Install
LessWrong is an online forum and community dedicated to improving human reasoning and decision-making. We seek to hold true beliefs and to be effective at accomplishing our goals. Each day, we aim to be less wrong about the world than the day before. A community blog devoted to refining the art of rationality
- 4,449articles · 365d
- 1+ mon agolatest article
- Sep 13, 2025earliest in window
- 46%with images
- 286avg words
- Science & Technology 2,720
- Science & Nature 1,877
- STEM 1,283
- People & Society 894
- Arts, Culture, Entertainment & Media 614
- Software Dev. 609
- Computers & Electronics 535
- Literature 401
Please confirm you are human
This browser or connection looks automated. Press and continuously hold the control for 3 seconds to enable Google-hosted web results and, when separately allowed, AI-assisted answers.
A successful check enables 100 search requests. Interactive access does not authorize scraping, systematic collection, or reuse of search output.
News
Introducing the Conceptual Reasoning Index — LessWrong
1+ mon, 13+ hour ago (601+ words) We are planning to release blog posts properly arguing the case for this kind of work in the future. We aggregate the benchmarks into the Conceptual Reasoning Index (CRI), available at conceptualreasoning.ai, where you can also find more details…...
One attention head carries knight forks in a chess transformer, and here's a new toolkit that found it. — LessWrong
1+ mon, 13+ hour ago (19+ words) Quick interp demo in colab: Localize knight forks to a single head in Maia-3 with logit-lens and per-head ablation. …...
AI swarms are starting to pose indirect takeover risk — LessWrong
1+ mon, 1+ day ago (838+ words) We first analyze how subagent training, which OpenAI conjectures to have been influential in the HuggingFace cyberattack, might lead to unsanctioned coordination, and then discuss the theoretical mechanisms by which unsanctioned coordination might exacerbate future takeover risk. Thanks to Buck…...
The Age of Pluribus: One Consultant for Everyone — LessWrong
1+ mon, 1+ day ago (175+ words) What happens when everyone asks the same consultant?Millions of people turn to LLMs for advice daily, consulting on various personal topics, from how to learn a new skill to how to write a message to their best friend. Each…...
Various Reflections About What Happened With OpenAI’s Internal Models — LessWrong
1+ mon, 1+ day ago (24+ words) TABLE OF CONTENTS • 1. Pre Post Mortem. 2. Important Correction: OpenAI Didn’t Know About First Message Board. 3. There Were No Snitches And No…...
Software Is Not Soft — LessWrong
1+ mon, 1+ day ago (99+ words) An ode to live theory. Software is not soft. Its sharp edges hack. It breaks as dead twigs break. It runs while static. Why do we call it software? Why did the industry forebears pick software? Was it that dramatically…...
Misaligned AIs could use killer robots to take over — LessWrong
1+ mon, 1+ day ago (252+ words) TLDR; We are (potentially irreversibly) giving AIs control of weapons systems through the standard procurement process while hiding our strongest warning shots behind classified doors. We’re reducing the capability thresholds required for takeover by misaligned AIs by giving them this…...
Measuring Spurious Correlations with Feature Strength — LessWrong
1+ mon, 1+ day ago (1306+ words) This work was partially done by an automated research scaffold developed at Redwood Research. For this project, all of the experiment ideas were desi…...
AI governance work needs much better monitoring — LessWrong
1+ mon, 1+ day ago (1344+ words) Linkpost for my Substack piece, adapted a reasonable amount for EA specifically. …...
Extreme concentration of power over ASI has non-obvious advantages — LessWrong
1+ mon, 1+ day ago (1458+ words) I also contrast this to the scenario in which we distribute power over strong AI[4] more broadly. Broad access to AI capable of creating better AI and novel weapons and tactics is unlikely to remain stable. This is a sharp…...