A compendium of Less Wrong.
- Don’t say straightforwardly false things.
- Track (for yourself) and distinguish (for others) your inferences from your observations.
- Estimate (for yourself) and make clear (for others) your rough level of confidence in your assertions.
- Make your claims clear, explicit, and falsifiable, or explicitly acknowledge that you aren’t doing so (or can’t).
- Aim for convergence on truth, and behave as if your interlocutors are also aiming for convergence on truth.
- Don’t jump to conclusions—maintain at least two hypotheses consistent with the available information.
- Be careful with extrapolation, interpretation, and summary/restatement—distinguish between what was actually said, and what it sounds like/what it implies/what you think it looks like in practice/what it’s tantamount to. If you believe that a statement A strongly implies B, and you are disagreeing with A because you disagree with B, explicitly note that “A strongly implies B” is a part of your model.
- Allow people to restate, clarify, retract, and redraft their points, if they say that their first attempt failed to convey their intended meaning; do not hold people to the first or worst version of their claim.
- Don’t weaponize equivocation/don’t abuse categories/don’t engage in motte-and-bailey shenanigans.
- Hold yourself to the absolute highest standard when directly modeling or assessing others’ internal states, values, and thought processes.
Less Wrong is a community resource devoted to refining the art of human rationality, sometimes known as rationalism.
Less Wrong believes in promoting lifestyle changes that lead to increased rationality and self-improvement. Writing posts that often focus on avoiding biases related to decision-making, the evaluation of evidence, and psychological barriers that prevent good decision-making.
In the early 2000s, American artificial intelligence researcher Eliezer Yudkowsky was frequently annoyed, frustrated, and disappointed in people’s inability to think in ways he considered obviously rational. He eventually focused on teaching the rationality skills necessary to do AI safety research until a sustainable culture would allow him to focus on AI safety research while also finding and training new AI safety researchers.
In February 2009, after the topics drifted more widely, Yudkowsky’s took his Overcoming Bias contributions to start the community blog Less Wrong and Overcoming Bias became Hanson’s personal blog.
LessWrong started with a series of daily blog posts written by Eliezer, originally called The Sequences. The Sequences were compiled into an edited volume, Rationality: A-Z. These writings attracted a large community of readers and writers interested in the art of human rationality.
Starting in 2012, many core members of the community stopped posting on Less Wrong due to the physical community’s growth in the Bay Area and increased demands and opportunities from other projects. Yudkowsky received enough support to focus on AI research instead of community-building. Prominent writers on Less Wrong left to their own blogs to develop their voice without asking if it was within the bounds of Less Wrong—a move which collectively formed the ‘rationalist movement.’
Less Wrong acknowledges this departure of top writers and lists blogs from the diaspora and rationalist movement.
Some prominent ideas that grew out of the Less Wrong community members include:
- Center for Applied Rationality
- Effective altruism
- Center for Effective Altruism
- Slate Star Codex
In 2015-2016 the site continued a steady decline of activity leading some to declare the site dead.
In 2016-2017, discussion of revival occurred and Oliver Habryka formed a team to relaunch LessWrong in 2017 on a new codebase—LessWrong 2.0.
LessWrong 2.0 was the first time LessWrong had a full-time dedicated development team behind it instead of only volunteer hours.
After nearly being put into read-only archive mode, LessWrong 2.0 brought the site back to life.
launched in 2018 with a new codebase and full-time team.
LessWrong is a place to 1) develop and train rationality, and 2) apply one’s rationality to real-world problems.
The LessWrong Books
For the first time, you can now buy the best new ideas on LessWrong in a physical book set.
In 2019, LessWrong started an annual review process to determine the best content on the site. The community reviewed all the posts on LessWrong posted in 2018 and voted to rank the best of them.
Of the over 2000 LessWrong posts reviewed, the LessWrong 2018 book contains 41 of the top voted essays, along with some comment sections, some reviews, a few extra essays to give context, and some preface/meta writing.
Rationality: From AI to Zombies
Rationality: A-Z, also referred to as the sequences, is an extensive exploration of how human minds come to understand the world they exist in and reasons they commonly fail to understand.
These sequences are a series of essays written by Eliezer Yudkowsky between 2006 and 2009 on the blogs Overcoming Bias and Less Wrong. About half of these essays were organized into a number of thematically linked “sequences” of blog posts.
The comprehensive work:
- Lays foundational conceptions of belief, evidence, and understanding.
- Reviews the systematic biases and common excuses which cause us to believe false things.
- Offers guidance on how to change our minds and how to use language effectively
- Depicts the nature of human psychology with reference evolution.
- Clarifies the kind of morality we can have in a reducible, physical world.
- Repeatedly reminds us that confusion and mystery exist only in our minds.
Abridged indexes of the sequences:
R:AZ 1 Map and Territory sequences
R:AZ 2 How to Actually Change Your Mind sequences
- Overly Convenient Excuses
- Politics and Rationality
- Against Rationalization
- Against Doublethink
- Seeing with Fresh Eyes
- Death Spirals
- Letting Go
R:AZ 3 The Machine in the Ghost
R:AZ 4 Mere Reality
R:AZ 5 Mere Goodness
R:AZ 6 Becoming Stronger