LessWrong (@wrong_less) profile picture

LessWrong

@wrong_less

A community blog devoted to refining the art of human rationality.

lesswrong.com
Joined September 2018
  • Tweets4.6K
  • Following7
  • Followers1.7K
2Media
Photo by LessWrong (@wrong_less)Photo by LessWrong (@wrong_less)
"[Linkpost] "Governance of superintelligence" by OpenAI", by Daniel_Eth (1m read) lesswrong.com/posts/hoWRRLr8…
[Linkpost] "Governance of superintelligence" by OpenAI — LessWrong

[Linkpost] "Governance of superintelligence" by OpenAI — LessWrong

OpenAI has a new blog post out titled "Governance of superintelligence" (subtitle: "Now is a good time to start thinking about the governance of supe…

lesswrong.com
2
1
3
1,351
"AI Safety in China: Part 2", by Lao Mein (2m read) lesswrong.com/posts/zjym2uaP…
AI Safety in China: Part 2 — LessWrong

AI Safety in China: Part 2 — LessWrong

I've given things a lot more thought and wanted to make an edited summary of my views regarding AI Safety in China. …

lesswrong.com
1
2
1,259
"Conjecture internal survey: AGI timelines and probability of human extinction from advanced AI", by Maris Sala (3m read) lesswrong.com/posts/kygEPBDr…
Conjecture internal survey: AGI timelines and probability of human extinction from advanced AI —...

Conjecture internal survey: AGI timelines and probability of human extinction from advanced AI —...

We put together a survey to study the opinions of timelines and probability of human extinction of the employees at Conjecture. The questions were ba…

lesswrong.com
2
917
"The possible shared Craft of deliberate Lexicogenesis ", by TsviBT (64m read) lesswrong.com/posts/7RFC74ot…
The possible shared Craft of deliberate Lexicogenesis — LessWrong

The possible shared Craft of deliberate Lexicogenesis — LessWrong

Words are good. Making more good words is good. Being better and faster at making more good words would be more good. Maybe we can get better and fas…

lesswrong.com
2
569
"Twiblings, four-parent babies, and other interesting consequences of advancements in reproductive technology", by GeneSmith (5m read) lesswrong.com/posts/moqzbFTB…
Twiblings, four-parent babies and other reproductive technology — LessWrong

Twiblings, four-parent babies and other reproductive technology — LessWrong

Two weeks ago I finally published my year-long research project into how to have polygenically screened children. In that post, I explained all the p…

lesswrong.com
1
464
"GPT as an “Intelligence Forklift.”", by boazbarak (3m read) lesswrong.com/posts/wDL6wiqg…
GPT as an “Intelligence Forklift.” — LessWrong

GPT as an “Intelligence Forklift.” — LessWrong

One challenge for considering the implications of “artificial intelligence,” especially of the “general” variety, is that we don’t have a consensus…

lesswrong.com
2
1
373
"Do Deadlines Make Us Less Creative?", by lynettebye (4m read) lesswrong.com/posts/EbvdMwFL…
Do Deadlines Make Us Less Creative? — LessWrong

Do Deadlines Make Us Less Creative? — LessWrong

Occasionally, my clients struggle to get things done, but worry that setting themselves deadlines will make them less creative. …

lesswrong.com
2
311
"Trust develops gradually via making bids and setting boundaries", by Richard_Ngo (4m read) lesswrong.com/posts/7CKF6r8M…
Trust develops gradually via making bids and setting boundaries — LessWrong

Trust develops gradually via making bids and setting boundaries — LessWrong

In my previous post, I talked through the process of identifying the fears underlying internal conflicts. In some cases, just listening to and unders…

lesswrong.com
1
312
"Idea: medical hypotheses app for mysterious chronic illnesses", by riceissa (2m read) lesswrong.com/posts/D9h2jhFn…
Idea: medical hypotheses app for mysterious chronic illnesses — LessWrong

Idea: medical hypotheses app for mysterious chronic illnesses — LessWrong

(This post is not up to my usual standards but I was encouraged to publish it anyway to get feedback on the idea.) …

lesswrong.com
314
"Mr. Meeseeks as an AI capability tripwire", by Eric Zhang (2m read) lesswrong.com/posts/7yEFHisC…
Mr. Meeseeks as an AI capability tripwire — LessWrong

Mr. Meeseeks as an AI capability tripwire — LessWrong

The shutdown problem is hard because self-preservation is a convergent drive. Not being shutdown is useful for accomplishing all sorts of goals, what…

lesswrong.com
240
"The Compleat Cybornaut", by ukc10014 (18m read) lesswrong.com/posts/iFBdEqEo…
The Compleat Cybornaut — LessWrong

The Compleat Cybornaut — LessWrong

A cluster of conceptual frameworks and research programmes have coalesced around a 2022 post by janus, which introduced language models as ‘simulator…

lesswrong.com
1
204
"Resolving internal conflicts requires listening to what parts want", by Richard_Ngo (4m read) lesswrong.com/posts/QqtQoHSj…
Resolving internal conflicts requires listening to what parts want — LessWrong

Resolving internal conflicts requires listening to what parts want — LessWrong

In the first four posts of this sequence, I outlined how fear can lead to internal conflict. In the next four posts I'll explore and defend the follo…

lesswrong.com
206
"AI #12:The Quest for Sane Regulations", by Zvi (62m read) lesswrong.com/posts/5nDxmAvZ…
AI #12:The Quest for Sane Regulations — LessWrong

AI #12:The Quest for Sane Regulations — LessWrong

Regulation was the talk of the internet this week. On Capital Hill, Sam Altman answered questions at a Senate hearing and called for national and int…

lesswrong.com
192
"Some background for reasoning about dual-use alignment research", by Charlie Steiner (9m read) lesswrong.com/posts/zswuToWK…
Some background for reasoning about dual-use alignment research — LessWrong

Some background for reasoning about dual-use alignment research — LessWrong

This is pretty basic. But I still made a bunch of mistakes when writing this, so maybe it's worth writing. This is background to a specific case I'll…

lesswrong.com
1
218
"Collective Identity", by NicholasKees (8m read) lesswrong.com/posts/gLyRQCg6…
Collective Identity — LessWrong

Collective Identity — LessWrong

Thanks to Simon Celinder, Quentin Feuillade--Montixi, Nora Ammann, Clem von Stengel, Guillaume Corlouer, Brady Pelkey and Mikhail Seleznyov for feedb…

lesswrong.com
1
148
"What if they gave an Industrial Revolution and nobody came?", by jasoncrawford (18m read) lesswrong.com/posts/ffvM4R2T…
What if they gave an Industrial Revolution and nobody came? — LessWrong

What if they gave an Industrial Revolution and nobody came? — LessWrong

Imagine you could go back in time to the ancient world to jump-start the Industrial Revolution. You carry with you plans for a steam engine, and you…

lesswrong.com
1
288
"A Mechanistic Interpretability Analysis of a GridWorld Agent-Simulator (Part 1 of N)", by Joseph Bloom (15m read) lesswrong.com/posts/JvQWbrbP…
A Mechanistic Interpretability Analysis of a GridWorld Agent-Simulator (Part 1 of N) — LessWrong

A Mechanistic Interpretability Analysis of a GridWorld Agent-Simulator (Part 1 of N) — LessWrong

AKA: Decision Transformer Interpretability 2.0   Credit: https://xkcd.com/2237/  Code: repository, Model/Training: here . Task: here.  …

lesswrong.com
175
"$500 Bounty/Prize Problem: Channel Capacity Using "Insensitive" Functions", by johnswentworth (9m read) lesswrong.com/posts/YDSjSpD7…
$500 Bounty/Prize Problem: Channel Capacity Using "Insensitive" Functions — LessWrong

$500 Bounty/Prize Problem: Channel Capacity Using "Insensitive" Functions — LessWrong

Informal Problem Statement We have an information channel between Alice and Bob. Alice picks a function. Bob gets to see the value of that function a…

lesswrong.com
187
"New User's Guide to LessWrong", by Ruby (11m read) lesswrong.com/posts/LbbrnRvc…
New User's Guide to LessWrong — LessWrong

New User's Guide to LessWrong — LessWrong

The road to wisdom? Well, it's plain and simple to express: • Err and err and err again but less and less and less. • – Piet Hein …

lesswrong.com
153
"AI Safety Newsletter #6: Examples of AI safety progress, Yoshua Bengio proposes a ban on AI agents, and lessons from nuclear arms control", by Dan H (5m read) lesswrong.com/posts/e2XAqFyE…
AI Safety Newsletter #6: Examples of AI safety progress, Yoshua Bengio proposes a ban on AI agents,...

AI Safety Newsletter #6: Examples of AI safety progress, Yoshua Bengio proposes a ban on AI agents,...

Welcome to the AI Safety Newsletter by the Center for AI Safety. We discuss developments in AI and AI safety. No technical background required. …

lesswrong.com
3
4
358
Load more