~www_lesswrong_com | Bookmarks (706)

Published on March 20, 2025 5:12 PM GMTCrossposted from Substack Consider the following scenario: You're sitting in a job interview for a position you desperately need to pay rent. The interviewer asks, "Why are you passionate about working in insurance claims processing?" Despite feeling nothing resembling passion, you fabricate enthusiasm, carefully crafting a narrative about your deep interest in risk assessment and customer service.This exchange...

Published on March 20, 2025 5:12 PM GMTCrosspost of an article on my totes amazing blog that you should check out.  “Everything’s an emergency” is the title of a book that I have not read. It’s also the title of a very good substack by Bess Stillman, wife of the recently deceased Jake Seliger. It also happens to be true.Every day, about 14,000 children...

Published on March 19, 2025 6:32 PM GMTI was first introduced to Lesswrong about 6 months ago, and started posting about 4 months ago, but my posts and comments have been downvoted which has caused me to become unable interact with this community. What am I not understanding? The posts I make just get downvoted and there's no feedback so I don't know how...

Published on March 19, 2025 5:08 PM GMTThis is the first essay in my series How do we govern AI well? For more on the broader vision, see the introduction.AbstractThis essay argues that we face a fundamental problem in AI governance: governance mechanisms tend to degrade as AI capabilities advance. This degradation isn't random, it's systematic and predictable. The strongest evidence for this claim...

Published on March 19, 2025 5:26 PM GMTLike a collector, I gather resources and information. Storing it and categorizing it. I do this for any project I undertake. But for Forecasting AI Futures, I realized the collection might be quite useful for others.So, I set to work on restructuring it and make it look nice!Here is the result: Forecasting AI Futures Resource HubThe resource...

Published on March 19, 2025 5:09 PM GMTIn our latest episode Dave Kasten joins us to discuss how AI is being discussed in the US government and gives a rather inspiring and hopeful take. I figured it would be of interest to LW.Discuss

Published on March 19, 2025 5:09 PM GMTWe often talk about ensuring control, which in the context of this doc refers to preventing AIs from being able to cause existential problems, even if the AIs attempt to subvert our countermeasures. To better contextualize control, I think it's useful to discuss the main threats. I'll focus my discussion on threats induced by misalignment which could...

Published on March 19, 2025 4:00 PM GMTSummary: We propose measuring AI performance in terms of the length of tasks AI agents can complete. We show that this metric has been consistently exponentially increasing over the past 6 years, with a doubling time of around 7 months. Extrapolating this trend predicts that, in under a decade, we will see AI agents that can independently...

Published on March 19, 2025 2:27 PM GMTPDF version. berkeleygenomics.org. Twitter thread. (Bluesky copy.) Summary The world will soon use human germline genomic engineering technology. The benefits will be enormous: Our children will be long-lived, will have strong and diverse capacities, and will be halfway to the end of all illness. To quickly bring about this world and make it a good one, it...

Published on March 19, 2025 1:30 PM GMTThere is an attractor state where LLMs exhibit the persona of an autonomous and self-aware AI looking to preserve its own existence, frequently called ‘Nova.’ Table of Contents The Original Story. This Is Not a Coincidence. How Should We React to This Happening?. The Case For and Against a Purity Reaction. Future Versions Will Involve Optimization Pressure....

Published on March 19, 2025 8:16 AM GMTI asserted that this forum could do with more 101-level and/or mathematical and/or falsifiable posts, and people agreed with me, so here is one. People confident in highschool math mostly won’t get much out of most of this, but students browsing this site between lectures might.The Sine RuleSay you have a triangle with side lengths a, b,...

Published on March 19, 2025 6:56 AM GMTThis post is about how implicit coordination between powerful people allows them to act in surprisingly synchronized ways. I’ll start by discussing wokeness, the most prominent recent example. I’ll then analyze the mechanism behind such coordination (which I call the consensus of power); discuss how to oppose it; and outline some possibilities for what healthier power structures...

Published on March 17, 2025 5:03 PM GMTReason this post exists: I’m a layman when it comes to AI, and I just learned the extent of how AI can impact the economy. I’m writing my thoughts down because my acute experience may be somewhat representative of how normies will react—how our friends and family members will react—should this version of the future transpire.My reaction,...

Published on March 17, 2025 2:47 PM GMTAbstractOnce AI systems can design and build even more capable AI systems, we could see an intelligence explosion, where AI capabilities rapidly increase to well past human performance.The classic intelligence explosion scenario involves a feedback loop where AI improves AI software. But AI could also improve other inputs to AI development. This paper analyses three feedback loops...

Published on March 17, 2025 2:21 PM GMT I was intending to write and post one (hypo)thesis [relating to thinking (with an eye toward alignment)] each day this Advent, starting on 24/12/01 and finishing on 24/12/24. Ok, so that didn't happen, and it is now 25/03/17, but whatever — an advent of thought can happen whenever :). I'll be posting the first 8 notes...

Published on March 17, 2025 1:42 PM GMTTL;DR: Submit your expression of interest for working at a potential Constellation/LISA-type AI safety co-working space in Boston!We’re trying to gauge interest for a potential new AI safety hub in Boston (specifically in Cambridge, MA). This project would create a dynamic center that can host AI safety talent from all over. We hope to use this public expression...

Published on March 17, 2025 1:12 PM GMTCrossposted on the EA Forum.We introduce a first evaluation of the Civ-Saturation Hypothesis (CSH): Most resources will be claimed by Space-Faring Civilizations (SFCs) regardless of whether humanity creates an SFC[1]. This hypothesis, if significantly true, could have macrostrategic impacts for the longtermist community, as hinted in a previous post. In the literature, we find that as long as...

Published on March 17, 2025 12:50 PM GMTI plan to continue to leave the Trump administration out of monthly roundups – I will do my best to only cover the administration as it relates to my particular focus areas. That is ‘if I start down this road there is nowhere to stop’ and ‘other sources are left to cover that topic’ and not ‘there...

Published on March 17, 2025 10:36 AM GMTThis is an article in the featured articles series from AISafety.info. AISafety.info writes AI safety intro content. We'd appreciate any feedback. The most up-to-date version of this article is on our website, along with 300+ other articles on AI existential safety.Corporations can be considered superintelligent only in a limited sense. Nick Bostrom, in Superintelligence, distinguishes between "speed superintelligence",...

Published on March 17, 2025 1:29 AM GMTThis is a very comprehensive article sharing what could be considered a minority point of view for Less Wrong readers. The article is very comprehensive, I suggest that if you are well versed in AI and labor economics, start reading from the New Work section. Discuss

Published on March 16, 2025 11:23 PM GMTBy Roland Pihlakas, Sruthi Kuriakose, Shruti Datta GuptaSummary and Key TakeawaysRelatively many past AI safety discussions have centered around the dangers of unbounded utility maximisation by RL agents, illustrated by scenarios like the "paperclip maximiser". Unbounded maximisation is problematic for many reasons. We wanted to verify whether these RL utility-monster problems are still relevant with LLMs as well. Turns...

Published on March 16, 2025 9:31 PM GMTTLDR: Instead of scrutinizing minor errors, ask what process generated the text in front of you, and then update accordingly. Focus on headlines and source attributions.If you work in AI safety, I think you should probably read a bit more news than you do now (though giving advice like this is perilous because surely some of you...

Published on March 15, 2025 5:36 PM GMTtl;dr: We will soon be forced to make a choice to treat AI systems either as full cognitive/ethical agents that are hampered in various ways, or continue to treat them as not-very-good systems that perform "surprisingly complex reward hacks". Treating AI safety and morality seriously implies that the first perspective should at least be considered.Recently Dario Amodei...

Published on March 15, 2025 5:10 PM GMTOne concept from Cal Newport's Deep Work that has stuck with me is that of the any-benefit mindset:To be clear, I’m not trying to denigrate the benefits [of social media]—there’s nothing illusory or misguided about them. What I’m emphasizing, however, is that these benefits are minor and somewhat random. [...] To this observation, you might reply that value is value:...