PODCAST

LessWrong (30+ Karma)

LessWrong

Audio narrations of LessWrong posts.

technology philosophy

All Episodes

9:35

“Loudly Give Up, Don’t Quietly Fade” by Screwtape

LessWrong (30+ Karma) ·

2023/11/13

1.There's a supercharged, dire wolf form of the bystander effect that I’d like to shine a spotlight on.First, a quick recap. The Bystander Effect is a phenomenon where people are less likely to help when there's a group around. When I took basic medical training, I was told to always ask one specific person to take actions instead of asking a crowd at large. “You, in the green shirt! Call 911!” (911 is the emergency services number in the United States.) One habit I worked hard to instill in my own

en-gb

37:18

[Linkpost] “Theories of Change for AI Auditing” by...

LessWrong (30+ Karma) ·

2023/11/13

Crossposted from the AI Alignment Forum. May contain more technical jargon than usual.Executive summaryOur mission at Apollo Research is to reduce catastrophic risks from AI by auditing advanced AI systems for misalignment and dangerous capabilities, with an initial focus on deceptive alignment.In our announcement post, we presented a brief theory of change of our organization which explains why we expect AI auditing to be strongly positive for reducing catastrophic risk from advanced AI systems.In

en-gb

3:46

“They are made of repeating patterns” by quetzal_rainbow

LessWrong (30+ Karma) ·

2023/11/13

Epistemis status: an obvious parody. — You won't believe me. I've found them.— Whom?— Remember that famous discovery by Professor Prgh'zhyne about pockets of baryonic matter in open systems that minimize the production of entropy within them? They went further and claimed that goal-oriented systems could emerge within these pockets. Crazy idea, but... it seems I've found them near this yellow dwarf!— You're kidding. We know that a good optimizer of outcomes over systems' states should have a model

en-gb

26:34

“Redirecting one’s own taxes as an effective altruism...

LessWrong (30+ Karma) ·

2023/11/13

About twenty years ago, I stopped paying U.S. federal income taxes. By law, the government has ten years to collect an unpaid tax bill, whereafter a sort of statute of limitations kicks in and the bill becomes permanently noncollectable. I’ve adopted the practice of waiting out this ten-year period and then donating the amount of the uncollected tax to charity, typically the Top Charities Fund organized by GiveWell. Over the past six years I’ve redirected over $30,000 from the U.S. Treasury to

en-gb

1:35

“You can just spontaneously call people you haven’t...

LessWrong (30+ Karma) ·

2023/11/13

Here's a conversation I had relatively recently:Me: "I wish I had more friends. You guys are great, but I only get to hang out with you like once or twice a week. It's really painful being holed up in my house the entire rest of the time."Friend: "You know ${X}. You could talk to him."Me: "I haven't talked to ${X} since 2019."Friend: "Why does that matter? Just call him."Me: "What do you mean 'just call him'? I can't do that."Friend: "Yes you can"Me: Later: I call ${X}, we talk for an hour and a

en-gb

5:07

“Zvi’s Manifold Markets House Rules” by Zvi

LessWrong (30+ Karma) ·

2023/11/13

All markets created by Zvi Mowshowitz shall be graded according to the rules described herein, including the zeroth rule.The version of this on LessWrong shall be the canonical version, even if other versions are later posted on other websites.Rule 0: If the description of a particular market contradicts these rules, the market's description wins, the way a card in Magic: The Gathering can break the rules. This document only establishes the baseline rules, which can be modified.Effort put into the

en-gb

31:44

[Linkpost] “Don’t Donate A Kidney To A Stranger” by...

LessWrong (30+ Karma) ·

2023/11/12

Donate a kidney to a stranger, is a battle cry picking up some fervor in EA circles. The stem of this seems to be certainty in a flawed understanding of medical research, combined with conviction about a rather acrobatic view of morality.My argument against donation is split into 3 parts, with the first being by far the most important:Why kidney donations do great harm to the donorWhy kidney donations are ethically fuzzy and might be a net negativeWhy the desire to donate a kidney is likely

en-gb

2:21

“The Fundamental Theorem for measurable factor...

LessWrong (30+ Karma) ·

2023/11/12

I present the fundamental theorem for all finitely factored measurable spaces. To understand the result, it is necessary to understand the definition of the history in this setting. All the maths is in this document. The core idea is to express mathematically that the history tells us, when the conditional probability of an event depends on which factor. I show that the history can be expressed mathematically exactly as this and show that this representation can be used to deduce that structural

en-gb

19:23

“Game Theory without Argmax [Part 1]” by Cleo Nardo

LessWrong (30+ Karma) ·

2023/11/11

Written during the SERI MATS program under the joint mentorship of John Wentworth, Nicholas Kees, and Janus.PrefaceIn classical game theory, we characterise agents by a utility function and assume that agents choose options which cause maximal utility. This is a pretty good model, but it has some conceptual and empirical limitations which are particularly troublesome for AI safety.Higher-order game theory (HOGT) is an attempt to rebuild game theory without appealing to either utility functions or

en-gb

9:39

“It’s OK to be biased towards humans” by dr_s

LessWrong (30+ Karma) ·

2023/11/11

Let's talk about art.In the wake of AI art generators being released, it's become pretty clear this will have a seismic effect on the art industry all across - from illustrators, to comic artists, to animators, many categories see their livelihood threatened, with no obvious "higher level" opened by this wave of automation for them to move to. On top of this, the AI generators seem to have mostly been trained with material whose copyright status is... dubious, at the very least. Images have been

en-gb

5:53

“Palisade is hiring Research Engineers” by Charlie...

LessWrong (30+ Karma) ·

2023/11/11

Palisade is looking to hire two Research Engineers. We are a small team consisting of Jeffrey Ladish (Executive Director), Charlie Rogers-Smith (Chief of Staff), and Kyle Scott (part-time Treasurer & Operations). In joining Palisade, you would be a founding member of the team, and would have substantial influence over our strategic direction. Applications are rolling, and you can fill out our short (~10-20 minutes) application form here. Palisade's missionWe research dangerous AI capabilities to

en-gb

15:23

[Linkpost] “Who is Sam Bankman-Fried (SBF) really,...

LessWrong (30+ Karma) ·

2023/11/11

As you may know, Sam Bankman-Fried ("SBF") was convicted of seven counts of fraud and conspiracy. He now faces the potential of more than 100 years in prison.I've been trying to figure out how someone who appears to believe deeply in the principles of effective altruism could do what SBF did. It has been no surprise to me to see that the actions he was convicted of are nearly universally condemned by the EA community. Could it be that he did not actually believe in EA ideas despite promoting EA and

en-gb

8:55

“We have promising alignment plans with low taxes” by...

LessWrong (30+ Karma) ·

2023/11/10

Epistemic status: I’m sure these plans have advantages relative to other plans. I'm not sure they're adequate to actually work, but I think they might be.With good enough alignment plans, we might not need coordination to survive. If alignment taxes are low enough, we might expect most people developing AGI to adopt them voluntarily. There are two alignment plans that seem very promising to me, based on several factors, including ease of implementation, and applying to fairly likely default paths to

en-gb

7:05

“Picking Mentors For Research Programmes” by Raymond D

LessWrong (30+ Karma) ·

2023/11/10

Several programmes right now offer people some kind of mentor or supervisor for a few months of research. I participated in SERI MATS 4.0 over the summer, and I saw just how different people's experiences were of being mentored. So this is my list of dimensions where I think mentors at these programmes can differ a lot, and where the differences can really affect people's experiences.When you pick a mentor, you are effectively trading between these dimensions. It's good to know which ones you care

en-gb

26:10

“GPT-2030 and Catastrophic Drives: Four Vignettes” by...

LessWrong (30+ Karma) ·

2023/11/10

I previously discussed the capabilities we might expect from future AI systems, illustrated through GPT2030, a hypothetical successor of GPT-4 trained in 2030. GPT2030 had a number of advanced capabilities, including superhuman programming, hacking, and persuasion skills, the ability to think more quickly than humans and to learn quickly by sharing information across parallel copies, and potentially other superhuman skills such as protein engineering. I’ll use “GPT2030++” to refer to a system that

en-gb

98:02

“AI Timelines” by habryka, Daniel Kokotajlo, Ajeya...

LessWrong (30+ Karma) ·

2023/11/10

IntroductionHow many years will pass before transformative AI is built? Three people who have thought a lot about this question are Ajeya Cotra from Open Philanthropy, Daniel Kokotajlo from OpenAI and Ege Erdil from Epoch. They have each spent hundreds of hours researching and forecasting when we might see very large impacts from AI development, but they substantially disagree about the relevant timescales. For instance, here are their median timelines for one operationalization of transformative

en-gb

18:11

“Text Posts from the Kids Group: 2021” by jefftk

LessWrong (30+ Karma) ·

2023/11/09

Another round of liberating kid posts from Facebook. For reference, in 2021 Lily turned 7, Anna turned 5, and Nora was born. (Some of these were from me; some were from Julia. Ones saying "me" could mean either of us.) 2021-01-02 Anna: Hello, I'm Mr. Hamburger. Me: It's time to brush teeth, Mr. Hamburger. Anna: I can't brush my teeth, I'm a hamburger. Me: It's still time to brush teeth. Anna: Hamburgers don't have teeth. 2021-01-04 "Anna, try bonking your head into the

en-gb

137:14

“AI #37: Moving Too Fast” by Zvi

LessWrong (30+ Karma) ·

2023/11/09

We had OpenAI's dev day, where they introduced a host of new incremental feature upgrades including a longer context window, more recent knowledge cutoff, increased speed, seamless feature integration and a price drop. Quite the package. On top of that, they introduced what they call ‘GPTs’ that can let you configure a host of things to set up specialized proto-agents or widgets that will work for specialized tasks and be shared with others. I would love to mess around with that, once I have the

en-gb

4:12

“Learning-theoretic agenda reading list” by Vanessa...

LessWrong (30+ Karma) ·

2023/11/09

Crossposted from the AI Alignment Forum. May contain more technical jargon than usual.Recently, I'm receiving more and more requests for a self-study reading list for people interested in the learning-theoretic agenda. I created a standard list for that, but before now I limited myself to sending it to individual people in private, out of some sense of perfectionism: many of the entries on the list might not be the best sources for the topics and I haven't read all of them cover to cover myself

en-gb

13:21

“Polysemantic Attention Head in a 4-Layer...

LessWrong (30+ Karma) ·

2023/11/09

Produced as a part of MATS Program, under @Neel Nanda and @Lee Sharkey mentorshipEpistemic status: optimized to get the post out quickly, but we are confident in the main claimsTL;DR: head 1.4 in attn-only-4l exhibits many different attention patterns that are all relevant to model's performanceIntroductionInprevious post about the docstring circuit, we found that attention head 1.4 (Layer 1, Head 4) in a4-layer attention-only transformer would act as either afuzzy previous token head or as

en-gb

4,491 results

LessWrong (30+ Karma)

All Episodes

Similar Podcasts

Lead With AI

Tamara Nall

Double Tap

Double Tap Productions Inc.

Tech Radar Voice

Tikal Knowledge

Tech Updates

Meir Tsvi

Fuckupy v IT

Matěj Kotrba, KAIZEMA