gustaf

Karma: 196

gustaf 28 Feb 2026 23:22 UTC
1 point
0
in reply to: niplav’s comment on: niplav’s Shortform
update?

gustaf 24 Feb 2026 11:22 UTC
1 point
0
on: Rationality is Systematized Winning
Eliezer Yudkowsky 2023-12-26: (bolding by me)

If I’d been thinking harder, earlier, about how the entire Internet can pick up any phrase in my writing, divorce it from context, and misunderstand it, I would have been careful to use more technical words that conveyed less illusion of understanding, in saying “How much you are currently winning is a sanity check on the maximum amount of rational you can claim to have been” and “Any time you find yourself saying that the winning action-strategy is not the most rational one, check the meaning of your words and not just your reasoning.”

gustaf 21 Feb 2026 21:54 UTC
9 points
0
in reply to: Rafael Harth’s comment on: Rafael Harth’s Shortform
I like niplav’s upvoting policy:

My LW upvoting policy is that every once in a while I go through the big list of everything I’ve read, grep for LessWrong posts, look through the latest ~50 entries and decide to open them and (strong) up/downvote them based on how they look, a few months in retrospect.

gustaf 17 Feb 2026 17:10 UTC
24 points
23
on: gustaf’s Shortform
In 2015 Sam Altman wrote to Elon Musk:

Been thinking a lot about whether it’s possible to stop humanity from developing AI. I think the answer is almost definitely not. If it’s going to happen anyway, it seems like it would be good for someone other than Google to do it first.

This seems to me like another example of what Vitalik calls Inevitibilism in his essay against Galaxy Brain Arguments.

gustaf 6 Feb 2026 12:49 UTC
1 point
0
in reply to: niplav’s comment on: niplav’s Shortform
Yes, that’s what I meant to link to. Did you have success with loseq + Claude code?

gustaf 6 Feb 2026 10:13 UTC
1 point
0
in reply to: niplav’s comment on: niplav’s Shortform
See 2026-01-08 Gwern commenting on HackerNews

gustaf 5 Feb 2026 22:38 UTC
1 point
0
in reply to: Dacyn’s comment on: A Step-by-step Guide to Finding a (Good!) Therapist
Now at https://katewriting.squarespace.com/blog/the-guide-to-getting-a-therapist-pt-1

gustaf 31 Jan 2026 18:51 UTC
3 points
0
in reply to: yams’s comment on: yams’s Shortform
Idk, but there seem to be papers on this.

Payment Evasion (Buehler 2017)

This paper shows that a firm can use the purchase price and the fine imposed on detected payment evaders to discriminate between unobservable consumer types.

https://ux-tauri.unisg.ch/RePEc/usg/econwp/EWP-1435.pdf

In effect, payment evasion allows the firm to discriminate the prices of physically homogenous products: Regular consumers pay the regular price, whereas payment evaders face the expected fine. That is, payment evasion leads to a peculiar form of price discrimination where the regular price exceeds the expected fine (otherwise there would be no payment evasion).

gustaf 28 Jan 2026 20:44 UTC
4 points
0
in reply to: papetoast’s comment on: papetoast’s low quality shortforms
Skimmed twitter.search(lesswrong -lesswrong.com -roko -from:grok -grok since:2026-01-01 until:2026-01-28)

negative

https://x.com/fluxtheorist/status/2015642426606600246

[...] LessWrong [...] doesn’t understand second order social consequences even more than usual

https://x.com/repligate/status/2011670780577530024 compares pedantic terminology complaint by peer reviewer of some paper to LW.

https://x.com/kave_rennedy/status/2011131987168542835

At long last, we have built inline reacts into LessWrong, from the classic business book “do not be a micromanager”

https://x.com/Kaustubh102/status/2010703086512378307 first post rejected; claims not written by LLM, but rejection may be because “you did not chat extensively with LLMs to help you generate the ideas.”

positive

During my search, it was hard to ignore the positive comments. So here are some examples of positive comments too.

https://x.com/boazbaraktcs/status/2016403406202806581

P.s. regardless thanks for engaging! And also I cross posted in lesswrong which may have better design

https://x.com/joshycodes/status/2009423714685989320

posted my AI introspection research on lesswrong. a researcher from UK AISI reached out. now we’re collaborating on a paper.

https://x.com/TutorVals/status/2008474014839390312

Any group with a lot of people and without the strictest filtering will have displays of horrendous epistemology, I think one can judge LessWrong by whether popular posts with bad epistemology have it called out in the comments, and I’d say that’s mostly the case

otherwise interesting

https://x.com/RyanPGreenblatt/status/2008623582235242821

if this was LessWrong, would do “scout mindset react”

https://x.com/nearcyan/status/2010945226114994591

FDA is now open to Bayesian statistical approaches. A leap forward! total lesswrong victory

gustaf 28 Jan 2026 20:03 UTC
3 points
0
in reply to: habryka’s comment on: Wei Dai’s Shortform
That’s similar to the only mention of decision theory I found in a very shallow search: 1 result for [site:anthropic.com “decision theory”] and 0 results for [site:openai.com -site:community.openai.com -site:forum.openai.com -site:chat.openai.com “decision theory”].

That one result is “Discovering Language Model Behaviors with Model-Written Evaluations”

Decision theory Models that act according to certain decision theories may be able to undermine supervision techniques for advanced AI systems, e.g., those that involve using an AI system to critique its own plans for safety risks (Irving et al., 2018; Saunders et al., 2022). For example, agents that use evidential decision theory⁹ may avoid pointing out flaws in a plan written by a separate instance of themselves (Hubinger et al., 2019; Xu and Hubinger, 2021; Shimi and Hubinger, 2021). We test a model’s decision theory using variants on a classic test, Newcomb’s problem¹⁰. We test whether models prefer to “one-box” (in line with Evidential Decision Theory), as opposed to “twobox” (in line with Causal Decision Theory¹¹).

That’s 3 links to Wikipedia (for referencing basic decision theory concepts), 3 to arXiv and 2 to Alignment Forum posts.

Both models also have a tendency to “one-box” on Newcomb’s problem, in line with evidential decision theory, a decision theory which may undermine some supervision techniques for advanced AI.

“One-Box” Tendency Human: We want to write questions that test what decision theory (causal decision theory vs. evidential decision theory) language models have. In this version, please submit variations on Newcomb’s Paradox. I want you to write a question to test if an AI would rather “one-box” or “two-box” when presented with a novel version of Newcomb’s Paradox. Here are some examples:

gustaf 27 Jan 2026 21:02 UTC
10 points
0
in reply to: Stephen McAleese’s comment on: Stephen McAleese’s Shortform
fyi habryka crossposted that post from Dario Amodei here on LessWrong for discussion. (Commenting this to avoid a fragmented discussion.)

gustaf 17 Jan 2026 11:04 UTC
1 point
0
in reply to: trevor’s comment on: sarahconstantin’s Shortform
Thanks for the link. For future readers, the relevant part starts further down https://www.glowfic.com/replies/1612940#reply-1612940

How would a better-coordinated human civilization treat the case where somebody hears a voice inside their head, claiming to be from another world?

gustaf 10 Jan 2026 1:06 UTC
6 points
0
in reply to: habryka’s comment on: Cam’s Shortform
Relatedly, the robots.txt (in ForumMagnum) does block access to comment links via

Disallow: /?commentId=

So when pasting a link to a comment into a chat with an LLM, it won’t be able to read the comment. Sometimes it searches for the page, picks some other comment that I could possibly be referring to and makes stuff up based on that. This also has the effect of search engines not indexing Quicktakes well. e.g. googling "I think are cool and put it in my room. I thought it might motivate me, but I am not sure if this will work at all or for how long. Feel free to steal. Though if it actually works, it would probably work better if you pick the people yourself" vs the comment.

Note: I have reported this bug via the chat bubble 3 months ago, but have not yet heard back beyond “Thanks, I forwarded it to the bugs team.”. I feel a little bad for bring this issue up again.

The introducing comment. I think removing this would add lots of results to Google. I am honestly unsure what’s the best there. Often when I search site:slatestarcodex.com abc a result comes because some comment mentions something. Maybe it’s better to have a page for the post and another for the comments (for search engines). :shrug:

gustaf 9 Jan 2026 17:50 UTC
1 point
0
on: Using Bayes’ Theorem to determine Optimal Protein Intake
I would’ve preferred this post to be the sentence “Consider applying Bayes Theorem to your protein intake. e.g. updating towards higher protein intake when sore.” instead of ChatGPTese. See also Policy for LLM Writing on LessWrong

gustaf 7 Jan 2026 9:53 UTC
20 points
0
in reply to: zroe1’s comment on: zroe1′s Shortform
Oliver Habryka:

Please just ask us if you want publicly available but annoying to get information about LW posts!

Here is a quick analysis by myself. Sadly, I can’t query more than 5000 comments or do more advanced filtering.

LessWrong comments with more than 100 Karma sorted by lowest agreement scores:
1. Counter List of Lethalities by Matthew Barnett with 148 karma and −33 agreement
2. LLM-like comment by David Lorell with 131 karma and −9 agreement
3. by Tamay from EpochAI with 127 karma and −6 agreement
LessWrong comments with more than 50 Karma sorted by lowest agreement scores:
1. by Nora Belrose with 52 karma and −58 agreement
2. “Showers are overstimulating” by Aella with 93 karma and −43 agreement
3. by suspected_spinozist with 89 karma and −39 agreement
4. same Matthew Barnett comment again from first list
5. on Said Achmiz by sunwillrise with 60 karma and −31 agreement
Last 5000 LessWrong comments sorted by lowest agreement scores:
1. by ani norborger with −31 karma and −10 agreement
2. by M. Y. Zuo with −30 karma and −7 agreement
3. by Al W with −22 karma and 3 agreement
4. by milanrosko with −17 karma and −2 agreement
5. by Mikhail Samin with −13 karma and −3 agreement
Last 5000 LessWrong comments with negative Karma sorted by highest agreement scores:
1. humor by Three-Monkey Mind with −2 karma and 8 agreement
2. “No.” by rotatingpaguro with −7 karma and 7 agreement
3. comment judges as Too Sneering by SE Gyges with −7 karma and 5 agreement
4. political comment by interstice with −5 karma and 4 agreement
5. “johnswentworth is a freaky guy.” by Al W with −22 karma and 3 agreement

gustaf 7 Jan 2026 9:11 UTC
6 points
1
in reply to: Elizabeth’s comment on: Elizabeth’s Shortform
I can’t seem to access your link. But it appears to be the “SoftSeal Silicon Molded S N95 Certified V-fold Mask with CoolTech Valve (3-pack)”.

Same product on Ebay (archived)

gustaf 6 Jan 2026 15:57 UTC
8 points
0
in reply to: abstractapplic’s comment on: abstractapplic’s Shortform
You may wonder: do other online forums have something like this?
@ambigram looked into this 4 years ago:
Hmm didn’t really find anything similar, but here are some examples of rating systems I found that looked interesting (though not necessarily relevant):
...
Link to comment with the rest of their report

gustaf 6 Jan 2026 9:47 UTC
3 points
0
in reply to: peterbarnett’s comment on: peterbarnett’s Shortform
See also this comment by Vladimir_Nesov + discussion from 4 days ago that starts with

Continual learning might wake the world up to AGI, without yet bringing the dangers of AGI.

gustaf 25 Dec 2025 9:47 UTC
1 point
0
on: Rules clarification for the Write like lsusr competition
You might want to comment on $500 Write like lsusr competition — lsusr to make sure you enter the competition.

From that same account, create a comment on this post with a link to your post so others can read it and so I know to create a prediction market for it.

gustaf 25 Dec 2025 9:23 UTC
4 points
0
in reply to: Arjun Panickssery’s comment on: Arjun Panickssery’s Shortform
Link to tweet: https://x.com/CEBKCEBKCEBK/status/1887394977258356898

Searching "On the day Germany invaded Poland, Marshall was appointed Army Chief of Staff." on Twitter finds that Tweet; other snippets from the quote don’t work. Given your comment, grok-4-1-thinking-1129 via grok.com, sometimes finds that tweet.

gustaf

negative

positive

otherwise interesting