PODCAST

LessWrong (30+ Karma)

LessWrong

Audio narrations of LessWrong posts.

technology philosophy

All Episodes

6:42

[Linkpost] “Sam Altman’s Chip Ambitions Undercut...

LessWrong (30+ Karma) ·

2024/02/10

This is a linkpost for https://garrisonlovely.substack.com/p/sam-altmans-chip-ambitions-undercut If you enjoy this, please consider subscribing to my Substack. Sam Altman has said he thinks that developing artificial general intelligence (AGI) could lead to human extinction, but OpenAI is trying to build it ASAP. Why? The common story for how AI could overpower humanity involves an “intelligence explosion,” where an AI system becomes smart enough to further improve its capabilities, bootstrapping

en-gb

19:42

“One True Love” by Zvi

LessWrong (30+ Karma) ·

2024/02/09

We have long been waiting for a version of this story, where someone hacks together the technology to use Generative AI to work the full stack of the dating apps on their behalf, ultimately finding their One True Love. Or at least, we would, if it turned out he is Not Making This Up. Fun question: Given he is also this guy, does that make him more or less credible? Alas, something being Too Good to Check does not actually mean one gets to not check it, in my case via a Manifold

en-gb

15:13

“Skills I’d like my collaborators to have” by Raemon

LessWrong (30+ Karma) ·

2024/02/09

I originally wrote this in 2021, when Lightcone was exploring a vision of building "a rationalist campus" or maybe "a rationalist society."Right now, neither I, nor Lightcone as a whole, are focused on this in the same way. You *might* make a guild that taught these skills on purpose, or a barrier-to-entry that filtered for them. But, I mostly am publishing it now because I still roughly like it as an aspirational world I'd like to be a part of some day. I like these skills, and I'd like my

en-gb

6:37

“Transfer learning and generalization-qua-capability...

LessWrong (30+ Karma) ·

2024/02/09

Tl/Dr: Generalisation is a capability in its own right, and in our experiments it scales less quickly than specific capabilities.Our experiments show that on a range of tasks, finetuning Babbage on task A transfers more to task B than finetuning Davinci on task A transfers to task B. This is relevant for identifying the nature of the alignment problem we might encounter (for things in this vague cluster seehere,here, andhere).Of independent interest is the fact that a transfer learning metric can

en-gb

7:43

“Running the Numbers on a Heat Pump” by jefftk

LessWrong (30+ Karma) ·

2024/02/09

I recently looked into installing a heat pump mini-split system to heat and cool our house. While it's expensive to install, I had thought a combination of it being nicer than what we currently have and cheaper to operate would make it worth it. Unfortunately, after running the numbers on our utility costs it turns out it would be substantially more expensive to power than our current gas system. I got excited about a potential heat pump system after learning that over the past few years

en-gb

22:23

“Updatelessness doesn’t solve most problems” by...

LessWrong (30+ Karma) ·

2024/02/08

In some discussions (especially about acausal trade and multi-polar conflict), I’ve heard the motto “X will/won’t be a problem because superintelligences will just be Updateless”. Here I’ll explain (in layman's terms) why, as far as we know, it's not looking likely that a super satisfactory implementation of Updatelessness exists, nor that superintelligences automatically implement it, nor that this would drastically improve multi-agentic bargaining.Epistemic status: These insights seem like the

en-gb

41:11

“AI #50: The Most Dangerous Thing” by Zvi

LessWrong (30+ Karma) ·

2024/02/08

In a week with two podcasts I covered extensively, I was happy that there was little other news. That is, until right before press time, when Google rebranded Bard to Gemini, released an app for that, and offered a premium subscription ($20/month) for Gemini Ultra. Gemini Ultra is Here I have had the honor and opportunity to check out Gemini Advanced before its release. The base model seems to be better than GPT-4. It seems excellent for code, for explanations and answering

en-gb

21:54

“Believing In” by AnnaSalamon

LessWrong (30+ Karma) ·

2024/02/08

“In America, we believe in driving on the right hand side of the road.”Tl;dr: Beliefs are like bets (on outcomes the belief doesn’t affect). “Believing in”s are more like kickstarters (for outcomes the believing-in does affect).Epistemic status: New model; could use critique.In one early CFAR test session, we asked volunteers to each write down something they believed. My plan was that we would then think together about what we would see in a world where each belief was true, compared to a world

en-gb

13:41

[Linkpost] “A Chess-GPT Linear Emergent World...

LessWrong (30+ Karma) ·

2024/02/08

This is a linkpost for https://adamkarvonen.github.io/machine_learning/2024/01/03/chess-world-models.html A Chess-GPT Linear Emergent World RepresentationIntroduction. Among the many recent developments in ML, there were two I found interesting and wanted to dig into further. The first was gpt-3.5-turbo-instruct's ability to play chess at 1800 Elo. The fact that an LLM could learn to play chess well from random text scraped off the internet seemed almost magical. The second was Kenneth Li's

en-gb

10:30

“Nitric oxide for covid and other viral infections”...

LessWrong (30+ Karma) ·

2024/02/07

Epistemic status: I spent about 5 hours looking into this, and the next day developed covid myself. I did a bit more research plus all of the writing while sick. So in addition to my normal warning that I have no medical credentials, you should keep in mind that this knowledge may be cursed. Introduction. Nitric Oxide Nasal Spray, sold under the brand name Enovid, is a reactive compound that kills viruses (and I suspect taxes your nasal tissue). It has recently been tested and marketed for

en-gb

17:33

[Linkpost] “Debating with More Persuasive LLMs Leads...

LessWrong (30+ Karma) ·

2024/02/07

Crossposted from the AI Alignment Forum. May contain more technical jargon than usual.This is a linkpost for https://github.com/ucl-dark/llm_debate/blob/main/paper.pdfWe've just completed a bunch of empirical work on LLM debate, and we're excited to share the results. If the title of this post is at all interesting to you, we recommend heading straight to the paper. There are a lot of interesting results that are hard to summarize, and we think the paper is quite readable. If you're pressed for

en-gb

2:20

[Linkpost] “More Hyphenation” by Arjun Panickssery

LessWrong (30+ Karma) ·

2024/02/07

This is a linkpost for https://arjunpanickssery.substack.com/p/more-hyphenation“MAN EATING PIRANHA MISTAKENLY SOLD AS PET FISH” — example news headline from Steven Pinker's The Sense of Style The rule is that you use hyphens for compound modifiers like the ones in natural-language processing, high-impact opportunities, cost-effectiveness measures, high-status employers, and so on. Don’t break up compound proper nouns (“New York-based company”) and don’t use them after adverbs ending in -ly but do

en-gb

2:18

“Why I think it’s net harmful to do technical safety...

LessWrong (30+ Karma) ·

2024/02/07

IMO it is harmful on expectation for a technical safety researcher to work at DeepMind, OpenAI or Anthropic. Four reasons: Interactive complexity. The intractability of catching up – by trying to invent general methods for AI corporations to somehow safely contain model interactions, as other engineers scale models' combinatorial complexity and outside connectivity. Safety-capability entanglements Commercialisation. Model inspection and alignment techniques can support engineering and

en-gb

6:58

“story-based decision-making” by bhauth

LessWrong (30+ Karma) ·

2024/02/07

A few times, I've talked to an executive manager or early-stage investor, and this happened: me: Here's the main plan. Now, we think the odds are good, but the most likely failure point is here. If necessary, we have an alternative plan for that part, which goes as follows... them: (visible disgust) I was so confused! Aren't contingency plans good to have? Sure, investors want to see confidence, but what they really want is confidence in the overall vision. They expect some things to go wrong

en-gb

3:10

“How to train your own ‘Sleeper Agents’” by evhub

LessWrong (30+ Karma) ·

2024/02/07

Crossposted from the AI Alignment Forum. May contain more technical jargon than usual.This post is a guide on how to replicate our "Sleeper Agents" work. I think it would be pretty valuable for someone to replicate our work on non-Anthropic models, which is not something we're likely to do ourselves. For training a code vulnerability chain-of-thought backdoored model: Download code_backdoor_train_data.jsonl from our GitHub repository. This is all the code fine-tuning data you will need, but you

en-gb

31:02

“My guess at Conjecture’s vision: triggering a...

LessWrong (30+ Karma) ·

2024/02/06

ContextThe first version of this document was originally written in the summer of 2023 for my own sake while interning at Conjecture and shared internally. It was written as an attempt to pass an ideological Turing test. I am now posting an edited version of it online after positive feedback from Conjecture. I left Conjecture in August, but I think the doc still roughly reflects the current Conjecture's strategy. Members of Conjecture left comments on a draft of this post.Reflecting on this doc 6

en-gb

10:21

“what does davidad want from boundaries?” by...

LessWrong (30+ Karma) ·

2024/02/06

ChipmonkAs the Conceptual Boundaries Workshop (link) is coming up, and now that we're also planning Mathematical Boundaries Workshop in April, I want to get more clarity on what exactly it is that you want out of boundaries.So I just want to check: Is your goal with boundaries just to formalize a moral thing? I'll summarize what I mean by that: Claim 1: By "boundaries", you mean "the boundaries around moral patients— namely humans". Claim 1b: And to some degree also the boundaries around plants and

en-gb

26:03

“Preventing exfiltration via upload limits seems...

LessWrong (30+ Karma) ·

2024/02/06

Crossposted from the AI Alignment Forum. May contain more technical jargon than usual.At some point in the future, AI developers will need to ensure that when they train sufficiently capable models, the weights of these models do not leave the developer's control. Ensuring that weights are not exfiltrated seems crucial for preventing threat models related to both misalignment and misuse. The challenge of defending model weights has previously been discussed in a RAND report.In this post, I’ll

en-gb

108:51

“On the Debate Between Jezos and Leahy” by Zvi

LessWrong (30+ Karma) ·

2024/02/06

Previously: Based Beff Jezos and the Accelerationists Based Beff Jezos, the founder of effective accelerationism, delivered on his previous pledge, and did indeed debate what is to be done to navigate into the future with a highly Worthy Opponent in Connor Leahy. The moderator almost entirely stayed out of it, and intervened well when he did, so this was a highly fair arena. It's Jezos versus Leahy. Let's get ready to rumble! I wanted to be sure I got the arguments right and fully

en-gb

11:15

“Toy models of AI control for concentrated...

LessWrong (30+ Karma) ·

2024/02/06

In this post, we analyze some simple games that model the problem of ensuring AIs never succeed at taking a single catastrophic action (which we'll refer to as "escaping") during deployment. This exercise highlights some of the difficulties associated with deploying misaligned models given a limited perfect auditing budget, and also suggests some opportunities for technique development. We analyzed these games while writing AI Control: Improving Safety Despite Intentional Subversion. (Thanks to

en-gb

4,491 results

LessWrong (30+ Karma)

All Episodes

Similar Podcasts

Lead With AI

Tamara Nall

Reflections

Livewire Labs

ForGeeks

ForGeeks

FOX on Tech

FOX News Podcasts

The Linux Cast

The Linux Cast