PODCAST

LessWrong (30+ Karma)

LessWrong

Audio narrations of LessWrong posts.

technology philosophy

All Episodes

1:24

“Predict 2025 AI capabilities by Sunday” by Jonas V,...

LessWrong (30+ Karma) ·

2025/01/15

Until this Sunday, you can submit your 2025 AI predictions at ai2025.org. It's a forecasting survey by AI Digest for the 2025 performance on various AI benchmarks, as well as revenue and public attention. You can share your results in a picture like this one. I personally found it pretty helpful to learn about the different benchmarks, and also to think through my timelines estimates. The survey will close on Sunday, January 19th (anywhere on Earth).If you know any AI public intellectuals or

en-gb

3:04

“Get your math consultations here!” by Gurkenglas

LessWrong (30+ Karma) ·

2025/01/14

This post is mostly an ad to make people click the https://calendly.com/gurkenglas/consultation link starring in my Would you like me to debug your math? post.But writing just an ad feels icky to me, so first: The advice that was most useful to most peoplewas on how to use (graphical) debuggers. The default failure mode I see in a programmer is that they're wrong about what happens in their program. The natural remedy is to increase the bandwidth between the program and the programmer. Many people

en-gb

116:01

“Heritability: Five Battles” by Steven Byrnes

LessWrong (30+ Karma) ·

2025/01/14

Audio note: this article contains 54 uses of latex notation, so the narration may be difficult to follow. There's a link to the original text in the episode description. 0.1 tl;drThis is an opinionated but hopefully beginner-friendly discussion of heritability: what is it, what do we know about it, and how we should think about it? I structure my discussion around five contexts in which people talk about the heritability of a trait or outcome: (Section 1) The context of guessing someone's likely

en-gb

6:31

“How do you deal w/ Super Stimuli?” by Logan Riggs

LessWrong (30+ Karma) ·

2025/01/14

From neel.funI remember watching Youtube videos and thinking "This is the last video, I will quit after this". However, as soon as the video ends, my preferences would suddenly change to wanting to do one more! Many of us understand this false dichotomy: Quit mid-way through the videoQuit after the video ends at 3 amSo I would (sometimes) stop binging videos, but only if I quit mid-way through.Then short-form videos wrecked me.These (tik-toks/shorts/reels) have NO "middle of the video" to have a

en-gb

12:41

“Implications of the inference scaling paradigm for...

LessWrong (30+ Karma) ·

2025/01/14

Scaling inferenceWith the release of OpenAI's o1 and o3 models, it seems likely that we are now contending with a new scaling paradigm: spending more compute on model inference at run-time reliably improves model performance. As shown below, o1's AIME accuracy increases at a constant rate with the logarithm of test-time compute (OpenAI, 2024).OpenAI's o3 model continues this trend with record-breaking performance, scoring: 2727 on Codeforces, which makes it the 175th best competitive programmer on

en-gb

16:47

“Finding Features Causally Upstream of Refusal” by...

LessWrong (30+ Karma) ·

2025/01/14

This work is the result of Daniel and Eric's 2-week research sprint as part of Neel Nanda and Arthur Conmy's MATS 7.0 training phase. Andy was the TA during the research sprint. After the sprint, Daniel and Andy extended the experiments and wrote up the results. A notebook that contains all the analyses is available here. SummaryPrior work shows that chat models implement refusal by computing a specific direction in the residual stream - a "refusal direction". In this work, we investigate how this

en-gb

14:09

“Chance is in the Map, not the Territory” by Daniel...

LessWrong (30+ Karma) ·

2025/01/14

"There's a 70% chance of rain tomorrow," says the weather app on your phone. "There's a 30% chance my flight will be delayed," posts a colleague on Slack. Scientific theories also include chances: “There's a 50% chance of observing an electron with spin up,” or (less fundamental) “This is a fair die — the probability of it landing on 2 is one in six.”We constantly talk about chances and probabilities, treating them as features of the world that we can discover and disagree about. And it seems you

en-gb

22:04

“Why Modelling Multi-Objective Homeostasis Is...

LessWrong (30+ Karma) ·

2025/01/13

I notice that there has been very little if any discussion on why and how considering homeostasis is significant, even essential for AI alignment and safety. Current post aims to begin amending that situation. In this post I will treat alignment and safety as explicitly separate subjects, which both benefit from homeostatic approaches.This text is a distillation and reorganisation of three of my older blog posts at Medium: Making AI less dangerous: Using homeostasis-based goal structures

en-gb

28:25

“Zvi’s 2024 In Movies” by Zvi

LessWrong (30+ Karma) ·

2025/01/13

Now that I am tracking all the movies I watch via Letterboxd, it seems worthwhile to go over the results at the end of the year, and look for lessons, patterns and highlights.Table of Contents The Rating Scale. The Numbers. Very Briefly on the Top Picks and Whether You Should See Them. Movies Have Decreasing Marginal Returns in Practice. Theaters are Awesome. I Hate Spoilers With the Fire of a Thousand Suns. Scott Sumner Picks Great American Movies Then Dislikes Them. I Knew Before the Cards

en-gb

0:51

“How quickly could robots scale up?” by Benjamin_Todd

LessWrong (30+ Karma) ·

2025/01/13

This is a link post.Once robots can do physical jobs, how quickly could they become a significant part of the work force?Here's a couple of fermi estimates, showing (among other things) that converting car factories might be able to produce 1 billion robots per year in under 5 years.Nothing too new here if you've been following Carl Shulman for years, but I thought it could be useful to have a reference article. Please let me know about corrections or other ways to improve the estimates. ---

en-gb

12:46

“Moderately More Than You Wanted To Know: Depressive...

LessWrong (30+ Karma) ·

2025/01/13

Depressive realism is the idea that depressed people have more accurate beliefs than the general population. It's a common factoid in “things I learned” lists, and often posited as a matter of settled science.In this post, I’ll explore whether it's true. Where It BeganThe depressive realism hypothesis was first studied by Lauren Alloy and Lyn Yvonne Abramson, in a paper called Judgment of contingency in depressed and nondepressed students: Sadder but wiser?. Undergraduates had access to a button,

en-gb

5:56

“Applying traditional economic thinking to AGI: a...

LessWrong (30+ Karma) ·

2025/01/13

Traditional economics thinking has two strong principles, each based on abundant historical data: Principle (A): No “lump of labor”: If human population goes up, there might be some wage drop in the very short term, because the demand curve for labor slopes down. But in the longer term, people will find new productive things to do, such that human labor will retain high value. Indeed, if anything, the value of labor will go up, not down—for example, dense cities are engines of economic

en-gb

9:50

“Building AI Research Fleets” by bgold, Jesse Hoogland

LessWrong (30+ Karma) ·

2025/01/12

From AI scientist to AI research fleetResearch automation is here (1, 2, 3). We saw it coming and planned ahead, which puts us ahead of most (4, 5, 6). But that foresight also comes with a set of outdated expectations that are holding us back. In particular, research automation is not just about “aligning the first AI scientist”, it's also about the institution-building problem of coordinating the first AI research fleets.Research automation is not about developing a plug-and-play “AI scientist”

en-gb

11:46

“The purposeful drunkard” by Dmitry Vaintrob

LessWrong (30+ Karma) ·

2025/01/12

I'm behind on a couple of posts I've been planning, but am trying to post something every day if possible. So today I'll post a cached fun piece on overinterpreting a random data phenomenon that's tricked me before.Recall that a random walk or a "drunkard's walk" (as in the title) is a sequence of vectors _x_1, x_2, ldots_ in some _mathbb{R}^n_ such that each _x_k_ is obtained from _x_{k-1}_ by adding

en-gb

1:47

“No one has the ball on 1500 Russian olympiad winners...

LessWrong (30+ Karma) ·

2025/01/12

We have contact details and can send emails to 1500 students and former students who've received hard-cover copies of HPMOR (and possibly Human Compatible and/or The Precipice) because they've won international or Russian olympiads in maths, computer science, physics, biology, or chemistry.This includes over 60 IMO and IOI medalists.This is a pool of potentially extremely talented people, many of whom have read HPMOR.I don't have the time to do anything with them, and people in the Russian-speaking

en-gb

13:22

“Rolling Thresholds for AGI Scaling Regulation” by Larks

LessWrong (30+ Karma) ·

2025/01/12

This is a plan for how ASI could be relatively safely developed. Abstract: A plan that puts all frontier model companies on a unified schedule of model training, evaluation and approval, with regulatory compliance promoted through market access. This aims to combine (most of) the economic benefits of unrestricted competition but with more safety, (most of) the time-to-think benefits of AI pauses but with better compliance incentives, and (most of) the central oversight of a Manhattan project but

en-gb

39:01

“Dmitry’s Koan” by Dmitry Vaintrob

LessWrong (30+ Karma) ·

2025/01/11

-I can be a nightmare conference attendee: I tend to ask nitpicky questions and apply a dose of skepticism to a speaker's claims which is healthy in doing one's own research, but probably not optimal when everyone else is trying to follow a talk. I'm working on being better at this, but for now I blame my background. There is one nitpick that comes up again and again. In fact in one conference I brought it up so often that Jake Mendel coined a term for it: "Dmitry's koan".In koan form, the nitpick

en-gb

6:37

“Last week of the Discussion Phase” by Raemon

LessWrong (30+ Karma) ·

2025/01/11

There's one week left in the Discussion Phase of the 2023 Review. Here's a few things you might want to spend that time doing, in escalating order of effort:Review three posts, without stressing much about the details. Go the Quick Review[1] page, pick some posts you remember and had opinions about. Reread them, and write a review.Think about which posts should be in the Best of LessWrong. Go to the Advanced Review[2] page, look at which posts are currently in the top 50, and at the posts that

en-gb

37:00

“Recommendations for Technical AI Safety Research...

LessWrong (30+ Karma) ·

2025/01/10

Anthropic's Alignment Science team conducts technical research aimed at mitigating the risk of catastrophes caused by future advanced AI systems, such as mass loss of life or permanent loss of human control. A central challenge we face is identifying concrete technical work that can be done today to prevent these risks. Future worlds where our research matters—that is, worlds that carry substantial catastrophic risk from AI—will have been radically transformed by AI development. Much of our work

en-gb

15:32

“Human takeover might be worse than AI takeover” by...

LessWrong (30+ Karma) ·

2025/01/10

Epistemic status -- sharing rough notes on an important topic because I don't think I'll have a chance to clean them up soon. SummarySuppose a human used AI to take over the world. Would this be worse than AI taking over? I think plausibly: In expectation, human-level AI will better live up to human moral standards than a randomly selected human. Because: Humans fall far short of our moral standards. Current models are much more nice, patient, honest and selfless than humans. Though human-level AI

en-gb

4,491 results

LessWrong (30+ Karma)

All Episodes

Similar Podcasts

The Automated Daily

TrendTeller

All On AI

All On AI

Elon Musk Podcast

Stage Zero

Reflections

Livewire Labs

科技早8点

之远不知远