Chipmonk

• ChipmonkMay 11, 2025, 4:18 AM

  2 points

  0

  on: Progress = Fewer Bad Moments

  others have reported this phenomenon too:

  • https://​​x.com/​​nickcammarata/​​status/​​1863305295013065008

  • https://​​x.com/​​meditationstuff/​​status/​​1863321287080243505

• ChipmonkMay 8, 2025, 4:58 AM

  2 points

  0

  in reply to: JustisMills’s comment on: Progress = Fewer Bad Moments

  I’m not sure the idea of steady progress is so bad as to be net harmful, though.

  You say this and then I don’t understand your explanation. Could you re-explain?

• ChipmonkMay 7, 2025, 9:20 PM

  2 points

  0

  in reply to: Kaj_Sotala’s comment on: Progress = Fewer Bad Moments

  oh nice

Progress = Fewer Bad Moments

• ChipmonkApr 14, 2025, 1:59 AM

  3 points

  0

  on: Un­bend­able Arm as Test Case for Reli­gious Belief

  tagged this post as self-fulfilling prophecies!

• ChipmonkApr 14, 2025, 1:59 AM

  3 points

  0

  on: Un­bend­able Arm as Test Case for Reli­gious Belief

  yes!! “focusing on what you want!” (i talk a little more about this and self-fulfilling prophecies here)

  **aiming** at what you want. vector. (teleology, not etiology)

• ChipmonkApr 10, 2025, 8:00 PM

  3 points

  0

  on: Learned pain as a lead­ing cause of chronic pain

  This is also my experience helping people with lifelong anxiety/​insecurity

• ChipmonkApr 3, 2025, 7:39 PM

  2 points

  0

  on: Ex­am­ples of self-fulfilling prophe­cies in AI al­ign­ment?

  

  https://​​x.com/​​saffronhuang/​​status/​​1907863453009867183

• ChipmonkApr 2, 2025, 7:03 PM

  3 points

  −1

  on: VDT: a solu­tion to de­ci­sion theory

  Now we just need to ask Sonnet to formalize VDT

Align­ment First, In­tel­li­gence Later

Soft­max, Em­mett Shear’s new AI startup fo­cused on “Or­ganic Align­ment”

• ChipmonkMar 17, 2025, 9:04 PM

  2 points

  0

  on: An­nounc­ing EXP: Ex­per­i­men­tal Sum­mer Work­shop on Col­lec­tive Cognition

  what are the examples of the curriculum/​activities you’re considering/​planning?

• ChipmonkMar 15, 2025, 9:08 PM

  2 points

  0

  on: An­nounc­ing EXP: Ex­per­i­men­tal Sum­mer Work­shop on Col­lec­tive Cognition

  @Ivan Vendrov think you’d be interested

• ChipmonkMar 3, 2025, 2:54 AM

  5 points

  0

  on: Self-fulfilling mis­al­ign­ment data might be poi­son­ing our AI models

  Can you think of examples like this in the broader AI landscape? - What are the best examples of self-fulfilling prophecies in AI alignment?

• ChipmonkMar 3, 2025, 2:50 AM

  8 points

  0

  on: Ex­am­ples of self-fulfilling prophe­cies in AI al­ign­ment?

  

  https://​​x.com/​​sama/​​status/​​1621621724507938816

• ChipmonkMar 3, 2025, 2:47 AM

  4 points

  0

  on: Ex­am­ples of self-fulfilling prophe­cies in AI al­ign­ment?

  Situational Awareness and race dynamics? h/​t Jan Kulveit @Jan_Kulveit

• ChipmonkMar 3, 2025, 2:46 AM

  8 points

  0

  on: Ex­am­ples of self-fulfilling prophe­cies in AI al­ign­ment?

  Training on Documents About Reward Hacking Induces Reward Hacking

[Question] Ex­am­ples of self-fulfilling prophe­cies in AI al­ign­ment?

• ChipmonkFeb 28, 2025, 8:06 PM

  4 points

  0

  in reply to: niplav’s comment on: Do clients need years of ther­apy, or can one con­ver­sa­tion re­solve the is­sue?

  I don’t feel like I learned anything new from the post.

  This surprises me! Wait so-

  • The “How does one-shotting happen?” section didn’t have anything interesting for you? (Have you seen stuff like this elsewhere?)

  • Did you already know one-shotting was possible?

• ChipmonkFeb 28, 2025, 8:04 PM

  2 points

  0

  in reply to: niplav’s comment on: Do clients need years of ther­apy, or can one con­ver­sa­tion re­solve the is­sue?

  since your bullet-point list in the beginning isn’t detailed enough for anyone to try to replicate the method.

  Wait I’m confused- this is not the purpose of the post

  Also notable is that you only have positive examples for your method

  The purpose of this post is not advertisement. It’s to discuss one-shots

  Especially, how would you be able to distinguish between your approach convincing your customers they were helped, instead of actually changing their behavior?

  See above