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
- 4,445articles · 365d
- 1+ mon agolatest article
- Sep 13, 2025earliest in window
- 46%with images
- 286avg words
- Science & Technology 2,717
- Science & Nature 1,874
- STEM 1,282
- People & Society 892
- Arts, Culture, Entertainment & Media 614
- Software Dev. 609
- Computers & Electronics 534
- 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
When (and when not) LLMs can verbalize awareness of J-Space concept injections - Initial results — LessWrong
1+ mon, 1+ day ago (1281+ words) Code for reproduction and cross-model extensions available here. …...
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…...
Claude Opus 5 Just Beat My Text-Based Adventure Game Benchmark — LessWrong
1+ mon, 2+ day ago (564+ words) ---------------------------------------- • First, here are my previous articles on this subject: …...
Misaligned AIs could use killer robots to take over — LessWrong
1+ mon, 2+ 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…...
Extreme concentration of power over ASI has non-obvious advantages — LessWrong
1+ mon, 2+ 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…...
How risky would it be to make powerful AI obey one or a few people? — LessWrong
1+ mon, 2+ day ago (989+ words) It seems fairly likely that the first powerful AIs will be instruction-following rather than value-aligned, and will be controlled by a small number of people. So it makes sense to worry what individual people might do with such immense power....
What Claude Saw Below — LessWrong
1+ mon, 2+ day ago (1801+ words) The first attempt disappointed. I wrote: “see the below —” and hit send. Claude responded: “Nothing arrived on my end: no file, no text, no image. If you want to attach something, try again.” So I did, leaving the prompt unchanged…...
Claude summarizes behavior as significantly less misaligned when the actor is Claude vs another model — LessWrong
1+ mon, 3+ day ago (26+ words) (This is a lower-effort research update. It reflects my current beliefs/understanding, but is less robust than other research I'm working on. It refl…...
Book Review: The Infinity Machine — LessWrong
1+ mon, 3+ day ago (78+ words) It looks like Demis Hassabis is stepping away from Google DeepMind. In honor of his rise and presumed fall, I wrote an essay on the powers and perils of seeing 90% of the future.Linkpost for: https://millicosm.substack.com/p/book-review-the-infinity-machineDiscuss Book Review: The Infinity…...
Is it ethical to work on general-purpose robots given the risk of totalitarianism? — LessWrong
1+ mon, 3+ day ago (540+ words) One potential risk of developing general-purpose robots is that they could greatly reduce the friction required to establish a totalitarian regime. If robots became physically capable of manufacturing additional copies of themselves, a small group of bad actors could potentially…...