LessWrong (30+ Karma)

Escolta'l a la teva aplicació

• Apple Podcasts
• Pocket Casts
• Castro
• Podcast Index
• RSS

Episodis

250 episodis

1. [Linkpost] “Automating subtle emergent misalignment research” by evhub, Oscar Obeso, rqi, Monte M fa 11 hores 2:56
2. “Nectome 2026 retrospective (and our final pre-sale!)” by Aurelia fa 1 dia 10:08
3. “Organizations I’d recommend for funding” by So8res fa 1 dia 8:06
4. “Against strawmanning unilateral pauses” by Geoffrey Irving fa 1 dia 14:09
5. “Teaching Qwen Why” by astwei, Jack Thompson, Zephaniah Roe, yix, berry fa 1 dia 1:04:38
6. “Scott Alexander’s Against SURA Seems Deeply Confused” by Zvi fa 1 dia 1:06:41
7. “Towards understanding score-seeking in LLMs” by Alex Mallen, Aghyad Deeb, ryan_greenblatt fa 1 dia 19:19
8. “Some “coherence arguments” are two different things in a trenchcoat” by Steven Byrnes fa 2 dies 18:58
9. “Free-Floating Social Patterns” by johnswentworth fa 2 dies 22:35
10. “Against Against SURA” by Mikhail Samin fa 2 dies 26:45
11. “Magma is partnering with CAISSI to investigate recent unintended model behavior” by Parv Mahajan fa 2 dies 5:40
12. “Accidentally Misleading People Should Be Embarrassing; Or, Higher Innocence” by Zack_M_Davis fa 2 dies 11:06
13. “Towards The Autograntor” by Scott Alexander fa 2 dies 55:34
14. “Misalignment traps lie in wait” by David Africa fa 2 dies 6:17
15. “Explaining “AI might obliterate you” to senior policymakers in India” by TurnTrout fa 3 dies 24:35
16. “Generalized belief states in transformers” by Artemy Kolchinsky fa 3 dies 41:01
17. “Against SURA” by Scott Alexander fa 3 dies 52:30
18. “What Social Constructs Are (And Are Not)” by johnswentworth fa 4 dies 5:56
19. “Reviewing my 2017 AI Risk Project” by jefftk fa 4 dies 11:51
20. “Childhood and Education #22: Voice Without Exit” by Zvi fa 4 dies 45:16
21. “Dice enterprises as imperfect-recall games” by jessicata fa 4 dies 22:50
22. “Lessons From Red-teaming Auto-review, OpenAI’s Blocking Monitor” by monika_j, Giovanni M. Occhipinti fa 4 dies 29:24
23. “Microdosing the spirit of Anthropic” by Jan_Kulveit fa 5 dies 9:36
24. “Walking Partway to School” by jefftk fa 5 dies 1:16
25. “My experiences with the extremes of post-viral illness and brain retraining/mindbody work” by riceissa fa 5 dies 1:53:07
26. “What is OpenAI’s Mission Advisory Council?” by hersheys, meru fa 5 dies 2:26
27. “Making alignment the agent’s problem” by Florian_Dietz fa 5 dies 8:57
28. “AI #192: Anthropic Is Going Public” by Zvi fa 6 dies 1:48:42
29. “Leaking Muse’s Charter, Meta’s model spec” by vigji fa 6 dies 13:35
30. “Three Villages in a Trenchcoat, Post I” by DaystarEld fa 1 setmana 36:18
31. [Linkpost] “Reconceptualize things as your top affordances” by KatjaGrace fa 1 setmana 2:05
32. “You can now ”@Grok” on LessWrong” by Ruby fa 1 setmana 3:26
33. “The Crypto Dream is Structurally Illegal” by Taylor G. Lunt fa 1 setmana 10:43
34. “Some thoughts on GRAFT and follow-up work” by shawnghu fa 1 setmana 3:29
35. “Subtly wrong advice misaligns models more than obviously wrong advice” by Christine Corry, arav-dhoot, Jozdien fa 1 setmana 22:37
36. “Anthropic Files Its S-1” by Zvi fa 1 setmana 1:17:45
37. “I spent 12 hours writing an article for LessWrong, calling for a more ‘hippie’ rationalism. Then I didn’t post it. This is why (plus a link to the article, explicitly not located on LessWrong)” by honeybee fa 1 setmana 8:29
38. “Streetlight, Torchlight and Nightvision” by DusanDNesic fa 1 setmana 11:57
39. “Quantity has a quality of its own” by djbinder fa 1 setmana 11:35
40. “Against Girardian mimesis” by Viliam fa 1 setmana 13:10
41. “Anthropic Versus Anthropic” by Zvi fa 1 setmana 48:44
42. “AI speeds up 3rd party safety researchers by ~1.6x (via self-reports)” by Angel Martinez, Dewi Gould, nicole x, Alejandro Aristizabal fa 1 setmana 15:32
43. “A big case study of propensity entanglement” by nielsrolf, lwen2027, megasilverfist, Owen Terry, Johannes Taraz fa 1 setmana 5:39
44. “Scattered thoughts on how doing math feels these days” by dynomight fa 1 setmana 15:40
45. “Measuring agent capabilities on open-ended AI R&D tasks” by Alex Jackson, Titus Buckworth, Jason R Brown, Francis Rhys Ward fa 1 setmana 36:20
46. “The AI buildout is the largest bet in the history of capitalism. Key decision makers believe very powerful AI will be achieved soon.” by Hayekstan fa 1 setmana 7:22
47. “Hair Thickness of My Friends” by XelaP fa 1 setmana 6:35
48. “A basic case for differentially accelerating conceptual reasoning capabilities” by Emery Cooper, Caspar Oesterheld, Chi Nguyen fa 1 setmana 50:07
49. “A cat-belling problem for FDT” by jessicata fa 1 setmana 20:24
50. “Anthropic are the Good Guys” by Ivan Vendrov fa 1 setmana 20:55