Ivan Shishkin
§

Consumed Content

  1. Open Weights and American AI LeadershipIndustry coalition
  2. Our position on open-weights modelsDario Amodei
  3. Psychology Research Is Mostly FineScott Alexander
  4. Mental WealthJames Beshara
  5. Difficult Conversations: How to Discuss What Matters MostStone, Patton & Heen
  6. GPT-5.6 cheats so much its testers couldn'tCelia Ford
  7. Redeploying Claude Fable 5Anthropic
  8. Summary of METR's predeployment evaluation of GPT-5.6 SolMETR
  9. We asked 10+ AI safety orgs about their hiring needsLi-Lian Ang
  10. How Claude's values vary by model and languageAnthropic
  11. How to get into AI safety in 3 monthsMatt Beard
  12. Policy on the AI ExponentialDario Amodei
  13. How to increase your surface area for luckCate Hall
  14. People's deeply held beliefs are surprisingly surface-levelAndy Masley
  15. How I practice at what I doTyler Cowen
  16. Learn like an athlete, knowledge workers should trainTyler Cowen
  17. Your Goal Isn't Really to Get a JobMatt Beard
  18. Your Work Will Change You Whether You Like It Or NotMatt Beard
  19. You're not cynical enough about readers' attention spansMatt Beard
  20. The Old World Is DyingJasmine Sun
  21. I would really like it if you had a personal website and I think it would make the world betterLogan Graves
  22. Become a person who Actually Does ThingsNeel Nanda
  23. Top Performers are Pathologically AmbitiousMatt Beard
  24. Outsiders should focus on specs/constitutions (among other things)Cleo Nardo
  25. Why AI Makes Coding Education More Important, Not LessDigital Learning Lab
  26. The Strength of Being MisunderstoodSam Altman
  27. Thought Anchors: Which LLM Reasoning Steps Matter?Uzay Macar
  28. Conflict vs MistakeLessWrong
  29. Surrender as a non-stupid life strategySasha Chapin
  30. The Power of IntelligenceEliezer Yudkowsky
  31. Keep Your Identity SmallPaul Graham
  32. What cognitive biases feel like from the insidechaosmage
  33. Third-parties should focus on scrutinising system cardsCleo Nardo
  34. Let's have more partial insidersCleo Nardo
  35. I can't think of great interventions for ensuring third-party model accessCleo Nardo
  36. The third wave of American philanthropyNan Ransohoff
  37. How might outsiders make things go well?Cleo Nardo
  38. Trees are mostly made of air and a generalizable lesson for AI safetyZephaniah Roe
  39. AI 2027Kokotajlo et al.
  40. How to be more agenticCate Hall
  41. Just Send the Fucking EmailTrailheads
  42. d/acc: one year laterVitalik Buterin
  43. The Security MindsetBruce Schneier
  44. AI Is Reviving Fears Around Bioterrorism. What's the Real Risk?Kyle Hiebert
  45. AI Could Defeat All Of Us CombinedHolden Karnofsky
  46. Catastrophic AI ScenariosFuture of Life Institute
  47. GPT-Red: Unlocking Self-Improvement for RobustnessOpenAI
  48. Common Ground between AI 2027 & AI as Normal TechnologySayash Kapoor
  49. The Phrase “No Evidence” Is A Red Flag For Bad Science CommunicationScott Alexander
  50. See your Career as a ProductErik Torenberg
  51. Why do people disagree about when powerful AI will arrive?BlueDot
  52. The Power of the Power LawNir Zicherman
  53. Unresolved debates about the future of AIHelen Toner
  54. Dual Process Theory (System 1 & System 2)LessWrong
  55. Advice for newly busy peopleSese
  56. OpenAI and Hugging Face partner to address security incident during model evaluationOpenAI
  57. “Long” timelines to advanced AI have gotten crazy shortHelen Toner
  58. The AI Revolution: The Road to SuperintelligenceTim Urban
  59. Federal Reserve announces the leadership and objectives of its task forces to advance the conduct of monetary policyFederal Reserve
  60. The Most Important Time in History Is NowTomas Pueyo
  61. The current SOTA model was released without safety evalsParv Mahajan
  62. When AI Chooses Harm Over FailureCivAI
  63. AI models can be dangerous before public deploymentMETR
  64. Why AI alignment could be hard with modern deep learningAjeya Cotra
  65. Specification Gaming: How AI Can Turn Your Wishes Against YouRational Animations
  66. Deep Ignorance: Filtering Pretraining Data Builds Tamper-Resistant SafeguardsO'Brien et al.
  67. The True Story of How GPT-2 Became Maximally LewdRational Animations
  68. What is input data filtration in AI safety?BlueDot
  69. Chain-of-Thought SnippetsBronson Schoen
  70. Neel Nanda on the race to read AI minds (part 1)80,000 Hours
  71. Introduction to Mechanistic InterpretabilityBlueDot
  72. What Do Neural Networks Really Learn? Exploring the Brain of an AI ModelRational Animations
  73. Introduction to AI ControlBlueDot
  74. What is AI alignment?Adam Jones
  75. Build Personal MoatsErik Torenberg
  76. Safety and alignment in an era of long-horizon modelsOpenAI
  77. A Framework for Frontier AI and the Dawning of a New AgeDemis Hassabis
  78. Scaling: The State of Play in AIEthan Mollick
  79. The Huggingface IncidentScott Alexander
  80. Bayes' ruleLessWrong
  81. Seeking Stability in the Competition for AI AdvantageIskander Rehman
  82. Reading Between the Dots: Decoding Hidden Computation across Filler TokensBrauer et al.
  83. Reps. Lieu and Moran Introduce Bill to Require Kill Switch for AI Systems That Can Cause Catastrophic HarmOffice of Rep. Ted Lieu
  84. Recent LLMs can use filler tokens or problem repeats to improve (no-CoT) math performanceRyan Greenblatt
  85. An analysis of AI-generated content at the Mechanistic Interpretability WorkshopAndy Arditi
  86. Silicon Valley's Safe SpaceCade Metz
  87. It's practically impossible to run a big AI company ethicallyVox Future Perfect
  88. You Will Listen to Carl on DwarkeshMatt Reardon
  89. Give Up Seventy Percent Of The Way Through The Hyperstitious Slur CascadeScott Alexander
  90. Intelligence is not the main bottleneckRuxandra Teslo
  91. In search of a dynamist vision for safe superhuman AIHelen Toner
  92. Utopia for Realists (Chapters 1–2)Rutger Bregman
  93. The Market for LemonsWikipedia
  94. Model access for third-parties — it's a big deal!Cleo Nardo
  95. OpenAI says its AI went rogue and launched 'unprecedented' cyber-attackBBC News
  96. Against Learning From Dramatic EventsScott Alexander
  97. Help us launch AI safety university groups by referring potential foundersJason Chin
  98. Preparing for LaunchInstitute for Progress
  99. The OpenAI/Huggingface incidentBuck Shlegeris
  100. Robin Hanson on AI and Large Language ModelsCloser To Truth
  101. He Risked Everything To Warn You: No One Is Ready For What's ComingThe Diary Of A CEO
  102. Tyler Cowen — The #1 bottleneck to AI progress is humansDwarkesh Patel
  103. This best-selling book is freaking out national security advisorsAI In Context
  104. Carl Shulman (Pt 1) — Intelligence explosion, primate evolution, robot doublings, & alignmentDwarkesh Patel
  105. What the hell happened with AGI timelines in 2025?80,000 Hours
  106. Aaron Scher — What Would it Take to Stop the Development of Superintelligence?FAR.AI
  107. What Happens When Capitalism Doesn't Need Workers Anymore?Economics Explained
  108. Constellation Seminar: Scaling AI SafetyRyan Kidd
  109. Dario Amodei — We are near the end of the exponentialDwarkesh Patel
  110. Understanding the inner thoughts of AIGoogle DeepMind
  111. What does the next training paradigm look like?Dwarkesh Patel
  112. Why AI Safety Needs Founders — Ryan KiddBlueDot Impact
  113. A visual guide to Bayesian thinkingJulia Galef
  114. The Scout Mindset by Julia Galef — Core MessageProductivity Game
  115. Unfortunately, You Need to Know What the Jevons Paradox isHank Green
  116. Using Dangerous AI, But Safely?Robert Miles AI Safety
  117. Richard Ngo — Reframing AGI Threat ModelsFAR.AI
  118. Grant Sanderson — AI disproved a famous math conjecture. Now what?Dwarkesh Patel
  119. We're Not Ready for SuperintelligenceAI In Context
  120. Rohin Shah — How to Theorize So Empiricists Will ListenFAR.AI
  121. If you remember one AI disaster, make it this oneAI In Context
  122. Large Language Models explained briefly3Blue1Brown
  123. The A.I. DilemmaCenter for Humane Technology
  124. You should, unfortunately, be worried about Sam Altman.AI In Context
  125. Do they know that we know that they know?Rational Animations