Everything you care about in one place
Follow feeds: blogs, news, RSS and more. An effortless way to read and digest content of your choice.
Get Feederlesswrong.com
LessWrong
Get the latest updates from LessWrong directly as they happen.
Follow now 115 followers
Latest posts
Last updated 1 day ago
Surprising facts about the slave trade
7 days ago
1. The obstacle to abolition was not the economic system, but an...
A Mechanistic Explanation of Prompt Injection (and why you should study roles)
12 days ago
SummaryWe've been building a theory of how prompt injections work under the...
The LLM shoggoth meme is weirder than you think
16 days ago
This article contains spoilers for At the Mountains of Madness, The Case...
Who Got Breasts First and How We Got Them
22 days ago
It really is Sydney Sweeney’s world, and we’re all just living in...
Trees are mostly made of air and a generalizable lesson for AI safety
26 days ago
At the risk of embarrassing myself, I’ll share a confession.For context, I...
Guardian Angels: LLM Personalization for Productivity and Security
30 days ago
Powerful LLMs will be deployed at global scale in the next few...
Estimating No-CoT Task-Completion Time Horizons of Frontier AI Models
about 1 month ago
(see full author list at the end)About a year ago, METR showed...
x-risk-themed
about 1 month ago
Sometimes, a friend who works around here, at an x-risk-themed organisation, will...
Mnemonic portraits for 19,023 human genes
about 1 month ago
Back in 2013, Scott Alexander wrote in Extreme mnemonics:JS-154 is one of...
Models finding software vulnerabilities is not the primary source of cybersecurity risk
about 2 months ago
I have tried and failed to write a longer post since 2024...
Empowerment, corrigibility, etc. are simple abstractions (of a messed-up ontology)
about 2 months ago
1.1 Tl;drAlignment is often conceptualized as AIs helping humans achieve their goals...