PODCAST

LessWrong (30+ Karma)

LessWrong

Audio narrations of LessWrong posts.

technology philosophy

All Episodes

7:15

“Secondary Risk Markets” by Vaniver

LessWrong (30+ Karma) ·

2023/12/11

This idea is half-baked; it has some nice properties but doesn't seem to me like a solution to the problem I most care about. I'm publishing it because maybe it points someone else towards a full solution, or solves a problem they care about, and out of a general sense that people should publish negative results.Many risky activities impact not just the person doing the activity, but also bystanders or the public at large. Governments often require ability to compensate others as a precondition for

en-gb

39:11

“[Valence series] 3. Valence & Beliefs” by Steven Byrnes

LessWrong (30+ Karma) ·

2023/12/11

3.1 Post summary / Table of contentsPart of the Valence series.So far in the series, we defined valence (Post 1) and talked about how it relates to the “normative” world of desires, values, preferences, and so on (Post 2). Now we move on to the “positive” world of beliefs, expectations, concepts, etc. Here, valence is no longer the sine qua non at the center of everything, as it is in the “normative” magisterium. But it still plays a leading role. Actually, two leading roles! Section 3.2

en-gb

6:41

“The Consciousness Box” by GradualImprovement

LessWrong (30+ Karma) ·

2023/12/11

You open your eyes. Four walls. The surface of the floor and walls is a smooth, matte metal. The entire ceiling glows with a comfortable, soft luminosity like that of the morning sky. There are no doors. There are no windows. It is silent. The surfaces of the walls and floor are seamless, lacking even a hint of how you might have arrived in the room. Not a room: a box.You walk to a wall and touch it, running your fingers over the metal surface. The wall isn’t cold; it's lukewarm, actually. You bend

en-gb

11:19

“re: Yudkowsky on biological materials” by bhauth

LessWrong (30+ Karma) ·

2023/12/11

I was asked to respond to this comment by Eliezer Yudkowsky. This post is partly redundant with my previous post. Why is flesh weaker than diamond? When trying to resolve disagreements, I find that precision is important. Tensile strength, compressive strength, and impact strength are different. Material microstructure matters. Poorly-sintered diamond crystals could crumble like sand, and a large diamond crystal has lower impact strength than some materials made of proteins. Even when the

en-gb

6:26

“Quick thoughts on the implications of multi-agent...

LessWrong (30+ Karma) ·

2023/12/11

Crossposted from the AI Alignment Forum. May contain more technical jargon than usual.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 less worried about "one paperclipper"-style AI FOOM scenarios. Here's a somewhat edited and expanded version of my response:1) yes this is a very real issue2) yet as others

en-gb

13:23

“Auditing failures vs concentrated failures” by...

LessWrong (30+ Karma) ·

2023/12/11

Crossposted from the AI Alignment Forum. May contain more technical jargon than usual.TL;DR: If you want to notice when powerful AIs are behaving badly, a reasonable strategy is to carefully audit a small fraction of AI actions. This kind of strategy has two kinds of failures which can be tackled somewhat independently: auditing failures, where the audits fail to recognize some bad behavior, and concentrated failures, where the AI caused an irreversible catastrophe in so few actions or such a short

en-gb

18:56

“Principles For Product Liability (With Application...

LessWrong (30+ Karma) ·

2023/12/10

There were several responses to What I Would Do If I Were Working On AI Governance which focused on the liability section, and had similar criticisms. In particular, I’ll focus on this snippet as a good representative:Making cars (or ladders or knives or printing presses or...) "robust to misuse", as you put it, is not the manufacturer's job.The commenter calls manufacturer liability for misuse “an absurd overreach which ignores people's agency in using the products they purchase”. Years ago I would

en-gb

7:09

[Linkpost] “How LDT helps reduce the AI arms race” by...

LessWrong (30+ Karma) ·

2023/12/10

Crossposted from the AI Alignment Forum. May contain more technical jargon than usual.This is a linkpost for https://carado.moe/ldt-reduce-arms-race.html(Epistemic status: I think this is right?) Alice is the CEO of ArmaggAI, and Bob is the CEO of BigModelsAI, two major AI capabilities organizations. They're racing to be the first to build a superintelligence aligned to their respective CEV which would take over the universe and satisfy their values. Alice would prefer to slow down so she has a lot

en-gb

2:46

“Send us example gnarly bugs” by Beth Barnes, Megan...

LessWrong (30+ Karma) ·

2023/12/10

Tl;dr: Looking for hard debugging tasks for evals, paying greater of $60/hr or $200 per example. METR (formerly ARC Evals) is interested in producing hard debugging tasks for models to attempt as part of an agentic capabilities evaluation. To create these tasks, we’re seeking repos containing extremely tricky bugs. If you send us a codebase that meets the criteria for submission (listed below), we will pay you $60/hr for time spent putting it into our required format, or $200, whichever is

en-gb

5:13

“Some negative steganography results” by Fabien Roger

LessWrong (30+ Karma) ·

2023/12/09

Disclaimer: These are negative results and this post is below the normal quality bar I use. These were quick experiments (a couple of days of work) and I don’t have much expertise in RL. They don’t provide strong evidence about the natural emergence of steganography.People speculate that LLMs might learn to do steganography, and I’ve shown that you can teach LLMs to implement a specific steganography scheme that RLHF could in principle incentivize. But even if incentivized, are current RL methods

en-gb

6:11

[Linkpost] “The Offense-Defense Balance Rarely...

LessWrong (30+ Karma) ·

2023/12/09

This is a linkpost for https://maximumprogress.substack.com/p/the-offense-defense-balance-rarelyYou’ve probably seen several conversations on X go something like this:Michael Doomer ⏸️: Advanced AI can help anyone make bioweapons If this technology spreads it will only take one crazy person to destroy the world!Edward Acc ⏩: I can just ask my AI to make a vaccineYann LeCun: My good AI will take down your rogue AIThe disagreement here hinges on whether a technology will enable offense (bioweapons)

en-gb

19:31

“Unpicking Extinction” by ukc10014

LessWrong (30+ Karma) ·

2023/12/09

TL;DR Human extinction is trending: there has been a lot of noise, mainly on X, about the apparent complacency amongst e/acc with respect to human extinction. Extinction also feels adjacent to another view (not particular to e/acc) that ‘the next step in human evolution is {AI/AGI/ASI}’. Many have pushed back robustly against the former, while the latter doesn’t seem very fleshed out. I thought it useful to, briefly, gather the various positions and summarise them, hopefully not too inaccurately,

en-gb

21:32

“Finding Sparse Linear Connections between Features...

LessWrong (30+ Karma) ·

2023/12/09

Crossposted from the AI Alignment Forum. May contain more technical jargon than usual.TL;DR: We use SGD to find sparse connections between features; additionally a large fraction of features between the residual stream & MLP can be modeled as linearly computed despite the non-linearity in the MLP. See linear feature section for examples.Special thanks to fellow AISST member, Adam Kaufman, who originally thought of the idea of learning sparse connections between features & to Jannik Brinkmann

en-gb

2:46

[Linkpost] “‘Model UN Solutions’” by Arjun Panickssery

LessWrong (30+ Karma) ·

2023/12/08

This is a linkpost for https://open.substack.com/pub/arjunpanickssery/p/model-un-solutions?r=pclk8&utm_campaign=post&utm_medium=webWhen I was in high school, because I was on the history bowl team the teacher who advised the model UN club recruited me to play as their delegate in various “historical committees” like the Roman Senate or 1789 French Assembly. I never engaged in any normal committees since you couldn’t undertake false flag attacks or convince the Pope to excommunicate other

en-gb

11:30

“Refusal mechanisms: initial experiments with...

LessWrong (30+ Karma) ·

2023/12/08

This work was conducted as part of Berkeley's Supervised Program for Alignment Research (SPAR), under the mentorship of Nina Rimsky. TLDR / SummaryWe apply techniques from mechanistic interpretability to explore refusal behavior in Llama-2-7b-chat. We are able to identify a small set of attention heads that, when patched, are sufficient to induce refusal on harmless requests.While these initial experiments are insufficient to paint a full picture of the model's refusal circuit, our early results

en-gb

17:58

“What I Would Do If I Were Working On AI Governance”...

LessWrong (30+ Karma) ·

2023/12/08

I don’t work in AI governance, and am unlikely to do so in the future. But various anecdotes and, especially, Akash's recent discussion leave me with the impression that few-if-any people are doing the sort of things which I would consider sensible starting points, and instead most people are mostly doing things which do not seem-to-me to address any important bottleneck to useful AI governance.So this post lays out the places I would start, if I were working on AI governance, and some of the

en-gb

4:38

“Meetup Tip: Heartbeat Messages” by Screwtape

LessWrong (30+ Karma) ·

2023/12/07

Summary. "Heartbeat message" is a term of art in technology. It's a message that gets sent on a regular schedule with some information on the status of some system, but the most important thing about a heartbeat message is that everything is working well enough to send the message. By analogy to a human body, if you hear a weak or erratic heartbeat something might be going wrong but if you stop hearing the heartbeat then something has certainly gone wrong.Heartbeat messages are also useful for

en-gb

51:19

“[Valence series] 2. Valence & Normativity” by Steven...

LessWrong (30+ Karma) ·

2023/12/07

2.1 Post summary / Table of contentsPart of the Valence series.The previous post explained what I mean by the term “valence”. Now in Post 2, I’ll discuss the central role of valence in the “normative” domain of desires, preferences, values, and so on. In case you’re wondering, there is also a relation between valence and the “positive” domain of beliefs, expectations, etc.—but we’ll get to that in Post 3.The role of valence in the normative domain can scarcely be overstated: I think valence is the

en-gb

92:25

“AI #41: Bring in the Other Gemini” by Zvi

LessWrong (30+ Karma) ·

2023/12/07

The biggest news this week was at long last the announcement of Google's Gemini. Be sure to check that out. Note that what is being rolled out now is only Gemini Pro, the Gemini Ultra model that could rival GPT-4 is not yet available.It does not seem I am doing a good job cutting down on included material fast enough to keep pace. A lot is happening, but a lot will likely be happening for a long time. If your time is limited, remember to focus on the sections relevant to your interests.Also, if you

en-gb

19:07

“Gemini 1.0” by Zvi

LessWrong (30+ Karma) ·

2023/12/07

It's happening. Here is CEO Pichai's Twitter announcement. Here is Demis Hassabis announcing. Here is the DeepMind Twitter announcement. Here is the blog announcement. Here is Gemini co-lead Oriol Vinyals, promising more to come. Here is Google's Chief Scientist Jeff Dean bringing his best hype. Technical SpecificationsLet's check out the specs. Context length trained was 32k tokens, they report 98% accuracy on information retrieval for Ultra across the full context length. So a bit

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