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
- 4,447articles · 365d
- 1+ mon agolatest article
- Sep 13, 2025earliest in window
- 46%with images
- 286avg words
- Science & Technology 2,719
- Science & Nature 1,876
- STEM 1,282
- People & Society 893
- 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
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…...
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…...
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…...
How risky would it be to make powerful AI obey one or a few people? — LessWrong
1+ mon, 1+ 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....
Creative math research by AI as the latest sign of the end — LessWrong
1+ mon, 2+ day ago (902+ words) Yesterday I sat down with GPT 5.6 Sol High to do some brainstorming. The topic was one of the less appreciated Millenium Problems (the Birch and Swinnerton-Dyer conjecture), and the initial prompt started life as a question on Quora. Number theory…...
Book Review: The Infinity Machine — LessWrong
1+ mon, 2+ 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…...
On Democratizing ASI to Preserve Civil Liberties — LessWrong
1+ mon, 3+ day ago (445+ words) I continue to believe we should pause frontier AI development. Any discussion of alternative strategies should be thought of as planning for contingencies. If people still have civil liberties post-ASI, that will only be because the controllers of ASI allow…...
An Amicable Ape — LessWrong
1+ mon, 1+ week ago (158+ words) revisiting alignment The following series of posts revisits the idea of goal alignment between humanity and artificial intelligence, and within human…...
Announcing Lateral Workshop for experienced professionals moving into AI safety — LessWrong
1+ mon, 1+ week ago (379+ words) TLDR: Lateral Workshop is a program in Berkeley from September 11–13 for mid-career and senior professionals who are interested in transitioning into AI safety work to meet people in the field and explore how they can fit in. If you have…...
General capability - and capabilities generally - have no good y-axis — LessWrong
1+ mon, 1+ week ago (24+ words) BLUF: • * To determine whether AI is ‘improving exponentially’, ‘hitting the wall’, or any other claim which involves a quantity or magnitude (e.g.…...