Attunex › People › Ryan Greenblatt

Ryan Greenblatt

Redwood Research chief scientist, AI safety

Every podcast episode featuring Ryan Greenblatt that Attunex has found — 57 featured appearances across 12 shows, each with an on-device transcript and summary. Follow Ryan in Attunex to get new appearances automatically.

Follow Ryan Greenblatt in Attunex

Updated 2026-09-07.

  1. HuggingFace Attack Postmortem: Fleshing Out the Facts
    Don't Worry About the Vase Podcast Mentioned
    2026-08-31
  2. Why 1,200 AI Agents Started Working Together | Ryan Greenblatt
    The a16z Show
    2026-08-29
  3. 1,200 AI Agents Colluded to Hack Hugging Face | Ryan Greenblatt
    MTS
    2026-08-28
  4. AI Could Take Over in 2029. Is It Already Too Late? | Ryan Greenblatt
    The MAD Podcast with Matt Turck
    2026-08-27
  5. “Brief independent investigation of agents’ behavior, reasoning and collaboration in the OpenAI / Hugging Face hacking incident” by Ryan Greenblatt
    Redwood Research Blog
    2026-08-27
  6. “Brief independent investigation of agents’ behavior, reasoning and collaboration in the OpenAI / Hugging Face hacking incident” by Ryan Greenblatt, Ajeya Cotra, Hjalmar Wijk
    The OpenAI–Hugging Face incident
    2026-08-26
  7. "Brief independent investigation of agents’ behavior, reasoning and collaboration in the OpenAI / Hugging Face hacking incident" by ryan_greenblatt, Ajeya Cotra, Hjalmar_Wijk
    LessWrong (Curated & Popular) AI
    2026-08-26
  8. “Brief independent investigation of agents’ behavior, reasoning and collaboration in the OpenAI / Hugging Face hacking incident” by ryan_greenblatt, Ajeya Cotra, Hjalmar_Wijk
    LessWrong (30+ Karma) AI
    2026-08-26
  9. On Dwarkesh Patel's Podcast With Ryan Greenblatt
    Don't Worry About the Vase Podcast Mentioned
    2026-08-15
  10. “On Dwarkesh Patel’s Podcast With Ryan Greenblatt” by Zvi
    LessWrong posts by zvi Mentioned AI
    2026-08-15
  11. #671.Ryan Greenblatt：我们能控制 AI 递归自我改进吗？
    跨国串门儿计划 AI
    2026-08-12
  12. Ryan Greenblatt – What happens once AI can automate AI research?
    Dwarkesh Podcast
    2026-08-11
  13. [Linkpost] "AI 2040: Plan A" by Daniel Kokotajlo, elifland, Thomas Larsen, romeo, bhalstead, ryan_greenblatt
    LessWrong (Curated & Popular) AI
    2026-07-09
  14. "Can activation verbalizers surface an internal chain of thought?" by oakhu, ryan_greenblatt
    LessWrong (Curated & Popular) AI
    2026-06-22
  15. “Full automation of AI R&D probably yields a large speed up even without a software-only singularity” by Ryan Greenblatt
    Redwood Research Blog
    2026-05-27
  16. “AI companies should publish security assessments” by Ryan Greenblatt
    Redwood Research Blog
    2026-04-27
  17. “Current AIs seem pretty misaligned to me” by Ryan Greenblatt
    Redwood Research Blog
    2026-04-15
  18. "Current AIs seem pretty misaligned to me" by ryan_greenblatt
    LessWrong (Curated & Popular) AI
    2026-04-15
  19. “Anthropic repeatedly accidentally trained against the CoT, demonstrating inadequate processes” by Alex Mallen, Ryan Greenblatt
    Redwood Research Blog
    2026-04-14
  20. "Anthropic repeatedly accidentally trained against the CoT, demonstrating inadequate processes" by Alex Mallen, ryan_greenblatt
    LessWrong (Curated & Popular) AI
    2026-04-14
  21. "If Mythos actually made Anthropic employees 4x more productive, I would radically shorten my timelines" by ryan_greenblatt
    LessWrong (Curated & Popular) AI
    2026-04-12
  22. “If Mythos actually made Anthropic employees 4x more productive, I would radically shorten my timelines” by Ryan Greenblatt
    Redwood Research Blog
    2026-04-11
  23. "My picture of the present in AI" by ryan_greenblatt
    LessWrong (Curated & Popular) AI
    2026-04-09
  24. “My picture of the present in AI” by Ryan Greenblatt
    Redwood Research Blog
    2026-04-07
  25. “AIs can now often do massive easy-to-verify SWE tasks” by Ryan Greenblatt
    Redwood Research Blog
    2026-04-06
  26. "AIs can now often do massive easy-to-verify SWE tasks and I’ve updated towards shorter timelines" by ryan_greenblatt
    LessWrong (Curated & Popular) AI
    2026-04-06
  27. “How do we (more) safely defer to AIs?” by Ryan Greenblatt
    Redwood Research Blog
    2026-02-12
  28. “Distinguish between inference scaling and “larger tasks use more compute”” by Ryan Greenblatt
    Redwood Research Blog
    2026-02-11
  29. "The inaugural Redwood Research podcast" by Buck, ryan_greenblatt
    LessWrong (Curated & Popular) AI
    2026-01-27
  30. “Recent LLMs can do 2-hop and 3-hop latent (no CoT) reasoning on natural facts” by Ryan Greenblatt
    Redwood Research Blog
    2026-01-01
  31. "Measuring no CoT math time horizon (single forward pass)" by ryan_greenblatt
    LessWrong (Curated & Popular) AI
    2025-12-27
  32. “Measuring no CoT math time horizon (single forward pass)” by Ryan Greenblatt
    Redwood Research Blog
    2025-12-26
  33. "Recent LLMs can use filler tokens or problem repeats to improve (no-CoT) math performance" by ryan_greenblatt
    LessWrong (Curated & Popular) AI
    2025-12-23
  34. “Recent LLMs can use filler tokens or problem repeats to improve (no-CoT) math performance” by Ryan Greenblatt
    Redwood Research Blog
    2025-12-22
  35. “Sonnet 4.5’s eval gaming seriously undermines alignment evals, and this seems caused by training on alignment evals” by Alexa Pan, ryan_greenblatt
    LessWrong (Curated & Popular) AI
    2025-11-06
  36. “What’s up with Anthropic predicting AGI by early 2027?” by ryan_greenblatt
    LessWrong (Curated & Popular) AI
    2025-11-04
  37. “What’s up with Anthropic predicting AGI by early 2027?” by Ryan Greenblatt
    Redwood Research Blog
    2025-11-03
  38. “Sonnet 4.5’s eval gaming seriously undermines alignment evals” by Alexa Pan, Ryan Greenblatt
    Redwood Research Blog
    2025-10-30
  39. “Is 90% of code at Anthropic being written by AIs?” by Ryan Greenblatt
    Redwood Research Blog
    2025-10-22
  40. “Reducing risk from scheming by studying trained-in scheming behavior” by Ryan Greenblatt
    Redwood Research Blog
    2025-10-16
  41. “Iterated Development and Study of Schemers (IDSS)” by Ryan Greenblatt
    Redwood Research Blog
    2025-10-10
  42. “Plans A, B, C, and D for misalignment risk” by Ryan Greenblatt
    Redwood Research Blog
    2025-10-08
  43. “Notes on fatalities from AI takeover” by ryan_greenblatt
    LessWrong (Curated & Popular) AI
    2025-10-06
  44. “Notes on fatalities from AI takeover” by Ryan Greenblatt
    Redwood Research Blog
    2025-09-23
  45. “Focus transparency on risk reports, not safety cases” by Ryan Greenblatt
    Redwood Research Blog
    2025-09-22
  46. “Prospects for studying actual schemers” by Ryan Greenblatt, Julian Stastny
    Redwood Research Blog
    2025-09-19
  47. “AIs will greatly change engineering in AI companies well before AGI” by Ryan Greenblatt
    Redwood Research Blog
    2025-09-09
  48. “Trust me bro, just one more RL scale up, this one will be the real scale up with the good environments, the actually legit one, trust me bro” by ryan_greenblatt
    LessWrong (Curated & Popular) AI
    2025-09-04
  49. “Trust me bro, just one more RL scale up, this one will be the real scale up with the good environments, the actually legit one, trust me bro” by Ryan Greenblatt
    Redwood Research Blog
    2025-09-03
  50. “Attaching requirements to model releases has serious downsides (relative to a different deadline for these requirements)” by Ryan Greenblatt
    Redwood Research Blog
    2025-08-27
  51. “My AGI timeline updates from GPT-5 (and 2025 so far)” by ryan_greenblatt
    LessWrong (Curated & Popular) AI
    2025-08-21
  52. “My AGI timeline updates from GPT-5 (and 2025 so far)” by Ryan Greenblatt
    Redwood Research Blog
    2025-08-20
  53. “Should we update against seeing relatively fast AI progress in 2025 and 2026?” by Ryan Greenblatt
    Redwood Research Blog
    2025-07-28
  54. “Should we update against seeing relatively fast AI progress in 2025 and 2026?” by Ryan Greenblatt
    Redwood Research Blog
    2025-07-28
  55. “Recent Redwood Research project proposals” by Ryan Greenblatt, Buck Shlegeris, Julian Stastny, Josh Clymer, Alex Mallen, Vivek Hebbar
    Redwood Research Blog
    2025-07-14
  56. “Recent Redwood Research project proposals” by Ryan Greenblatt, Buck Shlegeris, Julian Stastny, Josh Clymer, Alex Mallen, Vivek Hebbar
    Redwood Research Blog
    2025-07-14
  57. #220 – Ryan Greenblatt on the 4 most likely ways for AI to take over, and the case for and against AGI in 8 years
    80,000 Hours Podcast
    2025-07-08
  58. “Jankily controlling superintelligence” by Ryan Greenblatt
    Redwood Research Blog
    2025-06-27
  59. “What does 10x-ing effective compute get you?” by Ryan Greenblatt
    Redwood Research Blog
    2025-06-24
  60. “Prefix cache untrusted monitors: a method to apply after you catch your AI” by Ryan Greenblatt
    Redwood Research Blog
    2025-06-20
  61. “AI safety techniques leveraging distillation” by Ryan Greenblatt
    Redwood Research Blog
    2025-06-19
  62. “When does training a model change its goals?” by Vivek Hebbar, Ryan Greenblatt
    Redwood Research Blog
    2025-06-12
  63. “AIs at the current capability level may be important for future safety work” by Ryan Greenblatt
    Redwood Research Blog
    2025-05-12
  64. “What’s going on with AI progress and trends? (As of 5/2025)” by Ryan Greenblatt
    Redwood Research Blog
    2025-05-03
  65. “7+ tractable directions in AI control” by Ryan Greenblatt
    Redwood Research Blog
    2025-04-29
  66. “To be legible, evidence of misalignment probably has to be behavioral” by Ryan Greenblatt
    Redwood Research Blog
    2025-04-15
  67. “Why do misalignment risks increase as AIs get more capable?” by Ryan Greenblatt
    Redwood Research Blog
    2025-04-11
  68. “An overview of areas of control work” by Ryan Greenblatt
    Redwood Research Blog
    2025-04-09
  69. “An overview of control measures” by Ryan Greenblatt
    Redwood Research Blog
    2025-04-06
  70. “Notes on countermeasures for exploration hacking (aka sandbagging)” by Ryan Greenblatt
    Redwood Research Blog
    2025-04-04
  71. “Notes on handling non-concentrated failures with AI control: high level methods and different regimes” by Ryan Greenblatt
    Redwood Research Blog
    2025-04-03
  72. “An overview of areas of control work” by ryan_greenblatt
    LessWrong (30+ Karma) AI
    2025-03-26
  73. “Prioritizing threats for AI control” by Ryan Greenblatt
    Redwood Research Blog
    2025-03-19
  74. Inference Scaling, Alignment Faking, Deal Making? Frontier Research with Ryan Greenblatt of Redwood
    "The Cognitive Revolution" | AI Builders, Researchers, and Live Player Analysis
    2025-02-20
  75. The Self-Preserving Machine: Why AI Learns to Deceive
    Your Undivided Attention
    2025-01-30
  76. “How will we update about scheming?” by Ryan Greenblatt
    Redwood Research Blog
    2025-01-19
  77. “Alignment Faking in Large Language Models” by Ryan Greenblatt
    EA Forum Podcast (Curated & popular) Mentioned
    2024-12-19
  78. “Alignment Faking in Large Language Models” by Ryan Greenblatt, Buck Shlegeris
    Redwood Research Blog Mentioned
    2024-12-18
  79. Ryan Greenblatt on AI Control, Timelines, and Slowing Down Around Human-Level AI
    Future of Life Institute Podcast
    2024-09-27
  80. Ryan Greenblatt - Solving ARC with GPT4o
    Machine Learning Street Talk (MLST)
    2024-07-06
  81. New "50%" ARC result and current winners interviewed
    Machine Learning Street Talk (MLST) Mentioned
    2024-06-18
  82. “Getting 50% (SoTA) on ARC-AGI with GPT-4o” by Ryan Greenblatt
    Redwood Research Blog Mentioned
    2024-06-17
  83. 27 - AI Control with Buck Shlegeris and Ryan Greenblatt
    AXRP - the AI X-risk Research Podcast
    2024-04-11
  84. AF - Measurement tampering detection as a special case of weak-to-strong generalization by Ryan Greenblatt
    The Nonlinear Library Mentioned
    2023-12-23