text stringlengths 300 320k | source stringlengths 52 154 |
|---|---|
# Techniques to fix incorrect memorization?
One of the things I've been using Anki to memorize is friends' birthdays, but I realized after a few weeks that I had entered one of them wrong, in a way that was confusingly similar to another:
- Friend A: June 5
- Friend B: July 5 (should actually be 6)
Now when I study ... | https://www.lesswrong.com/posts/kwkQ8B6yaw3FEch7n/techniques-to-fix-incorrect-memorization |
# shoes with springs
There's something intuitively intriguing about the concept of shoes with spring elements, something that made many kids excited about getting "[moon shoes](https://www.youtube.com/watch?v=OXJAXlw2LcI)", but they found the actual item rather disappointing. Using springs somehow with legged movement... | https://www.lesswrong.com/posts/LwvfYGZTSud7Rsuyw/shoes-with-springs |
# Paper Summary: The Koha Code - A Biological Theory of Memory
[paper link](https://arxiv.org/abs/2109.02124)
[author’s presentation](https://www.youtube.com/watch?v=nR0SBUOoLRM)
This paper introduces the Koha model, which aims to explain how information is stored within the brain and how neurons learn to become pa... | https://www.lesswrong.com/posts/mhp5s4GEePjXGqcAm/paper-summary-the-koha-code-a-biological-theory-of-memory |
# Conversation Visualizer
Conversation Visualizer
=======================
Voice to text to image to reflection.
* * *
Contents
--------
* [Background](https://210ethan.github.io/thoughts/visualizer.html#background)
* [Construction](https://210ethan.github.io/thoughts/visualizer.html#construction)
* [Cost](ht... | https://www.lesswrong.com/posts/m7aA2niPf2hKPn7Fw/conversation-visualizer |
# The proper response to mistakes that have harmed others?
I have a tendency to feel very guilty when I have harmed others, especially when the harm was quite large. And I do think I've been legitimately quite hurtful and harmful to a number of people over the course of my life.
Some of my guilt has persisted for yea... | https://www.lesswrong.com/posts/GBkc5yRgomstBkabn/the-proper-response-to-mistakes-that-have-harmed-others |
# Taking responsibility and partial derivatives
A common pattern for myself over the years is to get into some kind of interpersonal ~"conflict", feel mildly to extremely indignant about how the other person is at fault, then later either through confrontation or reflection, realize that I actually held substantial re... | https://www.lesswrong.com/posts/MGFBQECgifXMeuoau/taking-responsibility-and-partial-derivatives |
# Planning to build a cryptographic box with perfect secrecy
Summary
=======
Since September 2023, I started learning a lot of math and programming skills in order to develop the safest cryptographic box in the world (and yes, I am [aiming high](https://forum.effectivealtruism.org/posts/CmGPp5p9RvTLuuzbt/aim-high-eve... | https://www.lesswrong.com/posts/TK8ptSJGvAqj2HaRr/planning-to-build-a-cryptographic-box-with-perfect-secrecy |
# AI Alignment Metastrategy
I call "alignment strategy" the high-level approach to solving the technical problem[^rea58985c4m]. For example, value learning is one strategy, while delegating alignment research to AI is another. I call "alignment metastrategy" the high-level approach to converging on solving the technic... | https://www.lesswrong.com/posts/TALmStNf6479uTwzT/ai-alignment-metastrategy |
# aisafety.info, the Table of Content
Here is a list of Q&A from [https://aisafety.info/](https://aisafety.info/) . When I discovered the site, I was impressed by the volume of material produced. However, the interface is optimized for beginners. The following table of contents is for individuals who wish to navigate ... | https://www.lesswrong.com/posts/ZqxP6pJe53xRnRb4j/aisafety-info-the-table-of-content |
# The Roots of Progress 2023 in review
2023 was another big year for me and The Roots of Progress.
It was a year when ROP as an organization really started to take off. Even though the org itself was [formed in 2021](https://rootsofprogress.org/nonprofit-announcement), at first it was just a vehicle for my own intell... | https://www.lesswrong.com/posts/gt2aFiicFZfWsNuS4/the-roots-of-progress-2023-in-review |
# A case for AI alignment being difficult
This is an attempt to distill a model of AGI alignment that I have gained primarily from thinkers such as Eliezer Yudkowsky (and to a lesser extent Paul Christiano), but explained in my own terms rather than attempting to hew close to these thinkers. I think I would be pretty ... | https://www.lesswrong.com/posts/wnkGXcAq4DCgY8HqA/a-case-for-ai-alignment-being-difficult |
# Dark Art: Inception
If rationality is about overcoming bias, then I think of the "dark arts" as the weaponization bias. While we don't believe bias ought to be wielded as a weapon, learning how to swing a sword is vital to defending against it.
In this post I'll discuss a dark art I call "inception". *Inception* is... | https://www.lesswrong.com/posts/sggqmKcf7fnWkJByg/dark-art-inception |
# Bayesian updating in real life is mostly about understanding your hypotheses
My sense is that an increasingly common viewpoint around here is that the last ~20 years of AI development and AI x-risk discourse are well-described by the following narrative:
> Eliezer Yudkowsky (and various others who were at least ini... | https://www.lesswrong.com/posts/jsKYmqHw4aGLw6fLZ/bayesian-updating-in-real-life-is-mostly-about-understanding |
# Rhythm Stage Setup Components
I typically play what I'm calling my "rhythm stage setup", though it needs a better name. It's a keyboard, foot drums, breath controller, and whistle-controlled synthesizer, connected with [code I write](https://github.com/jeffkaufman/jammer), that lets me give a full and varied sound w... | https://www.lesswrong.com/posts/DCQPenXsQKXrBqCos/rhythm-stage-setup-components |
# 2023 in AI predictions
Lots of people have made AI predictions in 2023. Here I compile a subset. I have a habit of setting an email reminder for the date of the prediction, when I see AI predictions, so that when they are resolved I can point out their accuracy or inaccuracy. I have compiled most of the email remind... | https://www.lesswrong.com/posts/EZxG6ySHCEjDvL5x4/2023-in-ai-predictions |
# A hermeneutic net for agency
*[Metadata: crossposted from [https://tsvibt.blogspot.com/2023/09/a-hermeneutic-net-for-agency.html](https://tsvibt.blogspot.com/2023/09/a-hermeneutic-net-for-agency.html). First completed September 4, 2023.]*
A hermeneutic net for agency is a natural method to try, to solve a bunch of ... | https://www.lesswrong.com/posts/TNQKFoWhAkLCB4Kt7/a-hermeneutic-net-for-agency |
# What should a non-genius do in the face of rapid progress in GAI to ensure a decent life?
How can I maximise my chances of having a decent life, given the very high likelihood that GAI will make all our intellectual labour useless in the next few years?
For example, I graduated from a good university a few years ag... | https://www.lesswrong.com/posts/4XTS2inDJQQr3KAzR/what-should-a-non-genius-do-in-the-face-of-rapid-progress-in |
# Stop talking about p(doom)
*Epistemic status: Complete speculation, somewhat informed by copious arguing about the subject on Twitter.*
As AI risk has moved into the mainstream over the past few years, I've come to believe that "p(doom)" is an actively harmful term for X-risk discourse, and people trying to mitigat... | https://www.lesswrong.com/posts/EwyviSHWrQcvicsry/stop-talking-about-p-doom |
# Mech Interp Challenge: January - Deciphering the Caesar Cipher Model
**I'm writing this post to discuss solutions to the November challenge, and present the challenge for this January.**
If you've not read the [first post in this sequence](https://www.alignmentforum.org/posts/dpRuey9MDHNACSDtn/mech-interp-challeng... | https://www.lesswrong.com/posts/tCnvjYxgBocXpDEMH/mech-interp-challenge-january-deciphering-the-caesar-cipher |
# [location-changed] 2024-01-07 hangout, cryonics and longevity montreal
this event is indoors.
**Location**: _Brûlerie St-Denis_ is a coffee shop on st-denis street between duluth and roy streets.
the event itself is free, but the coffee shop it's in might expect you buy a coffee/tea/other.
meetups for people... | https://www.lesswrong.com/events/TSgGLL79hNhoum3Wa/location-changed-2024-01-07-hangout-cryonics-and-longevity |
# Steering Llama-2 with contrastive activation additions
The effects of subtracting or adding a "sycophancy vector" to one bias term.
**TL;DR:** By just adding e.g. a "sycophancy vector" to one bias te... | https://www.lesswrong.com/posts/v7f8ayBxLhmMFRzpa/steering-llama-2-with-contrastive-activation-additions |
# The Old Year and The New
Meet inside The Shops at Waterloo Town Square - we will congregate in the seating area next to the Valu-Mart with the trees sticking out in the middle of the benches at 7pm for 15 minutes, and then head over to my nearby apartment's amenity room. If you've been around a few times, feel free ... | https://www.lesswrong.com/events/4kthJYhgpYW4dMHGj/the-old-year-and-the-new |
# Boston Solstice 2023 Retrospective
Saturday evening we held another year's secular solstice celebration ( [2022](https://www.jefftk.com/p/boston-solstice-2022-retrospective), [2020](https://www.jefftk.com/p/secular-solstice-2020), [2019](https://www.jefftk.com/p/boston-solstice-2019-retrospective), [2018](https://ww... | https://www.lesswrong.com/posts/NTdfTaumDFFz4oMC4/boston-solstice-2023-retrospective |
# Are Metaculus AI Timelines Inconsistent?
The Metaculus prediction markets on AI timelines are the most referenced ones that I've seen.
The two main ones are as follows:
[When will the first weakly general AI system be devised, tested, and publicly announced?](https://www.metaculus.com/questions/3479/date-weakl... | https://www.lesswrong.com/posts/FCgpT4sj5BSEdeHHy/are-metaculus-ai-timelines-inconsistent |
# AI Is Not Software
*Epistemic Status: This idea is, I think, widely understood in technical circles. I'm trying to convey it more clearly to a general audience. Edit: See related* [*posts like this one*](https://www.lesswrong.com/posts/cFzC996D7Jjds3vS9/arguing-by-definition) *by Eliezer for background on how we sho... | https://www.lesswrong.com/posts/iknSNTbb8deJwQuLJ/ai-is-not-software |
# Dating Roundup #2: If At First You Don’t Succeed
Developments around relationships and dating have a relatively small speed premium, also there are once again enough of them for a full post.
[The first speculated on why you’re still single.](https://thezvi.substack.com/p/dating-roundup-1-this-is-why-youre) We faile... | https://www.lesswrong.com/posts/y8g4bXF7sdT9RduT6/dating-roundup-2-if-at-first-you-don-t-succeed |
# OpenAI's Preparedness Framework: Praise & Recommendations
In this post, I summarize some of my thoughts on OpenAI's recent [preparedness framework](https://cdn.openai.com/openai-preparedness-framework-beta.pdf). The post focuses on my opinions/analysis of the framework, and it doesn't have much summary (for readers ... | https://www.lesswrong.com/posts/G2AghCGjGCkhWiFDd/openai-s-preparedness-framework-praise-and-recommendations |
# Cortés, AI Risk, and the Dynamics of Competing Conquerors
Cortés's conquest of the Aztecs reflects current anxieties surrounding AI risks. His technological and tactical edge over the Aztecs parallels the potential superiority of smarter-than-human AIs attempting to take over earth. This historical parallel implies ... | https://www.lesswrong.com/posts/ZyPguqo3HZwQWWSuC/cortes-ai-risk-and-the-dynamics-of-competing-conquerors |
# Apologizing is a Core Rationalist Skill
There’s this narrative about a tradeoff between:
* The virtue of [Saying Oops](https://www.lesswrong.com/posts/wCqfCLs8z5Qw4GbKS/the-importance-of-saying-oops), early and often, correcting course rather than continuing to pour oneself into a losing bet, vs
* The loss of s... | https://www.lesswrong.com/posts/xiTLBuhEMmoyeor6D/apologizing-is-a-core-rationalist-skill |
# Otherness and control in the age of AGI
(Cross-posted from [my website](https://joecarlsmith.com/2024/01/02/otherness-and-control-in-the-age-of-agi). PDF of the full series [here](https://jc.gatspress.com/pdf/otherness_full.pdf). Audio version of the full series [here](https://joecarlsmithaudio.buzzsprout.com/203473... | https://www.lesswrong.com/posts/TtkfjskkAurvEN8Fa/otherness-and-control-in-the-age-of-agi |
# Gentleness and the artificial Other
(*Cross-posted from [my website](https://joecarlsmith.com/2024/01/02/gentleness-and-the-artificial-other). Audio version [here](https://www.buzzsprout.com/2034731/14233021), or search "Joe Carlsmith Audio" on your podcast app.*
*This is the first essay in a series that I’m callin... | https://www.lesswrong.com/posts/mzvu8QTRXdvDReCAL/gentleness-and-the-artificial-other |
# Agent membranes and causal distance
For the goal of **maintaining the safety and autonomy** of particular agents (and potential moral patients), there seems to be something important about **maintaining the **[***membranes and boundaries***](https://www.lesswrong.com/tag/boundaries-membranes-technical) **of those ag... | https://www.lesswrong.com/posts/kJAAsrEqtMvsnauiX/agent-membranes-and-causal-distance |
# MonoPoly Restricted Trust
Now we know why he’s so happy
Two months ago (an eternity in podcastin... | https://www.lesswrong.com/posts/R4w2jhNCDzcbLtfFB/monopoly-restricted-trust |
# Trading off Lives
Let's say someone [proposes](https://news.ycombinator.com/item?id=38841008) that to reduce deaths from overly chaotic airplane evacuations we ban passenger distractions during the most dangerous parts: takeoff and landing. How could we decide whether a ban like this would be worth it?
The argument... | https://www.lesswrong.com/posts/7WoYpodK9vvch5pd6/trading-off-lives |
# Terminology: <something>-ware for ML?
Would it be useful to have a term, analogous to 'hardware', 'software', 'wetware', 'vaporware' etc.[^ware], which could be used **to distinguish learned/discovered components of software**, like gradient-trained DNNs, prompt-hacked LLMs, etc?
[^ware]: For some illuminating comp... | https://www.lesswrong.com/posts/6xFNFAiovnLcAbEfo/terminology-less-than-something-greater-than-ware-for-ml |
# Does LessWrong make a difference when it comes to AI alignment?
I see LessWrong is currently obsessed with AI Alignment. I spoke with some others on the unofficial LessWrong discord, and agreed that *LessWrong is becoming more and more specialised, thus scaring off any newcomers who aren't interested in AI*.
That ... | https://www.lesswrong.com/posts/bbnoqQq5TDBHE3off/does-lesswrong-make-a-difference-when-it-comes-to-ai |
# What Helped Me - Kale, Blood, CPAP, X-tiamine, Methylphenidate
Here are some things that in the last 6 months have helped me a lot, mainly with handling chronic fatigue and ADHD. This post is very quickly thrown together and it seems more appropriate for a short form, however just in case this is exactly the informa... | https://www.lesswrong.com/posts/BRwuYDsHr748jGJCz/what-helped-me-kale-blood-cpap-x-tiamine-methylphenidate |
# Johannes' Biography
[Schedule a meeting with me.](https://calendly.com/johannescmayer)
[Tsuyoku Naritai!](https://www.lesswrong.com/posts/DoLQN5ryZ9XkZjq5h/tsuyoku-naritai-i-want-to-become-stronger) I operate on a [modified version](https://www.lesswrong.com/posts/RXAmPpDGxABANaNrB/johannes-c-mayer-s-shortform?comm... | https://www.lesswrong.com/posts/mQESiNe9dQte2P5Gk/johannes-biography |
# Copyright Confrontation #1
Lawsuits and legal issues over copyright continued to get a lot of attention this week, so I’m gathering those topics into their own post. The ‘virtual #0’ post [is the relevant section from last week’s roundup](https://thezvi.substack.com/i/139988916/copyright-confrontation).
#### Four C... | https://www.lesswrong.com/posts/9WD8nkqLTcd8YJPpT/copyright-confrontation-1 |
# Practically A Book Review: Appendix to "Nonlinear's Evidence: Debunking False and Misleading Claims" (ThingOfThings)
*Subtitle: Taking nonprescription amphetamines across the U.S.-Mexico border is a felony*
I haven't followed the controversy long enough to be able to tell how correct Ozy is, but what I like about t... | https://www.lesswrong.com/posts/wNqufsqkicMNxabZz/practically-a-book-review-appendix-to-nonlinear-s-evidence |
# Safety First: safety before full alignment. The deontic sufficiency hypothesis.
It could be the case that these two goals are separable and independent:
* “AI *safety”:* avoiding existential risk, s-risk, actively negative outcomes
* “AI *getting-everything-we-want*” ([CEV](https://www.lesswrong.com/tag/cohere... | https://www.lesswrong.com/posts/dLbBiAkJFdWK3uu3u/safety-first-safety-before-full-alignment-the-deontic |
# Agent membranes/boundaries and formalizing “safety”
It might be possible to formalize what it means for an agent/moral patient to be **safe** via membranes/boundaries. This post tells one (*just one*)story for how the membranes idea could be useful for thinking about existential risk and AI safety.
Formalizing “saf... | https://www.lesswrong.com/posts/SuktPEmBGxPrnZRHd/agent-membranes-boundaries-and-formalizing-safety |
# $300 for the best sci-fi prompt: the results
Six months ago we [announced](https://www.lesswrong.com/posts/goq3HfxZaQcomuH6d/usd300-for-the-best-sci-fi-prompt):
> We would like to find the best prompt to make GPT-4 do the following:
>
> * write the first chapter of a science fiction novel
> * the result should... | https://www.lesswrong.com/posts/GDGFqiaj8ePujZWEc/usd300-for-the-best-sci-fi-prompt-the-results |
# What’s up with LLMs representing XORs of arbitrary features?
*Thanks to Clément Dumas, Nikola Jurković, Nora Belrose, Arthur Conmy, and Oam Patel for feedback.*
In the comments of the [post on Google Deepmind’s CCS challenges paper](https://www.lesswrong.com/posts/wtfvbsYjNHYYBmT3k/discussion-challenges-with-unsupe... | https://www.lesswrong.com/posts/hjJXCn9GsskysDceS/what-s-up-with-llms-representing-xors-of-arbitrary-features |
# Non-alignment project ideas for making transformative AI go well
This is a series of posts with lists of projects that it could be valuable for someone to work on. The unifying theme is that they are projects that:
* Would be especially valuable if transformative AI is coming in the next 10 years or so.
* Are n... | https://www.lesswrong.com/posts/4d2JZiiyvZshhbEeJ/non-alignment-project-ideas-for-making-transformative-ai-go-1 |
# Screen-supported Portable Monitor
I broke my MacBook screen [a few weeks ago](https://www.jefftk.com/p/reviving-a-2015-macbook) and it [didn't end up](https://www.facebook.com/jefftk/posts/pfbid032yd7aoKRfTUvToru7MZVYhSmisQnoTLz8JqGsixfVzaatSnobhoHDQGwvujov3cGl?comment_id=947250690100507) making sense to repair it. ... | https://www.lesswrong.com/posts/jPjhYmMdJ62cHgqhj/screen-supported-portable-monitor |
# AI #45: To Be Determined
The first half of the week was filled with continued talk about the New York Times lawsuit against OpenAI, which I covered in its own post. Then that talk seemed to mostly die down,, and things were relatively quiet. We got a bunch of predictions for 2024, and I experimented with prediction ... | https://www.lesswrong.com/posts/NfXF6MZTgae766aoX/ai-45-to-be-determined |
# Some Vacation Photos
In the past three months, I've flown overnight from San Francisco to the UK/Europe three times.
Three times, I got a window seat on the left side of the plane.
And three times... well, see for yourself.
. Audio version [here](https://www.buzzsprout.com/2034731/14246130), or search "Joe Carlsmith Audio" on your podcast app.*
*This essay is part of a series I'm calling "Otherness and control in the ... | https://www.lesswrong.com/posts/sJPbmm8Gd34vGYrKd/deep-atheism-and-ai-risk |
# Cellular reprogramming, pneumatic launch systems, and terraforming Mars: Some things I learned about at Foresight Vision Weekend
In December, I went to the Foresight Institute’s Vision Weekend 2023 in San Francisco. I had a lot of fun talking to a bunch of weird and ambitious geeks about the glorious abundant techno... | https://www.lesswrong.com/posts/2DgA72C8rLszuyT2w/cellular-reprogramming-pneumatic-launch-systems-and |
# The Gears of Argmax
In the [last post](https://www.lesswrong.com/posts/guqG7hmSt49i7LJ6y/when-can-optimization-be-done-safely), I made the high-level claim that optimization can sometimes be done safely. And went on to claim that a key factor enabling safe optimization is sufficiently understanding the optimization ... | https://www.lesswrong.com/posts/cLoxazqoGAPLjAhPL/the-gears-of-argmax |
# Safety Data Sheets for Optimization Processes
In the [last post](https://www.lesswrong.com/posts/cLoxazqoGAPLjAhPL/the-gears-of-argmax), I expressed that it would be nice to have [safety data sheets](https://en.wikipedia.org/wiki/Safety_data_sheet) for different optimization processes. These data sheets would descri... | https://www.lesswrong.com/posts/zY58PJHMReCKH6PTu/safety-data-sheets-for-optimization-processes |
# Best-Responding Is Not Always the Best Response
Suppose that you are designing a software system to [bargain](https://en.wikipedia.org/wiki/Ultimatum_game) on your behalf. Either side can walk away from negotiations, and we'll use this as our [baseline](https://www.investopedia.com/terms/b/best-alternative-to-a-nego... | https://www.lesswrong.com/posts/uguxceuv7ERp26kn9/best-responding-is-not-always-the-best-response |
# Project ideas: Governance during explosive technological growth
*This is part of a series of lists of projects. The unifying theme is that the projects are not targeted at solving alignment or engineered pandemics but still targeted at worlds where transformative AI is coming in the next 10 years or so. See *[*here*... | https://www.lesswrong.com/posts/pqJuLTAjehAKxtefh/project-ideas-governance-during-explosive-technological-1 |
# MIRI 2024 Mission and Strategy Update
As we [announced](https://www.lesswrong.com/posts/NjtHt55nFbw3gehzY/announcing-miri-s-new-ceo-and-leadership-team) back in October, I have taken on the senior leadership role at MIRI as its CEO. It’s a big pair of shoes to fill, and an awesome responsibility that I’m honored to ... | https://www.lesswrong.com/posts/q3bJYTB3dGRf5fbD9/miri-2024-mission-and-strategy-update |
# If I ran the zoo
*(I was musing about what it means for an incoherent lump of meat to "have preferences," and thought it might be illuminating to consider what I'd do if I were, approximately, God. It, uh, hasn't been illuminating yet, but it's been entertaining and still seems at least* potentially *fruitful.)*
Pr... | https://www.lesswrong.com/posts/TQNwdqrkPtArweSoo/if-i-ran-the-zoo |
# Striking Implications for Learning Theory, Interpretability — and Safety?
*This is a linkpost for the recent paper "*[*White-Box Transformers via Sparse Rate Reduction: Compression Is All There Is?*](https://arxiv.org/abs/2311.13110)*"*
It would obviously be lovely to have arguments about AI Safety techniques that ... | https://www.lesswrong.com/posts/KqWeRxuEoLPDwxndj/striking-implications-for-learning-theory-interpretability |
# Predictive model agents are sort of corrigible
*TLDR: Agents made out of conditioned predictive models are **not** utility maximisers, and, for instance, won't try to resist certain kinds of shutdown, despite being able to generally perform well.*
This is just a short cute example that I've explained in conversatio... | https://www.lesswrong.com/posts/jtHykMXyArKzkKEo4/predictive-model-agents-are-sort-of-corrigible |
# Forecast your 2024 with Fatebook
We've built [predict your year in 2024](https://fatebook.io/predict-your-year), a space for you to write down your forecasts for the year on [Fatebook](http://fatebook.io/). At the end of your year, you can return, resolve your predictions as YES, NO or AMBIGUOUS, and reflect.
, following up on a similar 2022 survey as well as one in 2016.
I encourage opening the ori... | https://www.lesswrong.com/posts/NfPxAp5uwgZugwovY/ai-impacts-survey-december-2023-edition |
# Catching AIs red-handed
We’ve been spending a lot of time recently thinking about how to mitigate risks posed by [scheming](https://www.lesswrong.com/s/J7JpFeijCK5urdbzv/p/yFofRxg7RRQYCcwFA) (a.k.a. deceptively aligned) models intentionally subverting your safety techniques in order to cause an unacceptable outcome.... | https://www.lesswrong.com/posts/i2nmBfCXnadeGmhzW/catching-ais-red-handed |
# What technical topics could help with boundaries/membranes?
I want to compile a list of topics that could$$ help inspire **formal/mathematical abstractions** for [boundaries/membranes](https://www.lesswrong.com/tag/boundaries-membranes-technical). (See my recent [explainer](https://www.lesswrong.com/posts/kJAAsrEqt... | https://www.lesswrong.com/posts/cpvTNsMepfu8ejBaD/what-technical-topics-could-help-with-boundaries-membranes |
# The Hippie Rabbit Hole -Nuggets of Gold in Rivers of Bullshit
I’ve been an aspiring rationalist ever since I read [the sequences](https://www.readthesequences.com/) a few years ago.
I also keep a deck of tarot cards on my desk.
I don’t believe there is an increased chance of drawing certain cards, or that “the un... | https://www.lesswrong.com/posts/Ekegw5vbfoyizbQe8/the-hippie-rabbit-hole-nuggets-of-gold-in-rivers-of-bullshit |
# Technology path dependence and evaluating expertise
I often wonder if I'd be able to tell if a technology were uncompetitive for fundamental viability reasons or for reasons of path dependence—if the ultimate performance would be a winner once we'd worked out a few trillion dollars of kinks, but we never go down tha... | https://www.lesswrong.com/posts/unesQvsKBjx6x2ZkY/technology-path-dependence-and-evaluating-expertise |
# AI Impacts 2023 Expert Survey on Progress in AI
AI Impacts just released the results from their latest survey! I've found this quite interesting to engage with. Here is the results section of the linked wiki page:
* * *
### **Results**
The full dataset of anonymized responses to the survey is available [here](ht... | https://www.lesswrong.com/posts/RkegCmCgjGhskiFvm/ai-impacts-2023-expert-survey-on-progress-in-ai |
# The Next ChatGPT Moment: AI Avatars
**Epistemic Status:** Speculative. Dependent on intuitions about near-term AI tech and human psychology.
**Claim:** Within the next 1-3 years, many people will have an interaction with an AI avatar that feels authentically human. This will significantly amplify the public percept... | https://www.lesswrong.com/posts/DsgHj5hxcPgb6rEnj/the-next-chatgpt-moment-ai-avatars |
# Almost everyone I’ve met would be well-served thinking more about what to focus on
> *Almost everyone I’ve ever met would be well-served by spending more time thinking about what to focus on.* —Sam Altman
In May 2020, we parked two moving trucks in the harbor and carried everything we owned from one to the other. J... | https://www.lesswrong.com/posts/Z55vXvngZxkPdKger/almost-everyone-i-ve-met-would-be-well-served-thinking-more |
# Project ideas: Epistemics
*This is part of a series of lists of projects. The unifying theme is that the projects are not targeted at solving alignment or engineered pandemics but still targeted at worlds where transformative AI is coming in the next 10 years or so. See *[*here*](https://lukasfinnveden.substack.com/... | https://www.lesswrong.com/posts/sYdT44uzMwoNXs2Lq/project-ideas-epistemics-1 |
# Survey of 2,778 AI authors: six parts in pictures
*Crossposted from* [*AI Impacts blog*](https://blog.aiimpacts.org/p/2023-ai-survey-of-2778-six-things)
The 2023 Expert Survey on Progress in AI is [out](https://aiimpacts.org/wp-content/uploads/2023/04/Thousands_of_AI_authors_on_the_future_of_AI.pdf), this time with... | https://www.lesswrong.com/posts/6uJBqeG5ywE6Wcdqw/survey-of-2-778-ai-authors-six-parts-in-pictures |
# AI Risk and the US Presidential Candidates
It's the new year, and the 2024 primaries are approaching, starting with the Iowa Republican causus on January 15. For a lot of people here on LessWrong, the issue of AI risk will likely be an important factor in making a decision. AI hasn't been mentioned much during any o... | https://www.lesswrong.com/posts/GyjFwYs4PSGTcr54q/ai-risk-and-the-us-presidential-candidates |
# The Sequences on YouTube
**tl;dr:** Do you have any thoughts/advice/warnings/ideas for someone who wants to turn the Sequences into a series of YouTube videos?
*Naturally, this post also exists as a* [*video*](https://youtu.be/ROahwRpF4sQ?si=kiC8fWY4TLKbpRom)*.*
**I will be:**
* Turning all 50 Sequence Highlig... | https://www.lesswrong.com/posts/kyfsCezT4ERvpF9Re/the-sequences-on-youtube |
# Defending against hypothetical moon life during Apollo 11
*\[Header image:* [*Photo*](https://history.nasa.gov/ap11ann/kippsphotos/5961.jpg) *of the lunar lander taken during Apollo 11.\]*
In 1969, after successfully bringing men back from landing on the moon, the astronauts, spacecraft, and all the samples from th... | https://www.lesswrong.com/posts/dedyJwQXZxakJLuDF/defending-against-hypothetical-moon-life-during-apollo-11 |
# Towards AI Safety Infrastructure: Talk & Outline
*Thanks to Esben Kran and the whole* [*Alignment Jam*](https://alignmentjam.com/) *team for setting this up.*
**Context**: We recently got the chance to share a bit more about the flavor of research we're particularly excited about at [Straumli AI](https://straumli.a... | https://www.lesswrong.com/posts/dAGAhAopppB33QJnN/towards-ai-safety-infrastructure-talk-and-outline |
# Bayesians Commit the Gambler's Fallacy
***TLDR:** Rational people who start out uncertain about an (in fact independent) causal process and then learn from unbiased data will rule out “streaky” hypotheses more quickly than “switchy” hypotheses. As a result, they’ll commit the **gambler’s fallacy**: expecting the pro... | https://www.lesswrong.com/posts/oZMNoxWsfsHQyzzB7/bayesians-commit-the-gambler-s-fallacy |
# Deceptive AI ≠ Deceptively-aligned AI
**Tl;dr:** A “deceptively-aligned AI” is different from (and much more specific than) a “deceptive AI”. I think this is well-known and uncontroversial among AI Alignment experts, but I see people getting confused about it sometimes, so this post is a brief explanation of how the... | https://www.lesswrong.com/posts/a392MCzsGXAZP5KaS/deceptive-ai-deceptively-aligned-ai |
# Project ideas: Sentience and rights of digital minds
*This is part of a series of lists of projects. The unifying theme is that the projects are not targeted at solving alignment or engineered pandemics but still targeted at worlds where transformative AI is coming in the next 10 years or so. See *[*here*](https://l... | https://www.lesswrong.com/posts/PT3A7DnDiDApkHjw5/project-ideas-sentience-and-rights-of-digital-minds-1 |
# A model of research skill
Doing research means answering questions no one yet knows the answer to. Lots of impactful projects are downstream of being good at this. A good first step is to have a model for what the hard parts of research skill are.
Two failure modes
=================
There are two opposing failure ... | https://www.lesswrong.com/posts/NyGgegFATfsyuGFo4/a-model-of-research-skill |
# Sledding Among Hazards
Some sledding hills are very open, with basically no hazards. As long as you start in a reasonable place there's nothing to run into.
[](https://www.jefftk.com/open-sledding-big.jpg)
*Robbins Farm Park, Arlington MA*
Other hills are not like this... | https://www.lesswrong.com/posts/Brp3uigeytpbzsKcm/sledding-among-hazards |
# Why There Is Hope For An Alignment Solution
*Note: I originally wrote the first draft of this on 2022/04/11 intending to post this to Less Wrong in response to the List of Lethalities post, but wanted to edit it a bit to be more rigorous and ended up taking a very long time before getting around to it. I’m still not... | https://www.lesswrong.com/posts/SahrhfoZpBk2jvrj7/why-there-is-hope-for-an-alignment-solution |
# Reflections on my first year of AI safety research
Last year, I wrote a [post about my upskilling in AI alignment](https://forum.effectivealtruism.org/posts/DnMg5q4Wyuuf99kkX/reflections-on-my-5-month-ai-alignment-upskilling-grant). To this day, I still get people occasionally reaching out to me because of this arti... | https://www.lesswrong.com/posts/EJqGTpinaHN3uoZnR/reflections-on-my-first-year-of-ai-safety-research |
# 2023 Prediction Evaluations
It is that time of the year. One must ask not only whether predictions were right or wrong, whether one won or lost, but what one was and should have been thinking, whether or not good decisions were made, whether the market made sense.
The main subject will be the 2023 ACX Predictions, ... | https://www.lesswrong.com/posts/6o98z3QMAQSkHf3gp/2023-prediction-evaluations |
# Why aren't Yudkowsky & Bostrom getting more attention now?
Now that the idea of AI Existential Risk has gained more acceptability, I am surprised that the thought leaders are not more often mentioned and contacted by AI executives and researchers in the cutting edge (OpenAI, Anthropic, etc.), recognized academic fig... | https://www.lesswrong.com/posts/i2wrQ28HLfAJizQae/why-aren-t-yudkowsky-and-bostrom-getting-more-attention-now |
# When "yang" goes wrong
*(Cross-posted from [my website](https://joecarlsmith.com/2024/01/08/when-yang-goes-wrong). Podcast version [here](https://www.buzzsprout.com/2034731/14271131), or search "Joe Carlsmith Audio" on your podcast app.*
*This essay is part of a series I'm calling "Otherness and control in
the age ... | https://www.lesswrong.com/posts/fPajcaopbQ7pxztf4/when-yang-goes-wrong |
# Hackathon and Staying Up-to-Date in AI
Listen to the first episode of a new sub-series, which will follow my team's process during the Evals Hackathon hosted by [AlignmentJamz](https://alignmentjam.com) in November of 2023. In the end, our submission was titled "Detecting Implicit Gaming through Retrospective Evalua... | https://www.lesswrong.com/posts/MPfPeukbnD2eJQBNX/hackathon-and-staying-up-to-date-in-ai |
# Project ideas: Backup plans & Cooperative AI
*This is part of a series of lists of projects. The unifying theme is that the projects are not targeted at solving alignment or engineered pandemics but still targeted at worlds where transformative AI is coming in the next 10 years or so. See *[*here*](https://lukasfinn... | https://www.lesswrong.com/posts/MJYmWrnpfvPMfmBcH/project-ideas-backup-plans-and-cooperative-ai-1 |
# Is it justifiable for non-experts to have strong opinions about Gaza?
Ok, so here's the prompt that we'll be discussing, which I proposed:
"I am of the opinion that the situation in Gaza is complicated and that a large majority of people are uninformed enough such that anything more than weak opinions is not ju... | https://www.lesswrong.com/posts/jW4AxnxLkM4j8sHet/is-it-justifiable-for-non-experts-to-have-strong-opinions |
# A starter guide for evals
This is a starter guide for model evaluations (evals). Our goal is to provide a general overview of what evals are, what skills are helpful for evaluators, potential career trajectories, and possible ways to start in the field of evals.
Evals is a nascent field, so many of the followi... | https://www.lesswrong.com/posts/2PiawPFJeyCQGcwXG/a-starter-guide-for-evals |
# Hiring decisions are not suitable for prediction markets
In a recent edition of [The Diff](https://www.thediff.co/archive/how-many-trillion-dollar-companies-should-there-be/) ($), Byrne Hobart pointed out [a talk](https://www.youtube.com/watch?v=cCNgBazFnjU) given by Robin Hanson at Manifest. The talk is well worth ... | https://www.lesswrong.com/posts/t3maGeMshAFN3xXcC/hiring-decisions-are-not-suitable-for-prediction-markets |
# Compensating for Life Biases
[. Podcast version [here](https://www.buzzsprout.com/2034731/14279206), or search for "Joe Carlsmith Audio" on your podcast app.*
*This essay is part of a series I'm calling "Otherness and co... | https://www.lesswrong.com/posts/aBmBaL95hrNLvqJse/does-ai-risk-other-the-ais |
# Announcing the Double Crux Bot
Edit (July 18, 2025): **While the Double Crux Bot is no longer actively maintained, the code is open-source and available** [**here!**](https://github.com/sofvanh/double-crux-bot) We phased out the Heroku deployment for the Discord and Slack bot in May 2025. Thanks to everyone who took... | https://www.lesswrong.com/posts/ENgwdgdHZ2a6HTWDL/announcing-the-double-crux-bot |
# 2024 ACX Predictions: Blind/Buy/Sell/Hold
My 2023 ACX predictions showed a clear lack of confidence in taking on the market. I won 30 markets for an average of +185 each, and lost 12 for an average loss of -185 each[1](#footnote-1). When one goes 30-12, hitting a 71% mark versus the about 58% average price initially... | https://www.lesswrong.com/posts/7j9JXpGvXNowCkhdf/2024-acx-predictions-blind-buy-sell-hold |
# Bent or Blunt Hoods?
One of the many new EV companies, [Canoo](https://www.canoo.com/), is making a van:
[](https://www.jefftk.com/canoo-walmart-big.jpg)
Looking at the shape of it, one of the reactions [I saw](https://news.ycombinator.com/item?id=38915311) was:
> This... | https://www.lesswrong.com/posts/ahqynJHXXPoudsJD6/bent-or-blunt-hoods |
# Goodbye, Shoggoth: The Stage, its Animatronics, & the Puppeteer – a New Metaphor
*Thanks to Quentin FEUILLADE--MONTIXI for the discussion in which we came up with this idea together, and for feedback on drafts.*
**TL;DR** A better metaphor for how LLMs behave, how they are trained, and particularly for how to think... | https://www.lesswrong.com/posts/mweasRrjrYDLY6FPX/goodbye-shoggoth-the-stage-its-animatronics-and-the-1 |
# On the Contrary, Steelmanning Is Normal; ITT-Passing Is Niche
Rob Bensinger argues that ["ITT-passing and civility are good; 'charity' is bad; steelmanning is niche"](https://www.lesswrong.com/posts/MdZyLnLHuaHrCskjy/itt-passing-and-civility-are-good-charity-is-bad).
The ITT—Ideological Turing Test—is an exercise i... | https://www.lesswrong.com/posts/jo5Fhkb7escrYE9cC/on-the-contrary-steelmanning-is-normal-itt-passing-is-niche |
# A discussion of normative ethics
Hey Gordon. Thanks for being willing to chat with me about this idea I have of [Reflective Consequentialism](https://www.lesswrong.com/posts/xYYrADpKD22bH6F6o/reflective-consequentialism). I've been wanting to do so for a while now.
To start, let me provide some context regarding wh... | https://www.lesswrong.com/posts/vufuuFcLxjgAtXKFr/a-discussion-of-normative-ethics |
# AI as a natural disaster
*\[This post is not meant for the rationalist audience directly, because it has already internalized this lesson, but is to be used as an analogy in discourse with an unfamiliar crowd. The end goal is to make people *intuitively understand *why AI is extremely risky* whether or not *it has m... | https://www.lesswrong.com/posts/hsLRGwLnmkpHhvMuf/ai-as-a-natural-disaster-Edta |
# Saving the world sucks
I don’t want to save the world. I don’t want to tile the universe with hedonium. I don’t want to be cuckolded by someone else’s pretty network-TV values. I don’t want to do anything I don’t want to do, and I think that’s what (bad) EAs, mother Teresa, and proselytizing Christians all get wrong... | https://www.lesswrong.com/posts/y8sHpeGMRW6YhvYpT/saving-the-world-sucks-rDYh |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.