niplav is @niplav.site · 01/10/2026I'm coming around to believing that us trying to Hail Mary-beam our civ info into being emulated by benevolent grabby aliens is not a completely crazy idea www.lesswrong.com/posts/2xHhe4... 000
niplav is @niplav.site · 14/09/2026I miss agent foundations, it was fun before we were forced to do policy and evals (Opus 5 on LittleDarkAgeBench) 030
niplav is @niplav.site · 09/09/2026Typology of multi-domain universes (From Grand Futures (Anders Sandberg, 2025)) 250
niplav is @niplav.site · 31/08/2026Note: A Toy Structural Causal Model For Psychological Agency niplav.site/notes.html#A... (Most important variable that could still be added: risk aversion?) 010
niplav is @niplav.site · 27/08/2026Wait, this is such a situation, I think? LW reacts keep winning 351
niplav is @niplav.site · 25/06/2026Short note in which I implement a set of commutative hyperoperators suggested by Ismael Ghalimi in an array programming language: niplav.site/hyperoperato... 000
niplav is @niplav.site · 01/06/2026Load-bearing "see full thread". Submissions welcome. @norvid-studies.bsky.social 270
niplav is @niplav.site · 01/06/2026dialectics.info scrapbook thread (I thought it was lost, but it's apparently still online!!!) 100
niplav is @niplav.site · 18/05/2026New LessWrong shortform with learnings from reading an 1899 anthropology of the Australian Aboriginals, featuring: ① Nocebo effect so strong it can kill ② Right-to-wife ③ Gruesome rituals ④ Division of labour/specialization between tribes and more t.co/MVHi1arF01 040
niplav is @niplav.site · 23/04/2026PSA (thx to abstractapplic & @noosphere89.bsky.social ¹): LLMs often think they're in an eval when they aren't, probabilities reported: 15%, 15%, 2%, 3%, 25%–40%, 15%–25%, 60%–75%, 10–15%, 5-10%, 70-80%, 55-70% This is way higher than what a baserate should say, which is ~10⁻⁴–10⁻⁷ 210
niplav is @niplav.site · 09/04/2026"Thinking about the ethical concerns with this request" often leads to funny moments 060
niplav is @niplav.site · 03/04/2026I am proud to announce that I now have LLM psychosis. I will not take any furhther questions 030
niplav is @niplav.site · 19/03/2026a perclique is a pergraph where for all a, b ∈ V ∪ P with a≠b there exists a peredge e∈P such that e \neq and e≠b, and e: a→b or e: b→a. there are five percliques: 011
niplav is @niplav.site · 25/02/2026this paper was written by a person with the title "Large Language Model", the {grand?}daughter of Mr. "Jesus is simply Physics at the End of Time", and another dude 120
niplav is @niplav.site · 21/02/2026Lumenator RCT on myself: Increases some variables by a lot, reliably, especially the subjective length of day and happiness, but also contentment, relaxation, and horniness. Original reasoning for lumenators stands up. Does not improve productivity (surprising!) or creativity. 151
niplav is @niplav.site · 21/02/2026I think composition is an under-rated way to really evaluate LLMs taste and pure generalization: Hard for LLMs (no audio input/output!) and fast for humans to check. Here's three piece Opus 4.6 wrote via MIDI, I'm… not impressed tbh. I think I could do better. 6170
niplav is @niplav.site · 16/02/2026I was originally gonna give you guys "Fornicating…" but that video was too wide 130
niplav is @niplav.site · 16/02/2026Probably the text highest in {annoyance, compulsion}/words I've ever written. Also plausibly the hardest from a pure writing standpoint. Charlatan Labyrinth Try to guess the constraint first! Explained at the link: niplav.site/microfiction... 7244
