846-9efd-92d4b9cf4162 | StampyAI/alignment-research-dataset/lesswrong | LessWrong | Research Report: Incorrectness Cascades
*[Edit: See updated version* [*here*](https://aizi.substack.com/p/research-report-incorrectness-cascades-259) *(*[*LW version*](https://www.lesswrong.com/posts/8fWuAYicWkqqmcGyt/research-report-incorrectness-cascades-corrected)*), which fixed a crucial bug and significantly chan... |
a6b75f87-942e-4304-b7da-c58927242bf1 | StampyAI/alignment-research-dataset/lesswrong | LessWrong | Let’s talk about uncontrollable AI
[*In another post*](https://www.lesswrong.com/posts/gEchYntjSXk9KXorK/uncontrollable-ai-as-an-existential-risk)*, I published four hypotheses on uncontrollable AI as an existential risk. Here I give some background on why I chose this particular framing.*
Talking about existential r... |
1e09080d-45a2-4ee0-b725-03b02a24471a | trentmkelly/LessWrong-43k | LessWrong | The track record of survey-based macroeconomic forecasting
I'm interested in forecasting, and one of the areas where plenty of forecasting has been done is macroeconomic indicators. This post looks at what's known about macroeconomic forecasting.
Macroeconomic indicators such as total GDP, GDP per capita, inflation, ... |
786df3ec-d4b3-42e8-9bbd-187c988bd39f | trentmkelly/LessWrong-43k | LessWrong | [SEQ RERUN] Disappointment in the Future
Today's post, Disappointment in the Future was originally published on 01 December 2008. A summary (taken from the LW wiki):
> A list of Ray Kurzweil's predictions for the period 1999-2009.
Discuss the post here (rather than in the comments to the original post).
This po... |
e4b05282-8d1d-48b0-a0a0-9a7653605c87 | trentmkelly/LessWrong-43k | LessWrong | The Trouble With "Good"
Related to: How An Algorithm Feels From Inside, The Affect Heuristic, The Power of Positivist Thinking
I am a normative utilitarian and a descriptive emotivist: I believe utilitarianism is the correct way to resolve moral problems, but that the normal mental algorithms for resolving moral prob... |
f654d1a2-c260-4c79-bb65-8d11f7ea6868 | trentmkelly/LessWrong-43k | LessWrong | Cognitive Science/Psychology As a Neglected Approach to AI Safety
|
53f63780-6a1d-4583-b3b5-1e22631feec6 | trentmkelly/LessWrong-43k | LessWrong | Would 2009 H1N1 (Swine Flu) ring the alarm bell?
Introduction
In order to assist in anticipating and responding to the next pandemic, I created an "alarm bell" questionnaire to suggest whether or not one should behave as if the world is about to experience a stock market-crashing pandemic.
To evaluate whether it is ... |
f0ca103c-48b4-4a53-add2-f3dc773e8074 | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | Bayesianism versus conservatism versus Goodhart
Key argument: if we use a non-Bayesian conservative approach, such as a minimum over different utility functions, then we better have a good reason as to why that would work. But if we have that reason, we can use it to make the whole thing into a Bayesian mix, which can... |
3347d894-75df-4036-8474-6e9337a2a298 | trentmkelly/LessWrong-43k | LessWrong | Principles of action: fundamentals of how action works
Introduction
For all action that an actor in the cosmos does, there can be determined general principles of action which are able to guide action which I call 'principles of action'. My principles of action are given and organised in what I call 'frameworks' which... |
e1e225d2-8583-410b-896e-5b7e4af096e2 | trentmkelly/LessWrong-43k | LessWrong | [QUESTION]: Academic social science and machine learning
I asked this question on Facebook here, and got some interesting answers, but I thought it would be interesting to ask LessWrong and get a larger range of opinions. I've modified the list of options somewhat.
What explains why some classification, prediction, a... |
2000eae0-a9bc-44e4-9d08-5d80f23f3c9e | StampyAI/alignment-research-dataset/special_docs | Other | Cooperation, Conflict, and Transformative Artificial Intelligence - A Research Agenda
LONGTERMRISK.ORGCooperation, Conflict,
and T ransformative
Artificial Intelligence
—
A RESEARCH
AGENDA
Jesse Clifton
—
March 2020
First draft: December 2019
Contents
1 Introduction 2
1.1 Cooperation failure: models and examples .... |
1298ba57-9999-4717-8f09-b009bb00d455 | trentmkelly/LessWrong-43k | LessWrong | Sunday August 23rd, 12pm (PDT) – Double Crux with Buck Shlegeris and Oliver Habryka on Slow vs. Fast AI Takeoff
This Sunday at 12pm (PDT), we're running a double crux between two curated LessWrong authors (see here for transcripts of events from previous weeks).
* We'll have 1.5–2 hours of Double Crux between Buck a... |
0ac8c557-dd25-41fd-9f88-eec40f1130eb | StampyAI/alignment-research-dataset/lesswrong | LessWrong | Ethics of Brain Emulation
I felt like this draft paper by Anders Sandberg was a well-thought-out essay on the morality of experiments on brain emulations. Is there anything you disagree with here, or think he should handle differently?
<http://www.aleph.se/papers/Ethics%20of%20brain%20emulations%20draft.pdf> |
ffffa4cc-fce8-4d6c-b942-6d436954666b | trentmkelly/LessWrong-43k | LessWrong | Editor Mini-Guide
,
Here is a brief overview of what you can do in the editor on LessWrong. This will change a great deal over time, so I’ll keep a note here of when the latest update was: Friday 5th July, 2019.
Three things you need to know
* Highlight text in the editor to get a popup menu.
* Whatever you make i... |
36a53a67-dfc2-456d-a071-298487461d57 | trentmkelly/LessWrong-43k | LessWrong | What are the advantages and disadvantages of knowing your own IQ?
I've seen some answers here: https://www.quora.com/What-are-the-advantages-disadvantages-to-know-your-own-IQ
But I would be curious to know the perspective from people here.
Third alternative: taking an IQ test and tracking your IQ, but not looking at... |
0774f382-d342-46a6-aaaf-469254de74b6 | trentmkelly/LessWrong-43k | LessWrong | Should We Tell People That Giving Makes Them Happier?
Why do people give to charity?
It seems strange to even ask. Most people would point to the fact that they’re altruistic and want to make a difference. Others are concerned with inequality and justice. Another group points to the concept of “paying it forward” or ... |
894a9acd-2d6b-49e6-866d-bc6d04c7bdb0 | trentmkelly/LessWrong-43k | LessWrong | In set theory, everything is a set
In Unix, everything is a file; in set theory, everything is a set.
----------------------------------------
This abstraction is very cool and allows us to prove facts that would not be provable if we thought that sets could include stuff that were not sets and therefore could not b... |
e4ec98fe-a048-4d07-bd55-8cfde0acc85e | trentmkelly/LessWrong-43k | LessWrong | H+ review of James Miller's Singularity Rising [link]
http://hplusmagazine.com/2014/01/16/review-singularity-rising-by-james-d-miller/ |
0b100329-9847-4bd1-8fcc-82f91da0e2d3 | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | What is a definition, how can it be extrapolated?
What is a definition? Philosophy has, ironically, a large number of [definitions of definitions](https://plato.stanford.edu/entries/definitions/), but three of them are especially relevant to ML and AI safety.
There is the [intensional definition](https://en.wikipedia... |
c7a26120-0ec6-403a-965f-41b33fb05d6e | trentmkelly/LessWrong-43k | LessWrong | Is Musk still net-positive for humanity?
Musk's behaviour has always been controversial and he's always been kind of a dick, but I don't think it is controversial at all to say that until some years ago he has been extremely net positive for society and humanity in general. However, he's behaviour and actions turned m... |
287bbcfd-0aca-47b7-b02f-eebb14af04eb | trentmkelly/LessWrong-43k | LessWrong | Understanding and controlling a maze-solving policy network
Previously: Predictions for shard theory mechanistic interpretability results
Locally retargeting the search by modifying a single activation. We found a residual channel halfway through a maze-solving network. When we set one of the channel activations to ... |
fe474692-ebe3-43ba-bc3f-55494dde7b09 | StampyAI/alignment-research-dataset/eaforum | Effective Altruism Forum | Tim Cook was asked about extinction risks from AI
Tim Cook (CEO of Apple) was on Good Morning America. The reporter asked about his concerns on AI, including extinction.
---
Q: You said, you want to remain deliberate and thoughtful in your approach. So what are your concerns?
> ...What people are now talking ab... |
decfc7b8-d93c-4e38-a528-471bcf9ffca8 | trentmkelly/LessWrong-43k | LessWrong | [Update: New URL] Today's Online Meetup: We're Using Mozilla Hubs
That's right! Tomorrow/Today (Sunday Mar 29th) we're having an online LessWrong meetup in Mozilla Hubs. You're welcome to join and chat with some of the interesting people who read LessWrong, SSC, Overcoming Bias, and so on. I am personally quite excite... |
05a3ded9-1504-4a75-bb8a-44a5aa31d640 | trentmkelly/LessWrong-43k | LessWrong | Phallocentricity in GPT-J's bizarre stratified ontology
TL;DR This relates to the findings reported in my posts Mapping the Semantic Void parts I and II. By creating a custom embedding at the token centroid (the mean vector of all 50,257 GPT-J token embeddings), prompting the model to define it and considering logits,... |
4d9e8e10-c663-4139-b8b9-3640597e526a | trentmkelly/LessWrong-43k | LessWrong | Employee Incentives Make AGI Lab Pauses More Costly
Disclaimer: I know approximately nothing about internal governance at AGI labs. My current understanding is that the main AGI labs have basically no plans for how to navigate a conditional pause, and this post is predicated on this assumption.
TL;DR: If an AGI lab p... |
db586c36-038f-4415-943d-a64a3ff8b1fc | StampyAI/alignment-research-dataset/arxiv | Arxiv | BERT-ATTACK: Adversarial Attack Against BERT Using BERT
1 Introduction
---------------
Despite the success of deep learning, recent works have found that these neural networks are vulnerable to adversarial samples, which are crafted with small perturbations to the original inputs Goodfellow et al. ([2014](#bib.bib6... |
469711c3-027a-44b4-828a-bd54670b902c | trentmkelly/LessWrong-43k | LessWrong | Project Moonbeam
Special thanks to Anders Sandberg for help in developing the core idea of this story, Remmelt for editorial feedback, and to Forrest Landry for identifying the underlying ideas.
The Mission
You have been tasked with a feasibility and safety analysis for "Project Moonbeam", a plan to seed the Moon w... |
1677314e-f769-4e12-b6d6-be3bd28ac2c4 | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | 200 COP in MI: Exploring Polysemanticity and Superposition
*This is the fifth post in a sequence called 200 Concrete Open Problems in Mechanistic Interpretability.*[*Start here*](https://www.alignmentforum.org/posts/LbrPTJ4fmABEdEnLf/200-concrete-open-problems-in-mechanistic-interpretability)*, then read in any order.... |
f42f226b-e5fa-47bd-8847-750d1caf3ad7 | trentmkelly/LessWrong-43k | LessWrong | Simulation arguments
> “If Nature herself proves artificial, where will you go to seek wildness? Where is the real out-of-doors?”
>
> – C.S. Lewis
(Cross-posted from Hands and Cities. Content warning: weird)
Summary:
* This post examines “simulation arguments” – i.e., arguments that we should assign significant p... |
d66275cf-f693-4867-bdd7-83cf84d1fa29 | trentmkelly/LessWrong-43k | LessWrong | [Link] Philosophers’ Non-Profit Offers Free Online Critical Thinking Course
https://dailynous.com/2021/01/25/philosophers-non-profit-offers-free-online-critical-thinking-course/
“A Boston-area non-profit organization has created a free, online, critical thinking course for students and teachers—a ‘boot camp for argum... |
a23128e2-d66d-43ed-bb89-8d589fdba9a4 | trentmkelly/LessWrong-43k | LessWrong | Fissure opens in chess AI scene [link]
http://boingboing.net/2012/01/19/fissure-opens-in-chess-ai-scen.html |
31f6a9e5-9ece-497f-8c3f-26ff97031337 | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | More findings on maximal data dimension
Produced as part of the [SERI ML Alignment Theory Scholars Program](https://www.serimats.org/) - Winter 2022 Cohort.
I’d like to thank Wes Gurnee, Aryan Bhatt, Eric Purdy and Stefan Heimersheim for discussions and Evan Hubinger, Neel Nanda, Adam Jermyn and Chris Olah for mentor... |
1ea6c9ea-8ade-4649-b021-7e4af512fa8d | StampyAI/alignment-research-dataset/eaforum | Effective Altruism Forum | Digital people could make AI safer
*Epistemic status: Speculation, I am not confident in my views here and I skim over a lot of issues.*
**TL;DR: I claim that accelerating the development of "Whole Brain Emulation", such that it arrives before human level AI, is an intervention that would reduce the risks posed by AI... |
6007607c-e825-4c79-b1fc-8c8df191acd8 | trentmkelly/LessWrong-43k | LessWrong | Why I'll Keep My Crummy Drawings—How Generative AI Art Won't Supplant... Art.
The other day the post below caught my attention. In its characteristic bad-grammar-ambiguous-origin-meme style, Neo’s sentiment is one I share.
As an artist of sorts (it’s the only thing I’m actually qualified for) I see art as more abou... |
78de6558-d9f5-4c27-93ad-88526d6ea3ba | trentmkelly/LessWrong-43k | LessWrong | Summer AI Safety Intro Fellowships in Boston and Online (Policy & Technical) – Apply by June 6!
TL;DR: Apply to AISST’s AI Policy Fellowship and Technical Fellowship by Friday, June 6. Refer others here. AISST (a group of undergrad and grad students at Harvard) is excited to be running two introductory reading groups ... |
c073dd98-2f60-444b-b94b-c4596d17b361 | LDJnr/LessWrong-Amplify-Instruct | LessWrong | "The core fallacy of anthropomorphism is expecting something to be predicted by the black box of your brain, when its casual structure is so different from that of a human brain, as to give you no license to expect any such thing.
The Tragedy of Group Selectionism (as previously covered in the evolution sequence) was a... |
300e1d06-0893-4712-8b1b-e054df56f16c | StampyAI/alignment-research-dataset/aisafety.info | AI Safety Info | Why should we prepare for human-level AI technology now rather than decades down the line when it’s closer?
Human-level AI may not be far in the future. Many top AI researchers think that human-level AI will be developed fairly soon, so urgency is warranted. Furthermore, waiting until the last second to start addressi... |
59ba3f84-8c42-4f36-9fd1-a4abf2715c51 | StampyAI/alignment-research-dataset/eaforum | Effective Altruism Forum | AIs accelerating AI research
*Note: This post was crossposted from [Planned Obsolescence](https://www.planned-obsolescence.org/ais-accelerating-ai-research/) by the Forum team, with the author's permission. The author may not see or respond to comments on this post.*
> *Researchers could potentially design the next... |
4d5fae3b-ea5d-4778-b063-68bcf6005546 | trentmkelly/LessWrong-43k | LessWrong | Life-changing decisions pertaining to education. Help.
I just graduated from high school and I want to continue my formal education by studying for a bachelor's degree in science.
I can go on to study in: Hungary, England, the USA.
Hungay. I'll graduate debt-free and I will spend little on my cost of living(e.g., ac... |
72589bc4-b199-497d-b99f-378f4e6d2a0c | trentmkelly/LessWrong-43k | LessWrong | Software tools for community truth-seeking
In reply to: Community Epistemic Practice
There are software tools, possibly helpful for community truth-seeking. For example, truthmapping.com is described very well here. Also, debategraph.org, and I'm sure there are others.
|
3437691a-6647-44ac-8837-abb7dd26dc2a | trentmkelly/LessWrong-43k | LessWrong | Deep sparse autoencoders yield interpretable features too
Summary
* I sandwich the sparse layer in a sparse autoencoder (SAE) between non-sparse lower-dimensional layers and refer to this as a deep SAE.
* I find that features from deep SAEs are at least as interpretable as features from standard shallow SAEs.
* I c... |
352e9788-f89f-4988-b1f8-8edf59627b49 | trentmkelly/LessWrong-43k | LessWrong | Notes from the Qatar Center for Global Banking and Finance 3rd Annual Conference
Today, I attended a crypto-related panel discussion at the 3rd annual conference at the Qatar Center for Global Banking and Finance, delving deep into the future of crypto regulation.
Leonardo Gambacorta (BIS): Discussed the types of nat... |
04da9feb-7031-4739-8a8b-917f2bf44906 | trentmkelly/LessWrong-43k | LessWrong | People are gathering 2 million USD to save a kid with a rare disease. I feel weird about it. Why?
If someone organizes a charity to gather 50,000 USD to save a kid from a rare disease, I'm all for it. Even though that same amount could help or save a few people if invested more effectively. Screw effectiveness! A life... |
505dba83-a0f4-4f6f-aecc-e4ea7c8c88f3 | trentmkelly/LessWrong-43k | LessWrong | The Xerox Parc/ARPA version of the intellectual Turing test: Class 1 vs Class 2 disagreement
I've been reading the excellent book The Dream Machine about ARPA and PARC, the research communities that invented the personal computer (before them computers were just tools for military research projects).
Xerox PARC was m... |
6419a653-d5b7-4e6e-8a0e-d137ac9b7db5 | trentmkelly/LessWrong-43k | LessWrong | Response to Oren Etzioni's "How to know if artificial intelligence is about to destroy civilization"
Oren Etzioni recently wrote a popular article titled "How to know if artificial intelligence is about to destroy civilization." I think it's a good idea to write publicly available responses to articles like this, in c... |
c38197fa-2d0f-46eb-8ca6-427564cbf205 | StampyAI/alignment-research-dataset/eaforum | Effective Altruism Forum | ChatGPT bug leaked users' conversation histories
To me, this is evidence that culturally, OpenAI is not operating with a "security mindset". In my experience this sort of thing is relatively uniform across a company's culture, so if user data is not being treated in a secure way, we might conclude that the AI developm... |
ad551e87-3633-4d94-872a-c2e6ddda1d03 | trentmkelly/LessWrong-43k | LessWrong | How to partition teams to move fast? Debating "low-dimensional cuts"
jacobjacob
One of my favorite pastimes is to study historical examples of teams who accomplished ambitious projects effectively. Recently on this topic I wrote the post "A Golden Age of Building? Excerpts and lessons from Empire State, Pentagon, Skun... |
65eac51b-56f1-49b8-85e8-3c86726ef747 | trentmkelly/LessWrong-43k | LessWrong | What if we should use more energy, not less?
Declining growth rates and technological stagnation since the 70s correlate with flatlining energy use per capita.
I've been reading "Where Is My Flying Car?: A Memoir of Future Past" by J Storrs Hall. Ostensibly a book about a promised future that never arrived, it's a bro... |
974b421b-bf6a-4ac1-958f-5692f8fb4c68 | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | Humans Reflecting on HRH
**TL;DR**: HRH places a theoretical bound on the best reflection process achievable by us. Notably, this is not necessarily guaranteed to converge to human values, nor is it something that is actually implementable in practice (analogously to HCH). In particular, this is intended to argue agai... |
c18afd07-2199-4409-916e-06cb4c556a29 | trentmkelly/LessWrong-43k | LessWrong | Distinctions of the Moment
I could use a better name for this (any suggestions?), but I'll use "distinctions of the moment" for the moment.
A distinction of the moment is a distinction made between two synonyms or near-synonyms (or sometimes antonyms), for the sake of pointing out a difference which is important to t... |
384b270a-497e-4156-a8a0-64c38a64f719 | trentmkelly/LessWrong-43k | LessWrong | Exploitation as a Turing test
|
5a096cf6-c6f6-4a31-b9d2-576e0053108e | trentmkelly/LessWrong-43k | LessWrong | Knowledge Base 6: Consensus theory of truth
One of the theories defining the truth says that: “(…) truth is whatever is agreed upon, or in some versions, might come to be agreed upon, by some specified group. Such a group might include all human beings, or a subset thereof consisting of more than one person. (…) be ag... |
cbdc468f-6e65-4439-9e20-b0e885565c41 | trentmkelly/LessWrong-43k | LessWrong | Reneging Prosocially
Author's note: this essay was originally published elsewhere in 2019. It's now being permanently rehosted at this link. I'll be rehosting a small number of other upper-quintile essays from my past writing over the coming weeks.
----------------------------------------
Quite often, I will make an... |
a8efc818-e3fa-4be7-b472-c1c561e5bbda | StampyAI/alignment-research-dataset/eaforum | Effective Altruism Forum | Early-warning Forecasting Center: What it is, and why it'd be cool
Summary
=======
*See companion post* [*here*](https://forum.effectivealtruism.org/posts/E5vp2LCEfkrrLWozJ/potentially-great-ways-forecasting-can-improve-the-longterm)*.*
Most work on [forecasting](https://forum.effectivealtruism.org/tag/forecasting),... |
462dfb90-c563-4fd3-8233-8f18493dcd78 | trentmkelly/LessWrong-43k | LessWrong | My AI Model Delta Compared To Christiano
Preamble: Delta vs Crux
This section is redundant if you already read My AI Model Delta Compared To Yudkowsky.
I don’t natively think in terms of cruxes. But there’s a similar concept which is more natural for me, which I’ll call a delta.
Imagine that you and I each model the... |
9b6893ab-d90a-4c95-8cdf-567fbd845b8a | trentmkelly/LessWrong-43k | LessWrong | -
- |
09baa0ca-4189-4d8e-b8b3-404d3df19f0e | trentmkelly/LessWrong-43k | LessWrong | Age gaps and Birth order: Reanalysis
This post follows on from my previous post detailing some areas where I was unable to reproduce Scott’s analysis of how the age gap between siblings modifies the SSC Birth order effect. I suggest you read that post first but here’s the summary:
> I attempted to reproduce Scott’s a... |
47843e0e-eb92-4c9e-bb58-26da2a457f92 | trentmkelly/LessWrong-43k | LessWrong | My guess for why I was wrong about US housing
Edit/update: this is receiving more upvotes and I think it is worth explicitly pointing out that while housing is up 60% over this time frame, the Sp500 is up 90+%, so this is not intended as a full on refutation of all the points made in the original post.
A few years ag... |
1fed1238-e068-4bb7-b9c2-8db561d63700 | trentmkelly/LessWrong-43k | LessWrong | Starting a rationality dojo; any tips?
I have a small (4 core members) existing group of SSC/LW readers who meet up for lunch once a week, which gives us the raw materials to start such a group. We plan to meet at least once a month, if not once a week. We'll probably start with the Hammertime sequence for content.
I... |
963c2960-1072-460d-b581-1e682914f7ef | StampyAI/alignment-research-dataset/lesswrong | LessWrong | Alien Axiology
I argue that there is a convergent [axiological](https://en.wikipedia.org/wiki/Axiology) framework by which all intelligent life in the universe trends towards — that there is a universal-like account of the resource constraints, psychology, and game theory of all species across all worlds. I argue for ... |
b10f0865-4f88-4c01-ac9a-2cf80f1bf46b | trentmkelly/LessWrong-43k | LessWrong | Personal blogging as self-imposed oppression
You Should Start a Blog Right Now!
Or should you? You can find some blog posts about the benefits of blogging but I can't think of many that explore the potential harms. I can think of some reasons why blogging — especially under your real identity — could waste your time,... |
0b3d8387-74b1-4e9d-861a-8eb9e4b21766 | StampyAI/alignment-research-dataset/lesswrong | LessWrong | Intent-aligned AI systems deplete human agency: the need for agency foundations research in AI safety
This is a summary of our recent preprint [Intent-aligned AI systems deplete human agency: the need for agency foundations research in AI safety](https://arxiv.org/abs/2305.19223) and submission to the [Open Philanthro... |
af02f582-d87b-49a4-a92b-71b0c8521c88 | trentmkelly/LessWrong-43k | LessWrong | Gravity and the Thermodynamics of Human Economy and Existence
The prerequisitions of life
"What an organism feeds upon is negative entropy. ... Thus the device by which an organism maintains itself stationary at a fairly high level of he orderliness (=fairly low level of entropy) really consists continually sucking... |
cdaef79e-d619-4983-a32b-9b5a143edb03 | trentmkelly/LessWrong-43k | LessWrong | Article idea: Good argumentation
I'm newly subscribed but looking to contribute, and I was wondering if anyone would be interested in an article (or series thereof) teaching, in a formal and well-defined manner, how to argue. It would cover things like:
* The general outline for a compelling argument;
* How to syst... |
16b4fe8d-3fbe-4bef-ad97-35c07c5a307b | StampyAI/alignment-research-dataset/eaforum | Effective Altruism Forum | Benefits/Risks of Scott Aaronson's Orthodox/Reform Framing for AI Alignment
[Scott Aaronson makes the argument that](https://scottaaronson.blog/?p=6821) an Orthodox vs Reform analogy works for AI Alignment. I don't think I've really heard it framed this way before. He gives his takes on the beliefs of each group and t... |
3d280a44-27f6-4afa-9dae-7cb550afd972 | trentmkelly/LessWrong-43k | LessWrong | Self-modification, morality, and drugs
No, not psychoactive drugs: allergy drugs.
This is my attempt to come to grips with the idea of self-modification. I'm interested to know of any flaws folks might spot in this analogy or reasoning.
Gandhi wouldn't take a pill that would make him want to kill people. That is to ... |
23b2e48a-fa0d-4330-8e9c-aca55b611c9f | trentmkelly/LessWrong-43k | LessWrong | April 2016 Media Thread
This is the monthly thread for posting media of various types that you've found that you enjoy. Post what you're reading, listening to, watching, and your opinion of it. Post recommendations to blogs. Post whatever media you feel like discussing! To see previous recommendations, check out the o... |
b8d52b75-0df6-44b6-bcf3-dd19bd0a823a | trentmkelly/LessWrong-43k | LessWrong | LQPR: An Algorithm for Reinforcement Learning with Provable Safety Guarantees
We describe LQPR (Linear-Quadratic-Program-Regulator), an algorithm for model-based reinforcement learning that allows us to prove PAC-style bounds on the behavior of the system. We describe proposed experiments that can be performed in the ... |
c04d9749-dc05-4f23-96eb-b97d4016d983 | trentmkelly/LessWrong-43k | LessWrong | "Originality is nothing but judicious imitation" - Voltaire
TLDR: Throughout human history pattern recognition, inspiration, and reasoning have been at the core of most innovations - including artificial intelligence. It is theorised that a code-generation model with the ability to demonstrate the aforementioned param... |
be8f688f-f39f-4123-90d8-6e7c79cbdd29 | trentmkelly/LessWrong-43k | LessWrong | Whither OB?
Robin plans to cut back posting shortly, after he and I have our long-awaited Disagreement about AI self-improvement. As for myself - I'm not finished, but I'm way over schedule and need to move on soon. I'm not going to stop posting entirely (I doubt I could if I tried) but I'm not going to be posting d... |
4dbe328c-6658-49a9-a18f-4daa19611dad | StampyAI/alignment-research-dataset/lesswrong | LessWrong | Why the technological singularity by AGI may never happen
Artificial general intelligence is often assumed to improve exponentially through recursive self-improvement, resulting in a technological singularity. There are hidden assumptions in this model which should be made explicit so that their probability can be ass... |
bbf35c13-bcd3-4a69-b984-9dbe57056e90 | StampyAI/alignment-research-dataset/arxiv | Arxiv | Information-Driven Adaptive Sensing Based on Deep Reinforcement Learning
1. Introduction
----------------
A fundamental use case in the Internet of Things (IoT) is to represent a phenomenon of the real world with an estimation or prediction model informed by measurements done by IoT sensing devices. When the phenom... |
59738766-6829-416d-aa14-7e4a8b97ab69 | trentmkelly/LessWrong-43k | LessWrong | The Rationality Wars
Ever since Tversky and Kahneman started to gather evidence purporting to show that humans suffer from a large number of cognitive biases, other psychologists and philosophers have criticized these findings. For instance, philosopher L. J. Cohen argued in the 80's that there was something conceptua... |
775a29c6-486b-4a08-a174-0f6b0417a805 | StampyAI/alignment-research-dataset/alignmentforum | Alignment Forum | Alignment Research = Conceptual Alignment Research + Applied Alignment Research
Instead of talking about alignment research, we should differentiate **conceptual** alignment research and **applied** alignment research.
I expect these two categories to be quite obvious to most people around here: conceptual alignment ... |
7f11f7db-e297-4863-a307-7936d09e7d39 | trentmkelly/LessWrong-43k | LessWrong | Masterpiece
A sequel to qntm's Lena. Reading Lena first is helpful but not necessary.
----------------------------------------
We’re excited to announce the fourth annual MMindscaping competition! Over the last few years, interest in the art of mindscaping has continued to grow rapidly. We expect this year’s competi... |
96c3036e-3e82-4929-975c-6c1adb51639c | trentmkelly/LessWrong-43k | LessWrong | Information theoretic model analysis may not lend much insight, but we may have been doing them wrong!
Introduction
For a SERI MATS research sprint under John Wentworth, teams were given the following research prompt:
> Everyone's assignment will be to figure out how to operationalize the "broad peaks induce data com... |
44650c42-4aa8-42ae-9a8d-f0e23ca6bae1 | trentmkelly/LessWrong-43k | LessWrong | In conclusion: in the land beyond money pumps lie extreme events
In a previous article I've demonstrated that you can only avoid money pumps and arbitrage by using the von Neumann-Morgenstern axioms of expected utility. I argued in this post that even if you're not likely to face a money pump on one particular decisio... |
5a18a3db-c96b-4074-8321-fe09cc2685f3 | trentmkelly/LessWrong-43k | LessWrong | Why I Define My Experience At the Monastic Academy As Sexual Assault
Trigger Warning: This post includes highly triggering details of sexual assault.
***The views expressed here are my own. Hereafter any italics used will indicate that I am quoting another person or source. ***
In December 2021 I published an OPEN L... |
8b8851d8-c25f-48f7-8032-e2b79677780c | StampyAI/alignment-research-dataset/arbital | Arbital | Diamond maximizer
A difficult, far-reaching open problem in [https://arbital.com/p/2v](https://arbital.com/p/2v) is to specify an [unbounded](https://arbital.com/p/) formula for an agent that would, if run on an [unphysically large finite computer](https://arbital.com/p/), create as much diamond material as possible. ... |
010bf11e-6e76-4ab3-819d-159399149b69 | trentmkelly/LessWrong-43k | LessWrong | What technologies was LessWrong 2.0 built on?
I'm asking this question not only because I am curious, but also because I might consider contributing to the codebase eventually and this is a relevant consideration. |
5d0e6dab-2148-4f9a-bb5e-7fcc3eecac60 | StampyAI/alignment-research-dataset/eaforum | Effective Altruism Forum | EAG DC: Meta-Bottlenecks in Preventing AI Doom
*Epistemic status: I haven’t thought about most of this for more than a few hours. Where I quote people directly, that’s exactly the quote, otherwise assume this is just my, possibly flawed understanding, of limited interaction with others. Practising thinking well and s... |
48139e69-d385-45ce-8274-c6c633f220f0 | trentmkelly/LessWrong-43k | LessWrong | Quick thoughts on the implications of multi-agent views of mind on AI takeover
There was conversation on Facebook over an argument that any sufficiently complex system, whether a human, a society, or an AGI, will be unable to pursue a unified goal due to internal conflict among its parts, and that this should make us ... |
9aaa192a-bd30-4ebb-98d9-ded12d7b353a | StampyAI/alignment-research-dataset/eaforum | Effective Altruism Forum | High Impact Careers in Formal Verification: Artificial Intelligence
*This work is supported by* [*CEEALAR*](https://ceealar.org/)*.*
*Many thanks to Vanessa Kosoy, Zac Hatfield-Dodds, Koen Holtman, Aaron Gertler, and Aman Patel*
Software is difficult to do. Sometimes software can be made less wrong.
Wikipedia: In ... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.